ThrombUS+ D2.3: Technical and functional requirements

D2.3. Technical and Functional Requirements V. Marozas [KTU], E. Kaldoudi [ATHENA], T. Prinz [VDE], A. Lukoševičius [KTU], R. Jurkonis [KTU], R. Eitminavičius [KTU], S. Didaskalou [ATHENA], A. Vehkaoja [TAU], R. Ahola [TAU], A. Slabov [KTU], D. Jegelevičius [KTU], S. Balling [MEDIS], A. Moutafidou [ATHENA], S. Daukantas [KTU], A. Rapalis [KTU] Due Date: 31 October 2024 Delivery Date: 31 October 2024 Revision Date: 12 November 2025 Horizon Innovation Action | Agreement No. 101137227 HORIZON-HLTH-2023-TOOL-05-05 Co-funded by the European Union D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 ii ThrombUS+ Consortium ATHENA Research and Innovation Center in Information, Communication and Knowledge Technologies, Greece Eleni Kaldoudi [email protected] KTU Kaunas University of Technology Lithuania Vaidotas Marozas [email protected] VERMON Vermon SA France Mathieu Legros [email protected] FRAUNHOFER Institute for Photonic Microsystems, Fraunhofer Germany Nicolas Lange [email protected] TELEMED Telemed Ultrasound Medical Systems Lithuania Dmitry Novikov [email protected] EchoNous EchoNous Inc USA Pavlos Moustakidis [email protected] MEDIS medis Medizinische Messtechnik GmbH Germany Susann Balling [email protected] ComfTech ComfTech SLR Italy Lara Alessia Moltani [email protected] TAU Faculty of Medicine and Health Technology Tampere University, Finland Antti Vehkaoja [email protected] LSMU Lithuanian University of Health Science Lithuania Andrius Macas [email protected] GNP Papageorgiou General Hospital Greece Maria Bigaki [email protected] CSS-IRCCS Home Relief of Suffering Hospital Italy Elvira Grandone e.grand[email protected] HSV Simon Veil Hospital France Maxime Gautier [email protected] VDE Association for Electrical, Electronic & Information Technologies, Germany Thorsten Prinz thorsten.pri[email protected] MEDEA MEDEA SRL Italy Pietro Dionisio [email protected] PHAZE Clinical Research and Pharma Consulting SA Greece Spiros Anagnostopoulos [email protected] PBY PredictBy Research and Consulting SL Spain Frans Folkvord [email protected] SciGen SciGen Technologies SA Greece Katerina Pavlidi [email protected] Disclaimer This document contains description of the ThrombUS+ project work, findings, and products. The authors of this document have taken any available measure for its content to be accurate, consistent and lawful. However, neither the project consortium as a whole nor the individual partners that implicitly or explicitly participated in the creation and publication of this document hold any sort of responsibility that might occur as a result of using its content. Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or HADEA. Neither the European Union nor the granting authority HADEA can be held responsible for them. In case you believe that this document harms in any way intellectual property held by you as a person or as a representative of an entity, please do notify us immediately. ThrombUS+ is an Innovation Action Project co-funded by the European Union, under HORIZON-HLTH-2023-TOOL05-05 “Harnessing the potential of real-time data analysis and secure Point-of-Care computing for the benefit of person-centred health and care delivery”. D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 iii Document Control Page Project Grant Agreement: 101137227 Acronym: ThrombUS+ Title: Wearable Continuous Point-of-Care Monitoring, Risk Estimation and Prevention for Deep Vein Thrombosis Type: Innovation Action Start: 1 January 2024 End: 30 June 2027 Programme: Horizon Europe Call Identifier: HORIZON-HLTH-2023-TOOL-05-05 Call Topic Harnessing the potential of real-time data analysis and secure Point-of-Care computing for the benefit of person-centred health and care delivery Website: https://thrombus.eu/ Deliverable # 12 No: D2.3 Deliverable Title: Technical and functional requirements Deliverable Type: R Classification: PU Task: T2.3. Technical and functional requirements [M01 - M10] Task Leader: KTU [V. Marozas] Work Package: Work package WP2 - Requirements and product co-design [M01 - 12] Work Package Leader: PBY [L. J. Segal] Responsible Partner: KTU Authors: V. Marozas [KTU], E. Kaldoudi [ATHENA], T. Prinz [VDE], A. Lukoševičius [KTU], R. Jurkonis [KTU], R. Eitminavičius [KTU], S. Didaskalou [ATHENA], A. Vehkaoja [TAU], R. Ahola [TAU], A. Slabov [KTU], D. Jegelevičius [KTU], S. Balling [MEDIS], A. Moutafidou [ATHENA], S. Daukantas [KTU], A. Rapalis [KTU] Input from: All consortium partners Peer Reviewers: PHAZE [I. Drougka], MEDEA [P. Dionisio] Due Date: M10 – 31 October 2024 Delivery Date: 31 October 2024 Revision Date: 12 November 2025 Document Status Version: 2.0 Status: Draft Consortium reviewed WP leader endorsed Coordinator endorsed D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 iv Revision History Version Date Modification Contributors 0.01 15 Apr 2024 New content – outline V. Marozas [KTU] 0.01 16 Apr 2024 New content – revision of outline V. Marozas [KTU] 0.02 30 May 2024 New content: 3.3 V. Marozas [KTU] 0.03 09 Jun 2024 New content: 3.3 V. Marozas [KTU] 0.04 19 Jun 2024 New content: structure V. Marozas [KTU]; T. Prinz [VDE] 0.04 21 Jun 2024 New content: 3.3 V. Marozas, A. Lukoševičius [KTU] 0.04 18 Jul 2024 New content: 3.3, 3.4 R. Jurkonis [KTU] 0.05 08 Aug 2024 New content: 3.3, 3.4 V. Marozas, R. Jurkonis [KTU] 0.05 14 Aug 2024 New content: 2.4, 3.3, 3.4 V. Marozas, MJ [KTU] 0.05 21 Aug 2024 New content: 2.4, 3.3, 3.4 V. Marozas, RE, [KTU] 0.05 26 Aug 2024 New content: 3.3.2 R. Eitminavičius [KTU] 0.05 28 Aug 2024 New content: 3.3.2 R. Eitminavičius [KTU] 0.1 04 Sep 2024 Changes in Ch.4 V. Marozas, A. Lukoševičius [KTU] 0.2 16 Sep 2024 New content: 3.3.3 V. Marozas, R. Eitminavičius [KTU] 0.3 20 Sep 2024 Changes in Ch.3.3.4, Ch. 3.4 V. Marozas, D. Jegelevičius [KTU]; A. Moutafidou [ATHENA] 0.4 28 Sep 2024 New content Ch. 4.8 A. Vehkaoja [TAU] 0.4 30 Sep 2024 New content Ch. 4.8 V. Marozas [KTU] 0.4 03 Oct 2024 New content Ch. 3.1, 3.2, 3.4, 4.1 S. Didaskalou, A. Moutafidou, E. Kaldoudi [ATHENA]; A. Lukoševičius, V. Marozas [KTU] 0.5 07 Oct 2024 Changes in Ch.2.5, Ch. 3.5 V. Marozas, A. Lukoševičius [KTU] 0.6 09 Oct 2024 New content Ch. 4.1, 4.3, 4.4 V. Marozas, A. Lukoševičius [KTU]; A. Sakalauskas [TELEMED] 0.6 10 Oct 2024 New content Ch. 3.1, restructuring S. Didaskalou, E. Kaldoudi [ATHENA] 0.7 12 Oct 2024 New content Ch. 3.1 V. Marozas [KTU] 0.8 22 Oct 2024 Review comments, changes Ch. 3.1 S. Balling [MEDIS]; T. Prinz [VDE]; R. Eitminavičius, V. Marozas [KTU] 0.8 24 Oct 2024 Review of whole text A. Rapalis [KTU] 0.9 27 Oct 2024 Correction of incorrect references. V. Marozas [KTU] 1.0 31 Oct 2024 Revised for conformity S. Didaskalou, E. Kaldoudi [ATHENA] 2.0 12 Nov 2025 Revised based on PR1 review comments: 1. Section 3: State-of-the-Art Analysis and Existing Gaps Implying Requirements was moved to Annex 3 2. The requirements for the hardware in TR_CHI_01 in Table 7 were updated as follows: − GPU: Nvidia RTX 4060 − Processor: i5-13600K − RAM: 16GB E. Kaldoudi [ATHENA] D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 v 3. The following technical requirements were added in Table 13: − TR_XGM_12: Game engagement is supported for each one of the 3 lower limb segments for at least one exercise and at least one story line [R] − TR_XGM_13: Game engagement is supported for patients in the lying, sitting and standing position [R] and the respective system requirements were added in the table of Annex 2: − SR_PR01_XGM_12_01: Game engagement is supported for each one of the 3 lower limb segments for at least one exercise and at least one story line − SR_PR01_XGM_13_01: Game engagement is supported for patients in the lying, sitting and standing position D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 vi Contents About ThrombUS+ ...................................................................................................................................................1 Task T2.3 and Deliverable D2.3 Description .............................................................................................................1 Cite this Document as ..............................................................................................................................................1 Terms and Definitions .............................................................................................................................................2 Executive Summary .................................................................................................................................................4 1. Introduction ....................................................................................................................................................5 1.1. Methodology ................................................................................................................................................. 5 2. Intended purpose, high-level functional and technical requirements .............................................................6 2.1. Intended purpose of the ThrombUS+ system ................................................................................................ 6 2.2. ThrombUS+ system components ................................................................................................................... 7 2.3. ThrombUS+ reference use cases (RUCs) ........................................................................................................ 7 2.4. Target users ................................................................................................................................................ 10 2.5. Main user requirements .............................................................................................................................. 11 3. Module-based analysis of requirements ....................................................................................................... 11 3.1. ThrombUS+ system modules ....................................................................................................................... 11 3.2. User (functional) requirements ................................................................................................................... 13 3.3. Technical requirements ............................................................................................................................... 15 4. Summary and Conclusions ............................................................................................................................ 21 ANNEX 1 | ThrombUS+ User Requirements ........................................................................................................... 23 ANNEX 2 | ThrombUS+ System Requirements ....................................................................................................... 33 ANNEX 3 | State-of-the-art analysis and existing gaps implying requirements ...................................................... 52 1. Risk assessment scores ................................................................................................................................. 52 1.1. Summary of knowledge gaps and potential solutions with implications for requirements ........................ 53 2. Prophylaxis methods ..................................................................................................................................... 53 2.1. Intermittent pneumatic compression .......................................................................................................... 54 2.2. Neuromuscular electrical stimulation ......................................................................................................... 54 2.3. Exoskeletons and soft robotics .................................................................................................................... 55 2.4. Physiology-guided stimulation .................................................................................................................... 57 2.5. Exercises ...................................................................................................................................................... 57 2.6. Serious games ............................................................................................................................................. 58 2.7. Extended reality .......................................................................................................................................... 64 2.8. Summary of knowledge gaps and potential solutions with implications for requirements ........................ 69 D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 vii 3. DVT monitoring and diagnostic methods and technologies ........................................................................... 70 3.1. B-scan and Doppler Ultrasound .................................................................................................................. 70 3.1.1. Summary of knowledge gaps and potential solutions with implications for requirements .............. 72 3.2. Venous occlusion plethysmography ............................................................................................................ 72 3.2.1. Strain-gauge plethysmography (SGP)................................................................................................. 73 3.2.2. Air plethysmography (APG) ................................................................................................................ 74 3.2.3. Electrical Impedance Plethysmography (EIP) ..................................................................................... 74 3.2.4. Approach and extracted parameters. ................................................................................................ 75 3.2.5. EIP implementation details ................................................................................................................ 76 3.2.6. Diagnostic accuracy ............................................................................................................................ 77 3.2.7. Challenges, errors, and limitations ..................................................................................................... 78 3.2.8. Future considerations ........................................................................................................................ 78 3.2.9. Commercially available VOP systems ................................................................................................. 79 3.2.10. Summary of knowledge gaps and potential solutions with implications for requirements .............. 80 3.3. Light reflection rheography (LRR) and photoplethysmography (PPG) ........................................................ 80 3.3.1. Background ........................................................................................................................................ 80 3.3.2. Approach ............................................................................................................................................ 81 3.3.3. Implementation details and extracted parameters ........................................................................... 81 3.3.4. Diagnostic accuracy ............................................................................................................................ 84 3.3.5. Challenges, errors, and limitations ..................................................................................................... 85 3.3.6. Future considerations ........................................................................................................................ 85 3.3.7. Commercially available LRR systems .................................................................................................. 85 3.3.8. Summary of knowledge gaps and potential solutions with implications for requirements .............. 86 3.4. Other promising DVT diagnostic methods and technologies ...................................................................... 87 3.4.1. Muscle stiffness assessment .............................................................................................................. 87 3.4.2. Summary of Knowledge Gaps and Potential Solutions with Implications for Requirements ............ 91 4. Actuators for tissue compression and US transducer positioning .................................................................. 91 4.1. Background ................................................................................................................................................. 92 4.2. Approaches ................................................................................................................................................. 92 4.3. Summary of knowledge gaps and potential solutions with implications for requirements ........................ 98 5. Wearable limb activity and orientation monitoring ...................................................................................... 99 5.1. Background ................................................................................................................................................. 99 5.2. Use cases of wearable activity monitoring ................................................................................................. 99 5.3. Approaches ............................................................................................................................................... 100 5.4. Calibration of IMU-based systems ............................................................................................................ 100 5.5. Accuracy, reliability and limitations .......................................................................................................... 100 5.6. Implementation details and output parameters ....................................................................................... 101 5.7. Commercially available IMU systems ....................................................................................................... 101 5.8. Summary of knowledge gaps and potential solutions with implications for requirements ...................... 101 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 1 About ThrombUS+ Deep vein thrombosis (DVT) is the formation of a blood clot within the deep veins, most commonly those of the lower limbs, obstructing blood flow. In 50% of people with DVT, the clot eventually breaks off and travels to the lung to cause pulmonary embolism. Clinical assessment of DVT is notoriously unreliable because up to 2/3 of DVT episodes are clinically silent and patients are symptom-free even when pulmonary embolism has developed. Early diagnosis of DVT is crucial and despite the progress made in ultrasound imaging and plethysmography techniques, there is a need for new methods to enable continuous monitoring of DVT diagnosis at the point of care. ThrombUS+ brings together an interdisciplinary team of industrial, technology, regulatory, social science, and clinical trial experts to develop a novel wearable diagnostic device for point-of-care, operator-free, continuous monitoring in patients with high DVT risk. The device will combine autonomous, AI-driven DVT detection based on novel wearable ultrasound hardware, impedance plethysmography, and light reflection rheography for immediate detection of blood clot formation in the lower limb. Activity and other physiological measurements will be used to provide a continuous assessment of DVT risk and support DVT prevention via serious gaming. The aggregated data will drive an intelligence decision support unit that will provide accurate monitoring and alerts. Extended reality will be used to guide experts to design exercises and patients to use the device optimally. ThrombUS+ is intended for use by postoperative patients in the ward, during long surgical operations, cancer patients or otherwise bedridden patients at home or in care units, and women during pregnancy and postpartum. ThrombUS+ will use big data sets for AI training collected in the project via 3 large-scale clinical studies and will validate the outcome in the clinical setting via 1 early feasibility study and 1 multi-center clinical trial. Task T2.3 and Deliverable D2.3 Description DoA provides this description of the task T2.3: “The goal of this task is to identify the high-level technical and functional criteria required by the wearable system to effectively and personally monitor, detect, and prevent DTV. Also, this task involves the identification and reviewing of existing integration tools of extended reality, sensor, and motion recognition technologies, to derive the technical specifications for the integration requirements between the different software and hardware components of ThrombUS+. Focus groups and brainstorming sessions will be conducted with clinicians to include the medical perspective. Algorithm requirements will be defined for integration and processing of data stream generated by multisensory system for AI-based decision support, and also for interface with intelligent autonomous wearable. Technical requirements for maximal size, power consumption, durability of reliable sensor contact, unobtrusiveness, and requirements for wireless data communication channels will be defined. The technical academic partner KTU, with extensive experience in medical wearable sensors and sensor networks, will lead this task. All partners will participate to ensure that technical and functional requirements represent all views and are fully understood by the entire consortium. A report on the high-level technical and functional requirements for the wearable system to effectively and personally monitor, detect, and prevent DVT. Output of Task 2.3.” Cite this Document as Marozas V, Kaldoudi E, Prinz T, Lukoševičius A, Jurkonis R, Eitminavičius R, Didaskalou S, Vehkaoja A, Ahola R, Slabov A, Jegelevičius D, Balling S, Moutafidou A, Daukantas S, Rapalis A. Technical and functional requirements, Deliverable 2.3, ThrombUS+ Horizon Europe Innovation Action, EC Grant Agreement No. 101137227, 31 October 2024. Revised 12 November 2025. https://doi.org/10.5281/zenodo.17642345 D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 2 Terms and Definitions Term Definition APG Air plethysmography ASGP Ambulatory strain gauge plethysmography CHI Identifier for ThrombUS+ module Central Hub & Intelligence CUS Compression ultrasound imaging DBQ Database query plan DoA Description of action DOF Degrees of freedom DVI Deep vein incompetence DVO Deep vein occlusion DVT Deep vein thrombosis EIM Identifier for ThrombUS+ Electrical impedance module EIP Electrical impedance plethysmography FV Flow volume ICC Intraclass coefficient ICU Intensive care unit IMU Inertial measurement unit IPCS Intermittent pneumatic compression system LAM Identifier for ThrombUS+ Limb Activity Module LRM Identifier for ThrombUS+ Light Reflection Module LRR Light reflection rheography MEMS Microelectromechanical systems NMES Neuromuscular electrical stimulation OF Outflow PE Pulmonary embolism PPG Photo plethysmography PPH Postpartum haemorrhage PV Peak velocity RUC Reference use case SGP Strain-gauge plethysmography SOTA State-of-the-art TAMEAN Intensity-weighted mean frequency TCA Tissue compression actuator D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 9 − Patient comfort: The device should be comfortable, avoiding excessive pressure, skin irritation, or reactions that could cause discomfort. It should consider patient anxiety by being easy to use and minimally intrusive. − Design considerations: A stocking-shaped design could be suitable, especially if it includes anti-embolism compression for high-risk patients who cannot be anticoagulated. The design should monitor both legs, as many surgeries are bilateral. However, it must avoid excessive pressure or irritation, especially for cardiovascular surgery patients with leg wounds or swelling and should not be used in cases where continuous compression is contraindicated (e.g., peripheral arterial disease or low arterial flow patients). RUC #4 – Pregnancy and postpartum − Continuous monitoring opportunities: For patients with sufficient suspicion or detection of VTE (venous thromboembolism), continuous DVT monitoring is beneficial. This applies to both hospitalized and selftreated patients, as continuous monitoring of blood clot formation is not feasible even in high-risk therapy units. This is especially crucial for conditions like postpartum haemorrhage (PPH), where thromboprophylaxis can’t be used, increasing VTE risk. − Early detection and resource utilization: Continuous scanning and alerting for DVT indicators allow earlier and more effective detection, optimizing clinical resources. − Post-discharge monitoring: Continuous monitoring can be used preventively at home or during travel for moderate-risk cases, aiding in follow-up and understanding prophylactic self-treatment outcomes. It also helps clinicians understand VTE recurrence better. − Current practices: Physical and non-pharmacological devices, such as graduated elastic compression stockings, are already welcomed in current practice to prevent VTE events. ThrombUS+ could be highly valued if it increases early detection rates and is accepted for use, potentially preventing maternal deaths and long-term health consequences for women. − Key functionalities: The device should detect key signs like saturation, pulse rate, and respiratory rate, sending alarms to doctors upon detecting DVT-related anomalies. Constant monitoring and real-time alerts are essential for timely interventions. − Simplicity and data storage: The device should be simple to use, deployable by nurses and patients, and capable of storing at least 24 hours of data for informed medical decisions upon receiving alarms. − Design considerations: As 90% of cases occur on the left leg, the device could be designed as a stocking, strap, belt, or band, prioritizing patient comfort. Continuous compression, while not dangerous, can be uncomfortable or painful, affecting adherence, especially outside clinical settings. − Material and comfort: The device should be light, breathable, and comfortable, considering pregnant women's tendency to overheat and general discomfort due to reduced mobility. − Interoperability: Ensuring compatibility with other electronic devices, including smartphones, can make the device more user-friendly and acceptable to patients. RUC #5 – Cancer and oncological treatment − Early detection: Continuous monitoring can detect asymptomatic DVT, as no specific monitoring is currently performed, especially in low-monitoring settings. − Outpatient management: It helps assess prophylactic treatment effectiveness after discharge, aiding in deciding when to adjust or stop treatment, improving patients’ quality of life. − Recurrence prevention: Beneficial even for patients who have responded well to treatment and regained mobility to prevent DVT recurrence. − Leg device preference: Most DVT cases occur in the leg, making a leg device most appropriate, but adaptability to other body parts for monitoring different symptoms is beneficial. D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 10 − Sensitivity and comfort: Device sensitivity determines usage duration; it should be comfortable for near 24/7 wear, including during sleep, especially for cancer patients. − Ease of use: The device must be user-friendly and non-intrusive, requiring minimal attention from patients, essential for long-term use, particularly for those undergoing multiple treatments. RUC #6 – Obesity − Customized design considerations: Designing ThrombUS+ devices must consider the patient’s health status and weight, ensuring minimal disruption and maximum effectiveness. This co-creation approach with doctors and patients aims to optimize device usability. − Benefits for obesity patients: Intermittent compression stockings integrated with ultrasound scanning offer mechanical prophylaxis without increasing bleeding risks. This design supports patients following lifestyle treatments to reduce weight and fat percentage, crucial for managing obesity-related risks like DVT. − Adaptability and mobility: Stocking designs are preferred as they provide stable compression without slipping, crucial during immobilization periods post-surgery. However, they may hinder mobility during recovery stages where natural muscle compression suffices. − Long-term use and follow-up: Patients with obesity may require prolonged device use, necessitating comfort, and materials that prevent skin irritation. ThrombUS+ usage could reduce the need for frequent Doppler ultrasound studies, cutting costs and streamlining follow-up procedures. − Comfort and usability: Devices must be comfortable and easy to use, especially for obese patients dealing with fatigue, mobility challenges, and multiple treatments. Comfort ensures patient adherence, even during sleep, if continuous monitoring is required. − Long-term wear: Since some patients may need to use the device for months or years, it must be nonintrusive and integrate seamlessly into daily life alongside other treatments. − Body adaptability: While leg devices are typically suitable for DVT monitoring, flexibility to adapt to other body parts with varying coagulability should be considered. This adaptation ensures the device meets specific patient needs, discussed in collaboration with healthcare providers. RUC #7 – Autoimmune diseases and genetic disorders − Suitability for autoimmune diseases and genetic disorders: Intermittent compression stockings with integrated ultrasound scanning provide mechanical prophylaxis essential for these patients. They reduce the need for risky anticoagulant treatments and can alert doctors promptly if critical conditions arise. − Flexibility and sensitivity: Devices must adapt to rapid health changes typical of autoimmune diseases, such as flare-ups in conditions like multiple sclerosis. Sensitivity and flexibility in device usage duration are crucial for patient comfort and adherence. − Long-term wear and comfort: Considering the chronic nature of autoimmune diseases, devices must be comfortable for long-term use, accommodating patient fatigue and the need for multiple treatments and devices. − Integration into daily life: Since patients may need to use the device for extended periods, ease of use and comfort are essential to ensure continuous adherence alongside other treatments and devices. 2.4. Target users Although the ThrombUS+ system intended purpose is to be used by healthcare professionals in the clinical environment to enhance patients’ treatment, the functional and technical requirements were designed based on the device’s overall lifespan. Therefore, in the context of user and system requirements, the target users are: D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 11 − patients, − medical specialists: doctor, physical therapist, nurse, medical technician, − administrators: system administrator, maintenance engineers, − researchers: medical researcher, biomedical engineer. 2.5. Main user requirements The ThrombUS+ system must provide the following main user functionalities: − Assist medical specialists with decision support for DVT diagnosis in the following RUCs: • neurosurgery, • lower limb orthopaedic surgery, • cardiovascular surgery and risk. − Assist medical specialists in continuously monitoring patients for changes in DVT digital biomarkers that indicate an increased risk of DVT in the following RUCs: • pregnancy and postpartum, • cancer and oncological treatment, • obesity, • autoimmune diseases and genetic disorders. − Assist patients with DVT prevention and prophylaxis in the following RUCs: • pregnancy and postpartum, • cancer and oncological treatment, • obesity, • autoimmune diseases and genetic disorders. − Assist administrators in maintaining the system by recording system startup, operation, and error logs. − Assist medical researchers and biomedical engineers in collecting raw medical data and signals for dataset acquisition. 3. Module-based analysis of requirements The ThrombUS+ system has a modular structure defined by the complexity of DVT monitoring functions and the multimodality of data sources and technologies. The modular concept of hardware and software enhances the system’s flexibility, scalability, and reusability, allowing for easy adaptation to user needs and use cases. At the same time, ThrombUS+ functions as a unified whole, communicating with the modules and integrating and aggregating module data in the Central Hub Intelligence. The technical requirements presented below are organized on a module-based level. 3.1. ThrombUS+ system modules The conceptual structure of the modules in the ThrombUS+ system, along with their main internal relationships, is shown in Figure 2. D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 12 Figure 2. Conceptual modules and their Interrelationships in the ThrombUS+ system. A project-wide unique identification system will incorporate the project outcomes and all related technical information and documentation, including individual software and hardware components, use cases, user requirements, and functional and technical specifications. At this stage, preliminary identifiers are proposed for the various planned project outcomes at the major module level, as summarized in Table 1. Table 1. Indicative unique identifiers of modules (the list will be updated during the project). ThrombUS+ Modules Identifier Central Hub & Intelligence CHI Wearable Actuator Module WAM Ultrasound Module USM Electrical Impedance Module EIM Light Rheography Module LRM Limb Activity Module LAM Extended Reality & Serious Gaming Module XGM The ThrombUS+ modules, as integral parts of the system, each serve specific purposes, described below: − The CHI consists of hardware (a laptop computer) and software (ThrombUS+ applications) used for controlling actuators, gathering data from sensors, processing data, implementing operator-controlled and independent DVT monitoring methods, patient guidance, and supporting decision-making. − The WAM performs thigh tissue compression to implement: a) the Compression Duplex Ultrasonography (CDUS) method, b) automatic steering of the ultrasound transducer using a feedback mechanism, enabling operator-independent DVT monitoring, and c) the Venous Occlusion Plethysmography (VOP) method, using EIM and LRM modules. − The USM performs intermittent measurements for operator-independent DVT monitoring, including: a) ultrasound imaging and online segmentation of deep veins in the thigh using the compression method, D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 13 b) detection of thrombus by tracking vein localization and dimensions during pressure application with the WAM, c) estimation of vein blood flow perturbations caused by distal DVT using Colour, Power, and Wave Doppler imaging, and d) ensuring a continuous data stream for decision support on DVT risk, thrombus detection, and characterization. − The EIM is a hardware module used by CHI for the implementation of the VOP method together with WAM. − The LRM is a hardware module used by CHI for the implementation of the VOP method together with the WAM or the tiptoe DVT test. − The LAM is a hardware module used by CHI for the implementation of a serious game used in DVT prophylaxis, as well as for providing biofeedback on lower limb maneuvers during ultrasound-based monitoring. − The XGM is a software module used for DVT risk estimation and prophylaxis. 3.2. User (functional) requirements User (functional) requirements of the ThrombUS+ system describe the specific behaviours, actions, or functions a system must perform to meet the needs of its users. These requirements focus on what the system should do from the end user’s perspective, defining the capabilities and tasks it must fulfil. They are user-oriented, specifying what users expect the system to do. In addition, they are actionable, as they outline specific tasks the system must perform. The requirements are also measurable, ensuring clarity for testing and validation. Lastly, they are non-technical, emphasizing what the system should achieve without delving into how it should be implemented. Four types of users were identified of the ThrombUS+ system: 1) Medical specialists (doctor, nurse, medical technician), 2) Patients, 3) Maintenance engineers, 4) Biomedical researchers. The list of main user functionalities of the ThrombUS+ system is provided in Table 2, Table 3, Table 4 and Table 5 below, for each user level. Table 2. Functional requirements: medical specialist level. ID Description Request (R) Wish (W) FR_MS_01 The system must support user authentication. R FR_MS_02 Allow medical specialists to input and review patient-related data, such as demographics and anamnesis. R FR_MS_03 The system must support interactive real-time measurements by using compression ultrasound (B-scan and Doppler), LRR, EIM, and IMU sensors. R FR_MS_04 The system must support online monitoring and data analysis of measurement data automatically. R FR_MS_05 The system must support automatic alerts based on the DVT detection model. R FR_MS_06 The system must support offline measurements and visual analytics of the total recorded data—including risk factors, ultrasound images (B-scan and Doppler) and related parameters, light reflection rheography and electrical impedance signals and parameters, physical activity, and limb orientation data—in an interactive and semi-automatic way. R FR_MS_07 The system should have a database with lower limb exercises for DVT prevention and rehabilitation and allow medical specialists to prescribe the rehabilitation scheme. R D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 14 FR_MS_08 The system must enable alternately performed operator-independent scanning of vein in two perpendicular planes. R FR_MS_09 The system of continuous ultrasonic monitoring must ensure minimal dimensions and ease of application R FR_MS_10 The possibility to use an ultrasound transducer for conventional scanning of peripheral veins must be ensured. R FR_MS_11 The resolution of ultrasound images must enable a detection of veins up to the depth of 12 cm. R FR_MS_12 The ultrasound monitoring must supply Dopler based data on possible deviations of the vein blood flow due to DVT. R FR_MS_13 The system must have a user interface adaptable for the ultrasound monitoring module for imaging, data presentation and control. R Table 3. Functional requirements: patient level. ID Description Request (R) Wish (W) FR_P_01 The system must support user authentication. R FR_P_02 Provide daily activity goals to prevent DVT through lower limb exercises. R FR_P_03 Display real-time updates on patient-specific DVT risk levels based on wearable sensor data. R FR_P_04 Alert patients to perform preventive actions (e.g., exercises) when their DVT risk increases. R FR_P_05 Offer a serious game to motivate patients to engage in preventive exercises. R FR_P_06 Offer an extended reality environment to guide patients on performing prescribed exercises correctly. R FR_P_07 Allow patients to enter their subjective symptoms and track any changes over time. W Table 4. Functional requirements: maintenance engineer level. ID Description Request (R) Wish (W) FR_ME_01 The system must support user authentication. R FR_ME_02 Automatically record system start-up, operation, and error logs. R FR_ME_03 Notify maintenance engineers of system errors or malfunctions through an alert system. R FR_ME_04 Provide real-time system health monitoring dashboards for all system components. R FR_ME_05 Log detailed diagnostic data for troubleshooting hardware modules. W FR_ME_06 Allow remote access to system logs for offsite diagnostics and troubleshooting. W D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 15 Table 5. Functional requirements: biomedical research level. ID Description Request (R) Wish (W) FR_MR_01 The system must support user authentication. R FR_MR_02 Collect and store raw medical data from the wearable sensors (e.g., ultrasound, plethysmography). R FR_MR_03 Allow researchers to export anonymized datasets for scientific analysis. R FR_MR_04 Enable customized data collection settings for specific research protocols. W FR_MR_05 Offer AI-powered tools to identify patterns and anomalies in large datasets for DVT research. W FR_MR_06 Support the integration of external research tools for dataset analysis and processing. W Two software applications will be developed, one for medical specialist and one for the patient. 3.3. Technical requirements Technical requirements are specific criteria that outline the capabilities, functions, characteristics, parameters, and standards the ThrombUS+ system must meet to fulfil its intended purpose. The tables below present only the main technical requirements related to the system and its modules. Detailed design specifications will follow as a result of the forthcoming Task T6.1. Table 6 outlines the key requirements for the CHI, designed to implement DVT monitoring and prophylaxis methods. The CHI will support two applications: the Medical Specialist App and the Patient App, both reliant on complex computations. Additionally, the CHI will power ultrasound imaging and actuators, necessitating the use of a modern laptop with advanced specifications as the central hub. Table 6. Technical requirements: Central Hub & Intelligence module. ID Description Request (R) Wish (W) TR_CHI_01 Computer: Central hub must CPU – at least i5-13600K; GPUat least Nvidia RTX 4060; RAM – at least 16GB; ROM – at least SDD 1TB; communication – Wi-Fi and BlueTooth; battery – at least 6 cells. Display– at least 17 in. R TR_CHI_02 Medical Specialist App: Central hub must be able to run an app for a medical specialist. R TR_CHI_03 Medical Specialist App: The app must support interactive real-time measurements by using compression ultrasound (B-scan and Doppler), LRR, EIM, and IMU sensors. R TR_CHI_04 Medical Specialist App: The application must support real-time monitoring and automated data analysis of measurement data, as well as provide automatic alerts based on the machine learning DVT detection model. R TR_CHI_05 Medical Specialist App: The application must support offline measurements and visual analytics of recorded data, including risk factors, ultrasound images (B-scan & Doppler) and related parameters, light reflection rheography and electrical impedance data, physical activity, and limb orientation data, in an interactive and semi-automatic manner. R TR_CHI_06 Patient App: The central hub must be capable of running the patient application. R TR_CHI_07 Patient App: The application must be capable of running a serious game for prophylaxis. R D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 16 TR_CHI_08 Patient App: The application must be capable of presenting risk factors to the patient. R Table 7 lists main requirements for the WAM, whose purpose is to perform thigh tissue compression for the implementation of (a) the compression duplex ultrasonography (CDUS) method, (b) steering the ultrasound transducer automatically using a feedback mechanism, thereby enabling operator-independent DVT monitoring, and (c) the venous occlusion plethysmography (VOP) method using EIM and LRM as sensors. Table 7. Technical requirements: Wearable actuator module. ID Description Request (R) Wish (W) TR_WAM_01 Structure: The WAM must include (a) a C-shell type semirigid wearable textile for the integration of a detachable ultrasound transducer and air bladders, and (b) an electronic actuator controller. The shell and textile must be robust and skincompatible to ensure durability and prevent irritation during prolonged use. R TR_WAM_02 Pressure Output Range: The actuator must generate sufficient force to compress the leg for implementation of CDUS and VOP methods without causing harm. Typical range of pressure 20-150 mmHg. R TR_WAM_03 Time Intervals: The actuator must be able to apply and release pressure at specific time intervals (ranging from minutes to several hours) for prolonged monitoring periods (lasting several days), depending on the application. R TR_WAM_04 Speed and Response Time for the CDUS method: The actuator must adapt to the CDUS requirements for compression and release speed. The inflation speed of the WAM should range from 3 to 6 mmHg per second, with a pump response time of less than 0.5 seconds. R TR_WAM_05 Angular Range of Transducer Steering: The actuator must be capable of adjusting the angle of US transducer by a minimum of 5 degrees in both directions, resulting in a total angular range of 10 degrees. R TR_WAM_06 Actuator Controller: The actuator controller must have the following sensors and inputs: 1) Two pressure sensors measuring pressure in two different bladders; 2) Feedback signal from Central HUB indicating the level of compression of the vein or other summarizing parameter such as the strain of tissue compression. R TR_WAM_07 Noise Level: Operational noise shall not exceed 40 dB to ensure patient comfort during prolonged use. R TR_WAM_08 Safety Features: The actuator must include overpressure protection to prevent excessive pressure that could lead to injury and emergency stop to end the operation immediately in case of malfunction. The safety threshold pressure setting is 180 mmHg. R TR_WAM_09 Power Supply: The power supply must ensure a stable voltage of 5V and a current of at least 450 mA. It is anticipated that the actuator will be powered via a USB-C cable from the Central Hub. R TR_WAM_10 Durability and Lifetime: The actuator is a reusable device and must withstand several thousand cycles of operation. Materials used in the actuator that come into direct or indirect contact with the body must meet biocompatibility standards (e.g., ISO 10993) to avoid adverse reactions. The device should withstand multiple sterilization cycles without compromising functionality or material integrity. R TR_WAM_11 Maintenance: The actuator must provide critical information (e.g., error codes about the status of piezo pumps) about its status for logging in to the Central Hub’s log files. R D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 17 TR_WAM_12 Usage for Research: The actuator controller must have an option to provide pressure levels from each cuff for recording and exporting raw data and signal files by the Central Hub. R TR_WAM_13 Reusability: WAM must be reusable, enabling washing and disinfection of it’s parts. R Table 8 lists the main requirements for the Ultrasound Module, whose purpose is to implement the compression duplex ultrasonography (CDUS) method for DVT monitoring. Table 8. Technical requirements: Ultrasound module ID Description Request (R) Wish (W) TR_USM_01 Transducer: The linear multi-element ultrasonic transducer for automated vein scanning must have at least 128 elements and have a central frequency in the range of 5 to 7 MHz. R TR_USM_02 A composite multielement ultrasonic transducer, enabling scanning of veins in two perpendicular planes (e.g. T or Row-Column (RC) type) must ensure scanning depth up to 12 cm. R TR_USM_03 The casing of the transducer must be small and suitable for integration into the WAM and removal from it. R TR_USM_04 The casing of the transducer should allow an attachment of a handle for manual Bscan operation. W TR_USM_05 Beamformer: The beamformer must ensure B-steer scanning with linear transducer having up to 128 elements. R TR_USM_06 The beamformer must support linear scanning with RC transducer having 96 elements and phased/sectorial scanning with 32 elements R TR_USM_07 The beamformer must provide adjustable focusing during transmission in 8 zones and dynamic focusing during reception. R TR_USM_08 The beamformer must support imaging modes including B-steer for linear probes, Color Doppler, Power Doppler, Pulse Wave Doppler, and duplex in two perpendicular scanning planes (with respect of the vein direction) R TR_USM_09 The beamformer must be connected to the Central Intelligence Hub via a USB-A or USB-C connector for both power supply and data transmission. R TR_USM_10 Data formats: The data format at the output of the beamformer for B-scan, Colour Doppler, and Pulse Wave Doppler modes must be an 8-bit image. R TR_USM_11 Beamformed RF data access through an SDK library must be ensured for parametrization of ultrasound signals. W TR_USM_12 The data format at the output of the beamformer for RF data must be 16 bits sampled at a rate of no less than 40 MHz. W TR_USM_13 Detection and registration: Online detection of a vein and the adjacent artery in the B-scan image, along with the segmentation of the vein's cross-section, must be ensured including provision of real-time data on the vein’s position in the scanning plane for the operator-less feedback control of the transducer’s position. R TR_USM_14 Online tracking of vein dimensions and the level of vein closure during the application of controllable pressure must be conducted using the results of online segmentation. R D2.3 |Technical and functional requirements v2.0 | 12 Nov 2025 18 TR_USM_15 Online registration of Colour and Wave/Spectral Doppler data stream must be enabled for automatic calculation of blood flow parameters. R TR_USM_16 Ultrasound scanning plane parallel to the vein direction must be maintained by online control of transducer position ensuring operator-less Wave Doppler data registration. R TR_USM_17 The USM beamformer must provide a consolidated stream of controllable scanning parameters, images, clips and signals to CHI by USB connection. R TR_USM_18 Parameters and characterization: Quantitative parameters must be available from the wave/spectral Doppler signal, such as vein flow pulsatility, respiratory phase parameters, and the reaction of flow changes due to possible distal augmentation. R TR_USM_19 A detected thrombus must be characterized using ultrasound parameters such as echogenicity, strain elastography under applied pressure, to evaluate the thrombus’s status and age—whether it is new or chronic. W TR_USM_20 Thrombus characterization using strain elastography with RF data obtained from beamformer must be ensured. W TR_USM_21 Safety and Security: Ultrasonic Monitoring module must be safe, and performance is as intended. This includes ensuring that it does not compromise patient safety or health, and that it has an acceptable risk/benefit profile. R TR_USM_22 The Ultrasonic Monitoring Module must comply with the safety requirements and safety indices on potential thermal and mechanical hazards associated with ultrasound usage described in IEC 60601 standard. R TR_USM_23 Ultrasound transducers used for continuous monitoring should be suitable for cuffs with different sizes and be detachable from the cuff. R Table 9 lists the main requirements for the EIM which will perform impedance plethysmography measurements during venous occlusion intervention and during other, experimental interventions such as deep respiration, and provide the output impedance signal to the Central Hub of the ThrombUS+ system. Table 9. Technical requirements: Electrical impedance module. ID Description Request (R) Wish (W) TR_EIM_01 Electrode Configuration: A tetrapolar electrode configuration, where the two outer electrodes are used for current injection, and the two inner electrodes for voltage measurement, must be used. This configuration is preferred over the bipolar configuration as it minimizes confounding factors, particularly the effects of skinelectrode impedance. R TR_EIM_02 Electrode Materials: Washable electroconductive textiles must be used for all electrodes. R TR_EIM_03 Excitation Frequency: One or more excitation frequencies in the frequency band of 10 – 500 kHz. At least one of the measurement frequencies shall be in the frequency band between 50 kHz. R TR_EIM_04 Excitation Current: The injected AC current must be in the range 0.1-3mA. R TR_EIM_05 Impedance Measurement Range: The measurement range of at least 0.1 – 400 Ohm, with measurement resolution of at least 0.1 Ohm and noise level of 0.1 Ohm rms maximum. R TR_EIM_06 Data Processing: The device must be able to send raw data, along with EIM-extracted parameters, to CHI. R D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 25 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_CHI_17_01 To perform manual initialization of examinations (ultrasound, plethysmography etc) Medical Expert CHI_17 1 2024-10-30 W Interview with Medical Expert Usability evaluation / Verification UR_PR01_CHI_18_01 To export examination results, reports and any patient related data Medical Expert CHI_18 1 2024-10-30 R Interview with Medical Expert Usability evaluation / Verification UR_PR01_CHI_19_01 To allow Medical Specialists view recorded medical data offline, including detailed analytics of their health metrics. Medical Expert CHI_19 1 2024-10-30 R Interview with Medical Expert Usability evaluation / Verification UR_PR01_CHI_20_01 To allow Medical Specialists access to anonymized patient data for scientific research and analysis. Medical Expert CHI_20 1 2024-10-30 W Regulations Usability evaluation / Verification UR_PR01_CHI_21_01 To provide Medical Specialists the ability to customize data collection settings to match specific research protocols. Medical Expert CHI_21 1 2024-10-30 W Interview with Medical Expert Usability evaluation / Verification UR_PR01_CHI_22_01 To enable interoperability of patient data with certified HIS Health Provider CHI_22 1 2024-10-30 W HL7 Standard Usability evaluation / Verification UR_PR01_CHI_23_01 ThrombUS+ modules communicate with CHI either with a wire or wireless via provided appropriate programming interfaces System Admin CHI_23 1 2024-10-30 W ThrombUS+ DoA Usability evaluation / Verification Wearable Actuator Module (WAM) UR_PR01_WAM_01_01 To perform compresion of the ultrasound transducer to achieve vein compression Medical Specialists WAM_01 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_02_01 To adjust the angle of the ultrasound transducer Medical Specialists WAM_02 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 26 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_WAM_03_01 To measure the current pressure for all bladders Medical Specialists WAM_03 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_04_01 To ensure patients comfort with respect to noise Patient WAM_04 1 2024-10-30 R State of the art analysis Usability evaluation / Verification UR_PR01_WAM_05_01 To ensure protection against overpressure that could lead to injury Patient WAM_05 1 2024-10-30 R State of the art analysis Usability evaluation / Verification UR_PR01_WAM_06_01 To sufficiently powered Medical Specialists WAM_06 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_07_01 To operate robustly over prolonged time periods Patient WAM_07 1 2024-10-30 R ThrombUS+ DoA IEC 60601 standard Usability evaluation / Verification UR_PR01_WAM_08_01 To be re-used by other patients Patient WAM_08 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_09_01 To be washable Patient WAM_09 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_10_01 To be adjustable with respect to thigh circumference Patient WAM_10 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_11_01 To be biocompatible ensuring no adverse reaction during prolonged wear Patient WAM_11 1 2024-10-30 R ISO 10993 IEC 60601 Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 27 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_WAM_12_01 To log critical information System Admin WAM_12 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification UR_PR01_WAM_13_01 To record and export raw data Researchers WAM_13 1 2024-10-30 R ThrombUS+ DoA Usability evaluation / Verification Ultrasound Module (USM) UR_PR01_USM_01_01 To image with sufficient quality veins and arteries in within the lower limb to perform DVT diagnosis Medical Specialists USM_01 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_USM_02_01 To have different imaging modes in two perpendicular planes and depth of 8cm to support DVT diagnosis Medical Specialists USM_02 1 2024-10-30 R State of the art analysis Usability evaluation / Verification UR_PR01_USM_03_01 The bulk and CMUT type ultrasound probes should be suitable both for manual and operator-less scanning Medical Specialists/ Researchers USM_03 1 2024-10-30 R Interview with medical doctors Usability evaluation / Verification UR_PR01_USM_04_01 To identify and segment in real-time veins and arteries within the imaging plane Medical Specialists USM_04 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_USM_05_01 The ultrasound probe should be able to be integrated in to the WAM Medical Specialists USM_05 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_USM_06_01 To support quantitative estimation of venous flow and thrombi Medical Specialists USM_06 1 2024-10-30 W State of the art analysis Usability evaluation / Verification UR_PR01_USM_07_01 To be biocompatible ensuring no adverse reaction during prolonged wear Patients USM_07 1 2024-10-30 R ISO 10993 IEC 60601 Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 28 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method Electrical Impedance Module (EIM) UR_PR01_EIM_01_01 To provide sufficient bioimpedance signal quality and resolution to discriminate physiological from pathological condition Medical Specialists EIM_01 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_EIM_02_01 To support automatic calibration of EIM module Medical Specialists EIM_02 1 2024-10-30 W State of the art analysis Usability evaluation / Verification UR_PR01_EIM_03_01 To be biocompatible ensuring no adverse reaction during prolonged wear Patients EIM_03 1 2024-10-30 R ISO 10993 IEC 60601 Usability evaluation / Verification UR_PR01_EIM_04_01 Electrodes should have reliable contact with the skin and be reusable. Patients EIM_04 1 2024-10-30 R State of the art analysis Usability evaluation / Verification UR_PR01_EIM_05_01 To permit easy wearing procedure Patients / Medical Specialists / Physical Therapist EIM_05 1 2024-10-30 W ThombUS+ DoA Usability evaluation / Verification Light Rheography Module (LRM) UR_PR01_LRM_01_01 To provide sufficient light rheography signal quality and resolution to discriminate physiological from pathological condition Medical Specialists LRM_01 1 2024-10-30 W ThombUS+ DoA Usability evaluation / Verification UR_PR01_LRM_02_01 Sensors should be reusable Patients LRM_02 1 2024-10-30 W ThombUS+ DoA Usability evaluation / Verification UR_PR01_LRM_03_01 To be biocompatible ensuring no adverse reaction during prolonged wear Patients LRM_03 1 2024-10-30 W ISO 10993 IEC 60601 Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 29 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_LRM_04_01 To permit easy wearing procedure Patients / Medical Specialists / Physical Therapist LRM_04 1 2024-10-30 W ThombUS+ DoA Usability evaluation / Verification Limb Activity Module (LAM) UR_PR01_LAM_01_01 To provide sufficient online position and orientation signal quality with high resolution for monitoring the activity of each lower limb segment Medical Specialists/ Patients LAM_01 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_LAM_02_01 Sensors should be reusable Patients LAM_02 1 2024-10-30 R ThombUS+ DoA Usability evaluation / Verification UR_PR01_LAM_03_01 To be biocompatible and ensuring no adverse reaction during prolonged wear Patients LAM_03 1 2024-10-30 R ISO 10993 IEC 60601 Usability evaluation / Verification UR_PR01_LAM_04_01 To permit easy wearing procedure Patient / Medical Expert / Physical Therapist LAM_04 1 2024-10-30 W ThombUS+ DoA Usability evaluation / Verification UR_PR01_LAM_05_01 To support automatic calibration of LAM module Medical Specialists LAM_05 1 2024-10-30 W State of the art analysis Usability evaluation / Verification Extended Reality & Serious Gaming Module (XGM) UR_PR01_XGM_01_01 To contain a collection of predefined exercises dedicated to DVT prevention. Physical therapist XGM_01 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 30 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_XGM_02_01 To contain a collection of predefined exercises dedicated to rehabilitation after orthopedic surgery. Physical therapist XGM_02 1 2024-10-30 W Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_03_01 To contain the correct pose for the ultrasound imaging ThrombUS+ device XGM_03 1 2024-10-30 W Diagnostic guidelines Usability evaluation / Verification UR_PR01_XGM_04_01 To contain the correct pose for the Electrical Impedance Plethysmography ThrombUS+ device XGM_04 1 2024-10-30 R Diagnostic guidelines Usability evaluation / Verification UR_PR01_XGM_05_01 To contain the feet exercise (tip-toe-test) for the Light Reflection Rheography ThrombUS+ device XGM_05 1 2024-10-30 R Diagnostic guidelines Usability evaluation / Verification UR_PR01_XGM_06_01 To update the database with new predefined exercises Researcher/ System Admin / Physical therapist XGM_06 1 2024-10-30 R ThrombUS+ use cases Usability evaluation / Verification UR_PR01_XGM_07_01 To set which exercises prescribed to be performed by the patient Physical therapist XGM_07 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_08_01 To personalize the predefined exercises on patient's needs, abilities and condition via live demonstration on the patient Physical therapist XGM_08 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_09_01 To set which movements are NOT allowed to be performed by the patient Physical therapist XGM_09 1 2024-10-30 W Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_10_01 To set the prescription: number of exercises, repetitions and number of repeats per day. Physical therapist XGM_10 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 31 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_XGM_11_01 To demonstrate exercises / movements / poses as performed by a 3D avatar (visual articulations) Physical therapist / Patient XGM_11 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_12_01 To visualize patient lower limb movement in real-time as performed by a 3D avatar (visual articulations) Physical therapist / Patient XGM_12 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_13_01 To provide a visual comparison of realtime activity and the predefined/ recorded exercises Physical therapist / Patient XGM_13 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_14_01 To provide visual guidance on how to perform the allowed exercises Patient XGM_14 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_15_01 To provide audio guidance on how to perform the allowed exercises Patient XGM_15 1 2024-10-30 W Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_16_01 To record lower-limb activity Physical therapist XGM_16 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_17_01 To visualize the recorded lower-limb activity Physical therapist XGM_17 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_18_01 To notify patient when not allowed movements are detected Patient XGM_18 1 2024-10-30 W Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_19_01 To prompt patients to perform the exercises/ movements Patient XGM_19 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 32 Unique Identifier User Requirements User or Stakeholder Ref.-No. Change Index Date Request (R) Wish (W) Source Evaluation Method UR_PR01_XGM_20_01 To generate detailed reports including visual elements on patient's performance Physical therapist / Patient XGM_20 1 2024-10-30 W Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_21_01 To export recorded data Researcher / System Admin XGM_21 1 2024-10-30 R ThrombUS+ DOA Usability evaluation / Verification UR_PR01_XGM_22_01 Game should align with exercises prescribed by the physical therapist Patient XGM_22 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_23_01 To have multiple game story lines Patient XGM_23 1 2024-10-30 R ThrombUS+ DOA Usability evaluation / Verification UR_PR01_XGM_24_01 Game should incentivize patients to complete exercises correctly Patient XGM_24 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_25_01 Game should adapt based on patients improvement Patient XGM_25 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification UR_PR01_XGM_26_01 Game should adapt based on physiotherapists prescription Patient XGM_26 1 2024-10-30 R Interview with physical therapist Usability evaluation / Verification D2.3|Technical and functional requirements v1.0 | 31 Oct 2024 33 ANNEX 2 | ThrombUS+ System Requirements Version 2, Date: 12 November 2025 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) Central Hub and Intellingence (CHI) SR_PR01_CHI_01_01 The CHI provides an interface for administrators to set up application settings, manage accounts for medical specialists and define operator roles. CHI_01 1 2024-10-30 UR_PR01_CHI_01_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_02_01 The CHI has a local database that stores patient medical data (as a personal health record segment), wearable activity data and system settings CHI_02 1 2024-10-30 UR_PR01_CHI_01_01 UR_PR01_CHI_14_01 UR_PR01_CHI_19_01 Passed software test: database functionality (data integrity, completeness, etc.) SR_PR01_CHI_03_01 The CHI provides an interface to initialize the device for a new patient, including patient identification details and medical history. CHI_03 1 2024-10-30 UR_PR01_CHI_02_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_04_01 The CHI has an interface for medical experts to prescribe ultrasound, impedance plethysmography, light reflection rheography examinations, and physical therapy, including relevant examination parameters (e.g., cuff pressure level, schedule). CHI_04 1 2024-10-30 UR_PR01_CHI_03_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_05_01 The CHI has a default mechanism for automatic initialization for each prescribed examination (e.g., ultrasound, plethysmography) within the thrombus project scope CHI_05 1 2024-10-30 UR_PR01_CHI_03_01 Passed software test: application functionality SR_PR01_CHI_06_01 The CHI utilizes a background service which orchestrates the whole prescribed examination CHI_06 1 2024-10-30 UR_PR01_CHI_01_01 UR_PR01_CHI_03_01 Passed software test: application functionality D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 34 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) and rehabilitation process (XGM), records and stores the raw data via provided SDKs from each wearable component (USM, LRM, EIM, WAM, LAM). SR_PR01_CHI_07_01 The CHI has an interface to enter patient health records related to DVT risk factors CHI_07 1 2024-10-30 UR_PR01_CHI_04_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_08_01 The CHI provides patients with a visualization of their personalized DVT risk estimation based on current patient condition. CHI_08 1 2024-10-30 UR_PR01_CHI_05_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_09_01 The CHI has notification alerts for patients' upcoming examinations (including position recommendations when necessary) and physical activity goals (prevention and rehabilitation exercises) CHI_09 1 2024-10-30 UR_PR01_CHI_06_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_10_01 The CHI has an interface to visualize the current examination progress and results of ultrasound, EIP, and LRR examinations for the patient, including correct positioning in the XR environment. CHI_10 1 2024-10-30 UR_PR01_CHI_07_01 UR_PR01_CHI_08_01 UR_PR01_CHI_09_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_11_01 The CHI provides an emergency stop function to immediately halt any ongoing examination in case of detected abnormalities or patient distress. CHI_11 1 2024-10-30 UR_PR01_CHI_10_01 UR_PR01_WAM_05_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_CHI_12_01 The CHI provides the ability to start/pause/stop the rehabilitation/prevention app (XGM) CHI_12 1 2024-10-30 UR_PR01_CHI_11_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 41 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) 2) sampled at a rate of no less than 40 MHz. from known depth scan using calibrated phantoms. SR_PR01_USM_13_01 The USM provides real-time: 1) detection of a vein and the adjacent artery, 2) segmentation of detected vein's crosssection, 3) vein's position in the scanning plane for the operator-less feedback control of the transducer's position during B-scan mode. USM_13 1 2024-10-30 UR_PR01_USM_02_01 UR_PR01_USM_04_01 UR_PR01_USM_05_01 Passed software test: application functionality. Passed usability test: segmentation of detected vein SR_PR01_USM_14_01 The USM provides and stores real-time measurements for the segmented vein's: 1) dimensions 2) level of compression (closure). USM_14 1 2024-10-30 UR_PR01_USM_02_01 UR_PR01_USM_04_01 UR_PR01_USM_05_01 Passed verification of vein closure during compression by pilot in-vivo trial SR_PR01_USM_15_01 The USM provides online registration of color and wave/spectral Doppler data stream enabling evaluation of blood flow parameters. USM_15 1 2024-10-30 UR_PR01_USM_01_01 UR_PR01_USM_02_01 UR_PR01_USM_06_01 Passed functionality test: availability of Doppler data Passed software test: performance of algorithm for blood flow evaluation SR_PR01_USM_16_01 The USM has the ability for online control of the transducer's orientation during ultrasound scanning to maintain the imaging plane parallel to the vein's direction, ensuring operatorindependent wave-Doppler data registration. USM_16 1 2024-10-30 UR_PR01_USM_01_01 UR_PR01_USM_02_01 UR_PR01_USM_04_01 Passed usability test verifying the control of the scanning plane position enabling the imaging of vein SR_PR01_USM_17_01 The USM beamformer must provide a consolidated stream of controllable scanning parameters, images, clips and signals to CHI by USB connection. USM_17 1 2024-10-30 UR_PR01_USM_01_01 UR_PR01_USM_02_01 UR_PR01_CHI_07_01 Passed usability test: SR_PR01_USM_18_01 [Optional] USM will provide quantitative data from wave/spectral Doppler signal, such as: USM_18 1 2024-10-30 UR_PR01_USM_06_01 (Optional, dependent on necessary research results) Passed D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 42 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) 1) vein flow pulsatility, 2) respiratory phasicity, 3) vein flow changes (e.g. due to compression on distal limb). functionality test of principal ability to obtain vein flow data with sufficient signal to noise ratio and subsequent evaluation of quantitative flow data. SR_PR01_USM_19_01 [Optional] The USM characterizes a detected thrombus (new or chronic) by its: 1) echogenicity, 2) strain elastography under applied pressure, 3) strain elastography with RF data. USM_19 1 2024-10-30 UR_PR01_USM_06_01 (Optional, dependent on necessary research results) Passed usability test on phantoms and pilot in-vivo trials SR_PR01_USM_20_01 The USM complies with IEC 60601 standard "Particular requirements for the basic safety and essential performance of ultrasonic medical diagnostic and monitoring equipment" including the safety indices on potential thermal and mechanical hazards associated with the wearable ultrasound usage. USM_20 1 2024-10-30 UR_PR01_USM_07_01 Passed compliance test for safety and essential performance SR_PR01_USM_21_01 The USM provides an SDK for raw ultrasound RF signal and data processing and programming interfaces. USM_21 1 2024-10-30 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: The SDK fulfills the requirements for command execution, data communication, processing, and programming functionality. Electrical Impedance Module (EIM) SR_PR01_EIM_1_01 The EIM contains of a tetrapolar electrode where: 1) the two outer electrodes are used for current injection, and 2) the two inner electrodes for voltage measurement. EIM_1 1 2024-10-30 UR_PR01_EIM_01_01 Passed functionality test: the tetrapolar electrode system is available for accurate bioimpedance measurements D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 43 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) SR_PR01_EIM_2_01 The EIM uses excitation frequencies in the frequency band of 10-500kHz with at least 5kHz discrete resolution. EIM_2 1 2024-10-30 UR_PR01_EIM_01_01 Passed performance test: Excitation validation with oscilloscope, ± 3% acceptable. SR_PR01_EIM_3_01 The EIM injects AC current in the range of 0.13.0mA EIM_3 1 2024-10-30 UR_PR01_EIM_01_01 Passed performance and safety tests: verification with reference resistors, ± 5% acceptable. SR_PR01_EIM_4_01 The EIM supports measurements in the range of 0.1-400 Ohm. EIM_4 1 2024-10-30 UR_PR01_EIM_01_01 Passed performance test: verification with test circuit, ± 5% acceptable. SR_PR01_EIM_5_01 The EIM supports measurements with a resolution of at least 0.1 Ohm and a noise level of 0.1 Ohm RMS (root mean squared) maximum. EIM_5 1 2024-10-30 UR_PR01_EIM_01_01 Passed performance test: noise level verification via test circuit. SR_PR01_EIM_6_01 The EIM measurements are accurate to within 12% to detect small physiological changes reliably. EIM_6 1 2024-10-30 UR_PR01_EIM_01_01 Passed performance test: verified with tissue modeling electronic circuit. SR_PR01_EIM_7_01 The EIM can transmit raw data and EIP-extracted parameters to the CHI module via the defined communication protocol specified in D6.1 "Specifications", over WiFi. Bluetooth is available in the hardware. EIM_7 1 2024-10-30 UR_PR01_EIM_01_01 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: The EIM can communicate with the CHI and transmit data via WiFi. SR_PR01_EIM_8_01 The EIM uses a rechargable battery for at least 2 days of intermittent monitoring with 5 VOP measurements done each day. EIM_8 1 2024-10-30 UR_PR01_CHI_23_01 UR_PR01_EIM_02_01 Passed performance test: verification via current consumption measurement and simulated use. Accepted value at minimum 5 VOP measurement in 2 days. SR_PR01_EIM_9_01 [Optional] The EIM supports automatic internal calibration. EIM_9 1 2024-10-30 UR_PR01_EIM_02_01 Passed functionality test: self-calibration is performed D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 44 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) before each intermittent measurement. SR_PR01_EIM_10_01 The EIM complies with the IEC 60601-1 electrical medical standards. EIM_10 1 2024-10-30 UR_PR01_EIM_03_01 Passed performance and safety tests: all standard requirements are fulfilled. SR_PR01_EIM_11_01 The EIM does not interfere with other medical equipment and ensures its reliable operation in noisy environments without degradation in performance. EIM_11 1 2024-10-30 UR_PR01_EIM_03_01 Passed performance test: all standard requirements are fulfilled. SR_PR01_EIM_12_01 Electrodes of the EIM are made of washable electroconductive textiles. EIM_12 1 2024-10-30 UR_PR01_EIM_04_01 Passed performance test: washing electroconductive textiles does not change EIM performance characteristics. SR_PR01_EIM_13_01 Electrodes of the EIM have adjustable circumference. EIM_13 1 2024-10-30 UR_PR01_EIM_04_01 Passed usability test: evaluation by users. SR_PR01_EIM_14_01 The EIM provides an SDK for data processing and programming interfaces, specified in D6.1 'Specifications'. EIM_14 1 2024-10-30 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: The SDK fulfills the requirements for command execution, data communication, processing, and programming functionality. SR_PR01_EIM_15_01 The EIM is integrated into a biocompatible wearable with the LRM and the LAM EIM_15 1 2024-10-31 UR_PR01_EIM_03_01 UR_PR01_EIM_04_01 UR_PR01_ EIM_05_01 Passed safety tests: the EIM fits into the biocompatible wearable. Light Rheography Module (LRM) D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 45 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) SR_PR01_LRM_01_01 The LRM has sampling rate not less than 100Hz. LRM_01 1 2024-10-30 UR_PR01_LRM_01_01 Passed performance test: the sampling rate is equal to 100Hz. SR_PR01_LRM_02_01 The LRM has a two-wavelength (red and infrared) light reflection rheography sensor. LRM_02 1 2024-10-30 UR_PR01_LRM_01_01 Passed performance test: the LRM sensor has twowavelength (red and infrared) LEDs. SR_PR01_LRM_03_01 The LRM can transmit raw data and extracted parameters to the CHI module via the defined communication protocol specified in D6.1 "Specifications" over WiFi. Bluetooth is available in the hardware. LRM_03 1 2024-10-30 UR_PR01_LRM_01_01 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: The LRM can communicate with the CHI and transmit data via WiFi. SR_PR01_LRM_04_01 The LRM is integrated into a biocompatible and washable wearable with the EIP and the LAM. LRM_04 1 2024-10-30 UR_PR01_LRM_02_01 UR_PR01_LRM_03_01 UR_PR01_LRM_04_01 Passed safety tests: the LRM fits into the biocompatible wearable. SR_PR01_LRM_05_01 The LRM uses a rechargeable battery for at least 2 days of intermittent monitoring. LRM_05 1 2024-10-30 UR_PR01_CHI_23_01 UR_PR01_LRM_02_01 Passed performance test: verified through current consumption measurement and simulated use. The accepted value is a minimum of 2 days of intermittent measurements. SR_PR01_LRM_06_01 The LRM complies with the safety indices on potential thermal and mechanical hazards as described in IEC 60601-1 standard. LRM_06 1 2024-10-30 UR_PR01_LRM_03_01 Passed performance and safety tests: all standard requirements are fulfilled. SR_PR01_LRM_07_01 The LRM provides an SDK for data processing and programming interfaces, specified in D6.1 'Specifications'. LRM_07 1 2024-10-30 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: The SDK fulfills the requirements for command execution, data communication, processing, and programming functionality. D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 46 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) Limb Activity Module (LAM) SR_PR01_LAM_01_01 The LAM module integrates inertial measurement unit (IMU) and optionally photoplethysmography (PPG) sensors. LAM_01 1 2024-10-30 UR_PR01_LAM_01_01 Passed functionality test: the LAM module integrates IMU and PPG sensors. SR_PR01_LAM_02_01 The LAM has a minimum sampling rate of 50Hz for activity and orientation measurements LAM_02 1 2024-10-30 UR_PR01_LAM_01_01 Passed performance test: the sampling rate is equal to 100Hz. SR_PR01_LAM_03_01 The LAM measurements are in the range of: 1) accelerometer (3 axes): ±2 to ±16 g, 2) gyroscope (3 axes): ±250 to ±2000 deg/s, 3) magnetometer (3 axes): ±4900 μT LAM_03 1 2024-10-30 UR_PR01_LAM_01_01 Passed performance test: Validation against a reference (electromagnetic orientation and position measuring device) is acceptable within ±3%. SR_PR01_LAM_04_01 The LAM consists of a minimum of three IMU sensors per leg, with one sensors per leg segment: 1) foot, 2) calf, 3) thigh. LAM_04 1 2024-10-30 UR_PR01_LAM_01_01 Passed functionality test: The LAM module integrates at least three IMU sensors per leg and one sensor per leg segment. SR_PR01_LAM_05_01 [Optional] The LAM contains an extra IMU sensor placed on torso. LAM_05 1 2024-10-30 UR_PR01_LAM_01_01 Passed functionality test: the LAM module integrates an additional IMU torso sensor for zero calibration. SR_PR01_LAM_06_01 [Optional] The LAM contains extra IMU sensors for redudancy and data quality enhancement. LAM_06 1 2024-10-30 UR_PR01_LAM_01_01 Passed functionality test: the LAM module integrates an additional IMU sensor for data quality enhancement. SR_PR01_LAM_07_01 For each connected IMU sensor the LAM outputs: 1) 3D orientation data (in degrees, radians, quaternions), LAM_07 1 2024-10-30 UR_PR01_LAM_01_01 Passed functionality test: for each connected IMU sensor, the LAM outputs and CHI receives D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 47 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) 2) raw sensors data: a) 3-axes acceleration (m/s2) b) 3-axes angular velocity (deg/s) c) 3-axes magnetic flux (μT) 3D orientation data and raw sensor data. SR_PR01_LAM_08_01 The LAM communicates with the CHI via WiFi LAM_08 1 2024-10-30 UR_PR01_CHI_23_01 Passed functionality test: the settings, commands and data are successfully communicated to and from the LAM to the CHI via WiFi. SR_PR01_LAM_09_01 LAM uses a rechargeable battery for at least 2 days of intermittent monitoring LAM_09 1 2024-10-30 UR_PR01_LAM_02_01 Passed performance test: verified through current consumption measurement and simulated use. The accepted value is a minimum of 2 days of intermittent measurements. SR_PR01_LAM_10_01 The LAM complies with the safety indices on potential thermal and mechanical hazards as described in the IEC 60601-1 standard. LAM_10 1 2024-10-30 UR_PR01_LAM_03_01 Passed performance and safety tests: all standard requirements are fulfilled. SR_PR01_LAM_11_01 The LAM provides an SDK for data processing and programming interfaces. LAM_11 1 2024-10-30 UR_PR01_CHI_23_01 Passed functionality and compatibility tests: the SDK fulfills the requirements for command execution, data communication, processing, and programming functionality. SR_PR01_LAM_12_01 The LAM is integrated into a biocompatible and washable wearable with the EIP and the LRM. LAM_12 1 2024-10-30 UR_PR01_LAM_02_01 UR_PR01_LAM_03_01 UR_PR01_ LAM_04_01 Passed safety tests: the LAM fits into the biocompatible wearable. Extended Reality & Serious Gaming Module (XGM) D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 48 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) SR_PR01_XGM_01_01 The XGM has a database containing lower limb exercises (at least 50) and poses which are recommended for: 1) DVT prevention/ rehabilitation, 2) rehabilitation after orthopedic surgery, 3) US, EIM and LRM examination, 4) [Optional] rehabilitation of patients defined in the ThrombUS+ Use Cases (D2.1). Each exercise is represented by: 1) IMU data 2) [Optional] video 3) metadata (name of exercise, description, textual guidance, indication (ICD11), photo, keywords, can be used in the serious game, etc.). XGM_01 1 2024-10-30 UR_PR01_XGM_01_01 UR_PR01_XGM_02_01 UR_PR01_XGM_03_01 UR_PR01_XGM_04_01 UR_PR01_XGM_05_01 UR_PR01_XGM_06_01 Passed software test: database functionality (data integrity, completeness, etc.) SR_PR01_XGM_02_01 The XGM has a private database per patient (as a personal health record segment) that stores exercise prescriptions, personalized exercises, recordings of lower-limb activity, patient progress calculated data, and XR/SG state data. XGM_02 1 2024-10-30 UR_PR01_XGM_07_01 UR_PR01_XGM_08_01 UR_PR01_XGM_09_01 UR_PR01_XGM_10_01 Passed software test: database functionality (accessibility, data integrity, completeness, etc.) SR_PR01_XGM_03_01 The XGM has an interface for recording exercises: 1) start/ stop recording option, 2) accept/decline recording option, 3) meta data input (name of exercises, description, photo, keywords, can be used in the serious game, etc.) XGM_03 1 2024-10-30 UR_PR01_XGM_01_01 UR_PR01_XGM_02_01 UR_PR01_XGM_03_01 UR_PR01_XGM_04_01 UR_PR01_XGM_05_01 UR_PR01_XGM_06_01 UR_PR01_XGM_16_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_XGM_04_01 The XGM has an interface to be used by the physiotherapists to prescribe rehabilitation schemes, such as: XGM_04 1 2024-10-30 UR_PR01_XGM_07_01 UR_PR01_XGM_08_01 UR_PR01_XGM_09_01 Passed software test: application functionality, D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 49 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) 1) browse/search, preview and select exercises, 2) set the sequence of exercises, 3) for each exercise: a) number of repetitions, b) number of sets, c) record a demonstration of exercise on patient (set the goal for exercise - personalization of the exercise), 4) visualize lower limb in real time during demonstration, 5) set the number of scheme repetition per day, 6) set the duration for the whole scheme, 7) set multiple schemes for the whole rehabilitation program (calendar), 8) to set prohibited exercises. UR_PR01_XGM_10_01 UR_PR01_XGM_11_01 UR_PR01_XGM_12_01 UR_PR01_XGM_13_01 Passed usability test: application interface evaluation by users SR_PR01_XGM_05_01 The XGM has an interface to select from the prescribed exercises and the ability to: 1) visualize in 3D each exercise, 2) visualize in 3D the current lower limb pose (real-time), 3) visualize in 3D an overlap / comparison of real-time activity and the recorded exercise, 4) provide guidance on how to perform each exercise, 5) notify the patient if prohibited exercise is detected. XGM_05 1 2024-10-30 UR_PR01_XGM_11_01 UR_PR01_XGM_12_01 UR_PR01_XGM_13_01 UR_PR01_XGM_14_01 UR_PR01_XGM_15_01 UR_PR01_XGM_16_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_XGM_06_01 The XGM is controlled (starts/stops/pause) on demand by the central intelligence application XGM_06 1 2024-10-30 UR_PR01_XGM_18_01 UR_PR01_XGM_19_01 Passed software test: start/stop/pause EXG system D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 50 Unique Identifier System Requirement Ref No. Change Index Date Ref. to User Requirement Acceptance Criteria (including Test Procedure, as applicable) through the central intellignece application SR_PR01_XGM_07_01 The XGM captures all data that are related to exercises and application metadata when application runs for future activity visualization and reporting. XGM_07 1 2024-10-30 UR_PR01_XGM_16_01 UR_PR01_XGM_17_01 UR_PR01_XGM_18_01 UR_PR01_XGM_19_01 UR_PR01_XGM_20_01 UR_PR01_XGM_21_01 Passed software test: service functionality SR_PR01_XGM_08_01 The XGM has an interface to: 1) display reports including visual elements on patient's progress (including activity during gaming), 2) to export stored/recorded data. XGM_08 1 2024-10-30 UR_PR01_XGM_20_01 UR_PR01_XGM_21_01 Passed software test: application functionality, Passed usability test: application interface evaluation by users SR_PR01_XGM_09_01 The XGM has at least 3 different story lines. XGM_09 1 2024-10-30 UR_PR01_XGM_23_01 Passed software test: game functionality, Passed usability test: game evaluation by users SR_PR01_XGM_10_01 The game adapts the level of difficulty based on the prescription scheme (e.g. number of repetitions, sets, duration, etc.) and the patient's progress. XGM_10 1 2024-10-30 UR_PR01_XGM_22_01 UR_PR01_XGM_24_01 UR_PR01_XGM_25_01 UR_PR01_XGM_26_01 Passed software test: game functionality SR_PR01_XGM_11_01 Each story line accepts as input at least 5 exercises (i.e., as HCI commands). XGM_11 1 2024-10-30 UR_PR01_XGM_22_01 UR_PR01_XGM_26_01 Passed software test: game functionality, Passed usability test: game evaluation by users SR_PR01_XGM_12_01 Game engagement is supported for each one of the 3 lower limb segments for at least one exercise and at least one story line XGM_12 1 2025-11-12 UR_PR01_XGM_22_01 UR_PR01_XGM_26_01 Passed software test: game functionality, Passed usability test: game evaluation by users D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 57 2.4. Physiology-guided stimulation It was hypothesized that the effectiveness of the compression device could be further improved by adaptively synchronizing the prophylaxis device with the heart rate 32 or respiration rate31. In the first case (Figure 6), a significant increase in blood velocity (56%) was found when the stimulation duration reached 56% of the beat-to-beat interval. Figure 6. Schematic representation of the heart rate adaptive prophylaxis device (left), duplex ultrasonography, with a ratio of stimulation duration to beat-to-beat interval of 56% (right) 32. Breathing affects venous return 33 , 34 by increasing venous blood velocity during inhalation; therefore, it could be used for the adaptation of stimulation. Adaptation to respiration rate is more medically accepted for device activation due to the decreased stimulation frequency and reduced discomfort for patients. 2.5. Exercises Another method of DVT prophylaxis after surgery involves patients performing prescribed exercises according to given instructions. In the study 35 , patients were asked to perform an ankle movement sequence consisting of 20° dorsal flexion, 30° varus, 40° plantar flexion, and 30° valgus flexion at a frequency of 30 times per minute, with 8-minute cycles, 1 cycle every 30 minutes, and 20 cycles each day (for 2 weeks postoperation: 7 cycles in the morning, 7 in the afternoon, and 6 in the evening). The control group (not performing exercises) had a significantly higher overall incidence of DVT (8 with asymptomatic DVT, n=97) compared to the case group (1 with asymptomatic DVT, n=96). On the 11th and 14th days of the experiment, the case group showed significantly higher maximum venous outflow and maximum venous capacity than the control group. Regarding the preferred frequency of flexion, the study 36 showed no difference in effects between the traditional approach of 3 cycles/min (holding flexion for 10 seconds) and 30 cycles/min. However, the patients expressed that performing flexion at a frequency of 30 cycles/min is more comfortable and causes less limb fatigue. 32 Weyer et al. RheoStim: Development of an Adaptive Multi-Sensor to Prevent Venous Stasis. Sensors. 2016; 16(4):428. https://doi.org/10.3390/s16040428 33 Kwon et al. Effects of ankle exercise combined with deep breathing on blood flow velocity in the femoral vein. Aust J Physiother. 2003;49(4):253-258. https://doi.org/10.1016/s0004-9514(14)60141-0 34 Zsenák et al. Effect of active and passive techniques used in thromboembolic prophylaxis on venous flow velocity in the post-procedure period. Frontiers in physiology vol. 15 1323840. 27 Mar. 2024, https://doi.org/10.3389/fphys.2024.1323840 35 Li et al. Active Ankle Movements Prevent Formation of Lower-Extremity Deep Venous Thrombosis After Orthopedic Surgery. Medical science monitor: international medical journal of experimental and clinical research vol. 22 3169-76. 7 Sep. 2016, https://doi.org/10.12659/msm.896911 36 Li et al. Which Frequency of Ankle Pump Exercise Should Be Chosen for the Prophylaxis of Deep Vein Thrombosis? INQUIRY: The Journal of Health Care Organization, Provision, and Financing. 2022;59. https://doi.org/10.1177/00469580221105989 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 58 2.6. Serious games The risk of thrombosis is greatly increased in patients who are immobilized and bedridden. In such cases, physiotherapists prescribe clinical exercises to reduce the risk of thrombosis, together with anticoagulant treatment. Depending on the patient status and comorbidities, these exercises may be combined with other rehabilitation regimes. Game-based exercises (exergames) could be used to increase patient compliance with ankle and foot exercise-based DVT prevention 37 . By aligning the game mechanics with the clinical exercises prescribed by the physiotherapist, the serious game incentivizes patients to perform the essential physiotherapy exercises for DVT. These game elements reinforce motivation through achievements and rewards, transforming the rehabilitation process into an interactive journey of achievement. The justification for the use of extended reality lies in its potential to go beyond traditional exercise methods. Through interactive experiences, patients can grasp the mechanics of movement more intuitively and strengthen the link between movement and thrombosis prevention. The study 38 utilized a wireless foot physical activity sensor (LEGSys; BioSensics LLC, Watertown, Mass) to navigate a computer cursor on a screen and control the game. After the game finished, the investigators measured each patient’s mean flow volume, peak flow velocity, mean flow velocity, and the cross-sectional area of the right femoral vein using ultrasound. They found enhanced blood flow in the femoral vein, with mean flow volume, mean flow velocity, and peak systolic velocity increasing by around 50% above baseline. Less forceful but more consistent contractions were found to be the most effective in improving venous blood flow. Serious games have started to yield important results in healthcare and especially in physical rehabilitation where motivation and frequent patient compliance is crucial for the success of a therapeutic plan. These games exploit gamification approaches to transform the boring physiotherapy exercises into engaging activities, where entertainment and therapeutic aims are hand in hand. Here, we review the serious games already based on rehabilitation and therapy, offering technological and clinical solutions in the field. Each game is designed with a particular rehabilitation or therapy need in mind: to be played with stroke or brain-injury patients, with surgery patients, with patients suffering from neurological conditions, even with elderly patients who just need some mobility rehab. Many of those games are played on PCs or mobile phones, and several games are even coupled with Microsoft’s Kinect or VR for more immersive play. Most frequently, these games feature a potent combo of high-tech sensing technologies, such as motion tracking, VR, AI and wearable sensors and moreover with exoskeletons. The input methods, however, are varied: the motion sensors and wearable devices are common entrant technologies alongside the neural sensors, but standard PC/mobile controls can be deployed, too. In the following, we will present representative games that relate to rehabilitation. − MIRA 39 : This is a physical rehabilitation platform designed for elderly patients and individuals recovering from strokes or surgeries. Adapting to patient movements through its advanced motion sensors, it uses artificial intelligence algorithms to follow users through the specified movements and ensure good execution of their prescribed exercises. Using cutting-edge gamification techniques, MIRA replaces traditional at-home physiotherapy exercises with interactive gamified products, thereby addressing the common problem that leads patients to readily skip their home workout sessions, providing both 37 Shimizu et al. A novel exercise device for venous thromboembolism prophylaxis improves venous flow in bed versus ankle movement exercises in healthy volunteers. Journal of orthopaedic surgery (Hong Kong) vol. 25,3 (2017): 2309499017739477. https://doi.org/10.1177/2309499017739477 38 Rahemi et al. Pilot study evaluating the efficacy of exergaming for the prevention of deep venous thrombosis. Journal of vascular surgery. Venous and lymphatic disorders vol. 6,2 (2018): 146-153. https://doi.org/10.1016/j.jvsv.2017.08.019 39 MIRA Rehab. [Online]. Available: https://www.fitness-gaming.com/profiles/company/mira-rehab.html D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 59 individualized immediate feedback adjusted to the patient’s performance and generating automatically gathered data to allow remote monitoring of progress by healthcare professionals. Example screenshot is shown in Figure 7. Figure 7. An indicative image of a user interacting with MIRA Rehabilitation Platform 40 . − Jintronix 41 : is a motion-based rehabilitation platform for stroke survivors and post-surgery patients. It tracks movement-usually through Microsoft Kinect-and guides the patient on how to make their movements better while doing exercises for physical therapy. The mission of Jintronix is to make rehabilitation more accessible, easy, and gamified to increase adherence with home-based exercises and to allow therapists to track improvements over time. An example screenshot is shown in Figure 8. Figure 8. An indicative image of a user interacting with Jintronix 42 user interface. − Kinapsys 43 : is a 3D motion-tracking rehabilitation system designed for patients with motor impairments. It is a personalized treatment plan that leads patients through exercises using a PC-based platform. It gives therapists real-time feedback with comprehensive tracking to track recovery and modify treatment 40 MIRA Rehabilitation Platform Lets Patients Play Their Way to Recovery. [Online]. Available: https://www.fitnessgaming.com/news/health-and-rehab/mira-rehabilitation-platform-lets-patients-play-their-way-to-recovery.html 41 Jintronix. [Online]. Available: https://jintronix.com/ 42 Jintronix. [Online]. Available: https://hcfinc.com/rehabilitation/specialized-services/jintronix/ 43 Kinapsys. [Online]. Available: https://www.medicareliefcenter.com/content/kinapsys-0 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 60 strategies. Embedded game elements keep users engaged and compliant with prescribed therapy routines. Example screenshot is shown in Figure 9. Figure 9. An indicative image showing a user interacting with Kinapsys43 software and some indicative images of the virtual environment options. − MindMaze 44 : is a different neuro-rehabilitation game that focuses on stroke patients. It uses VR and neural sensors to immerse users in virtual environments, activating brain plasticity and recovering motor functions. The aim was to expose patients to real-life tasks through the platform to regain control over their movements and cognitive abilities. Its scope is more significant than traditional therapy because it uses extended reality tuned with AI for faster rehabilitation. Figure 10 shows an example screenshot. Figure 10. An indicative image showing a user interacting with MindMaze 45 . − Neofect Smart Balance 46 : focuses on patients’ balance rehabilitation after a stroke or a neurological condition. This platform is equipped with wearable sensors and AI, following the movements of patients in tracking exercises accurately challenging leg lifts and ankle rotations. Its gamification and real-time feedback increase patient motivation and adherence to a therapy plan. Example screenshot is shown in Figure 11. 44 Mindmaze. [Online]. Available: https://mindmaze.com/ 45 Mindmaze. [Online]. Available: https://mindmaze.com/digital-therapies-for-neurorehabilitation/ 46 Smart Balance. [Online]. Available: https://www.neofect.com/us/smart-balance D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 61 Figure 11. An indicative image of a user interacting with Neofect Smart Balance46. − Virtual Rehab 47 : is an innovative neurorehabilitation platform that combines motion sensors with Kinect in a very interactive way. Virtual Rehab targets patients diagnosed with Parkinson’s, multiple sclerosis, or stroke and transforms physical therapy exercises into appealing tasks a patient will have to go through either at home or in the clinic. It also provides real-time feedback while allowing professionals to monitor their condition remotely. Example screenshot is shown in Figure 12. Figure 12. An indicative image of a user interacting with the Virtual Rehab 48 platform. − FysioGaming 49 : offers to patients undergoing physical therapy personalized, gamified rehabilitation exercises of. It converts conventional routines of therapy into entertaining games that encourage patients to be more proactive in treatment. It applies the use of motion sensors in monitoring exercise 47 Virtual Reality in Healthcare: a New Solution for Rehabilitation? [Online]. Available from: https://programace.com/blog/virtual-reality-in-healthcare-a-new-solution-for-rehabilitation/ 48 Virtual Reality Rehabilitation System. [Online]. Available: https://sanraffaele.it/en/technology/virtual-realityrehabilitation-system/ 49 FysioGaming Expands Rehabilitation Options with Kinect Games. [Online]. Available from: https://images.app.goo.gl/YfLMhkdWXapHNvqJ9 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 62 performance and provides data to therapists for recovery progress monitoring and adjustment of the course of therapy when necessary. Example screenshot is shown in Figure 13. Figure 13 An indicative image showing a virtual scene within the FysioGaming49 platform. − Playwork 50 : offers physical therapy exercises designed for rehabilitation through motion-tracking technology. By using Kinect sensors, it creates interactive sessions where patients can engage in exercises tailored to their needs. The work plays mode targets generally physical therapy, including also exercises for the elderly or those recovering from injuries, helping them regain mobility through fun, game-based therapy. Example screenshot is shown in Figure 14. Figure 14 An image showing a user performing physical exercises using the Playwork 51 . Data are streamed and visualized in a portable display. − SWORD Phoenix 52 : is an online platform focused on post-surgery rehabilitation, utilizing AI and movement sensors in wearables to lead patients through their exercises that have been prescribed by a physiotherapist. Emphasizing remote care, this platform enables patients to recover in the comfort of their home and receive real-time feedback from both users and healthcare professionals. Example screenshot is shown in Figure 15. 50 Smart PLAYBALL by PLAYWORK: The revolutionary Therapy Ball. [Online]. Available: https://www.playwork.me/playball 51 Introducing the Pilates 25.5"/65cm Smart PLAYBALL: Revolutionizing Recovery and Exercise. [Online]. Available: https://www.playwork.me/post/introducing-the-pilates-25-5--65cm-smart-playball-revolutionizing-recovery-andexercise 52 Sword Health Introduces Phoenix, the First AI Care Specialist and Raises $130 Million in a Mix of Primary and Secondary Sale, Increasing Valuation to $3 Billion. [Online]. Available: https://swordhealth.com/newsroom/introducing-phoenix D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 63 Figure 15. An indicative image of the SWORD Phoenix52 software, showing a user interacting with the platform. − Keeogo 53 : is a powered walking exoskeleton that assists individuals with mobility problems by helping the weakening muscles of one’s legs through a supporting device that enhances the movement of its wearer. Keeogo, in particular, is an indication in the rehabilitation of patients with impaired gait and mobility because it could have them return to ambulation in a shorter period and with less physiological stress for the patient. An example screenshot is shown in Figure 16. Figure 16. Image of the Keeogo53 exoskeleton system. Real-time feedback is another feature of all these games, enabling the user and healthcare professional to monitor the progress instantaneously. Most of the games are single player, though some have options for playing as multiplayer to further social or competitive aspects in treatment. All collect active user data for rehabilitation progress tracking and may be used in research or further customization of treatment. Most current serious games in health, focused on rehabilitation and physical therapy, are compared in Table 13. 53 SMART Powered Knee Orthosis. [Online]. Available: https://b-temia.com/keeogo/ D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 64 Table 13. Comparison among available serious games. Game Title Goal Target Audience Platform Technology Used Input Method Realtime Feedback No of players Data Collection Year MIRA Physical Therapy Elderly Patients PC Motion Sensors, AI Motion Sensors Yes Single Yes 2014 Jintronix Rehabilitation Post-Surgery Stroke, PostSurgery PC, Kinect Motion Tracking, AI Motion Sensors Yes Single, Multi Yes 2012 Kinapsys Physical Rehabilitation Patients with Motor Issues PC 3D Motion Tracking Motion Sensors Yes Single Yes 2013 MindMaze Stroke Rehabilitation Stroke Patients VR VR, AI, Neural Sensors Neural Sensors, VR Yes Single Yes 2017 Neofect Smart Balance Balance Training for Rehab Stroke, Neurological Patients PC, Mobile Wearable Sensors, AI Sensors Yes Single Yes 2019 VirtualRehab Neurological Rehabilitation Stroke, Parkinson’s Patients PC, Kinect Motion Sensors Kinect Motion Sensors Yes Single, Multi Yes 2012 FysioGaming Rehabilitation Exercises Physical Therapy Patients PC, Mobile Game-based Rehab Motion Sensors Yes Single Yes 2011 Playwork Physical Therapy and Exercise Patients and Elderly PC, Kinect Motion Tracking Kinect Motion Sensors Yes Single Yes 2015 SWORD Phoenix Physical Therapy Patients Recovering from Surgery Mobile AI, Digital Therapy Sensors, Mobile Device Yes Single Yes 2015 Keeogo (B-TEMIA) Exoskeleton Technology People with Mobility Issues Wearable Exoskeleton Technology Wearable Exoskeleton Yes Single Yes 2016 The research on sensor-based visualization tools for rehabilitation has focused on different types of technology or methodology, these include wearable sensors (e.g., Inertial Measurement Units, IMUs), Electromyography (EMG) signal captured by biometric sensors active electrodes available as T-Shirts, accelerometers data measuring wheelchair kinematics etc., optical systems such as Kinect from Microsoft or depth cameras and multisensor fusions systems. These tools are essential in providing snapshot views of the patient at a time of immediate concern for yield better outputs to clinicians, patients alike. These programs are developed to enhance the effectiveness of physical therapy exercises, educate patients and encourage compliance in addition to support personalized treatment plans. 2.7. Extended reality The review of wearable sensors applied in healthcare with a specific focus on rehabilitation was given by Pantelopoulos & Bourbakis 54 . In this paper, the authors detail which information can be obtained from patient movement data (e.g., accelerometers, gyroscopes, EMG sensors) in real-time to correct incorrect 54 Pantelopoulos A, Bourbakis N. A survey on wearable biosensor systems for health monitoring. Annu Int Conf IEEE Eng Med Biol Soc. 2008;2008:4887-90. doi: 10.1109/IEMBS.2008.4650309. PMID: 19163812. D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 65 exercise performances. Beshara et al. 55 utilized motion sensors such as IMUs and the Microsoft Kinect for capturing patient movements, with the data being visualized to aid in modifying cises based on their performance. Porciuncula and colleagues 56 proposed a multi-wearable sensor system measure with user-specific data modalities, such as IMUs and EMG for continuous monitoring of body motion patterns to assist both the therapists remotely track the rehabilitation progress and adapt therapy according to patient needs. CuestaVargas et al. 57 explained the role of inertial sensors as a tool to monitor the human kinematics behaviour in rehabilitation, showing that the visualization of instantaneous data acquired by Inertial Sensors can help patients even in real-time to improve posture and range of motion. Li et al. 58 developed an IMU-enabled tracking and display system for patient posture, that can provide realtime corrective feedback to ensure proper alignment during exercise. Da Gama et al. 59 used Microsoft Kinect, an optical depth sensor for monitoring patients’ movement without the need to use any additional hardware on patient’s body and giving the feedback in real time of their joining angles and body performance. Similarly, Liao et al. 60 performed the evaluation of rehabilitation systems in real time using depth sensors, such as Kinect and Intel RealSense sensing technologies for tracking and visualization of movements in different therapies, mainly aiming at physical therapy after ischemic stroke. The effectiveness of electromyographic biofeedback in rehabilitation was studied using EMG sensors that can monitor muscle activity further providing feedback on patterns of muscle activation essential for neuromuscular rehabilitation 61 . Patel et al. 62 provided an overview of wearable EMG sensors and results in the field, including post-stroke recovery and help with sports injuries—from muscle engagement in crutches after ultrasounds to compare rehabilitation techniques using EMG visualization to adjust therapy plans. To analyse the movement of patients during physical rehabilitation, fusion of multiple sensors was tried for example inertial, optical and EMG sensors to achieve a higher accuracy and consistency in getting deeper insight into patient movements 63 . Moreover, implants how to extract complex movements using wearable sensors, such as IMUs, EMG and accelerometers combined along a well-defined way of the spine and over 55 Beshara P, Anderson DB, Pelletier M, Walsh WR. The Reliability of the Microsoft Kinect and Ambulatory SensorBased Motion Tracking Devices to Measure Shoulder Range-of-Motion: A Systematic Review and Meta-Analysis. Sensors (Basel). 2021 Dec 8;21(24):8186. https://doi.org/10.3390/s21248186 56 Porciuncula F, Roto AV, Kumar D, Davis I, Roy S, Walsh CJ, Awad LN. Wearable Movement Sensors for Rehabilitation: A Focused Review of Technological and Clinical Advances. PM R. 2018 Sep;10(9 Suppl 2):S220-S232. doi: 10.1016/j.pmrj.2018.06.013. Erratum in: PM R. 2018 Dec;10(12):1437. https://doi.org/10.1016/j.pmrj.2018.12.002 57 Cuesta-Vargas AI, Galán-Mercant A, Williams JM. The use of inertial sensors system for human motion analysis. Phys Ther Rev. 2010 Dec;15(6):462-473. https://doi.org/10.1179/1743288X11Y.0000000006 58 Li R, Peterson N, Walter HJ, Rath R, Curry C, Stoffregen TA. Real-time visual feedback about postural activity increases postural instability and visually induced motion sickness. Gait Posture. 2018 Sep;65:251-255. https://doi.org/10.1016/j.gaitpost.2018.08.005 59 Da Gama A, Fallavollita P, Teichrieb V, Navab N. Motor Rehabilitation Using Kinect: A Systematic Review. Games Health J. 2015 Apr;4(2):123-35. https://doi.org/10.1089/g4h.2014.0047 60 Liao Y, Vakanski A, Xian M, Paul D, Baker R. A review of computational approaches for evaluation of rehabilitation exercises. Comput Biol Med. 2020 Apr;119:103687. https://doi.org/10.1016/j.compbiomed.2020.103687 61 Mackay EJ, Robey NJ, Suprak DN, Buddhadev HH, San Juan JG. The effect of EMG biofeedback training on muscle activation in an impingement population. J Electromyogr Kinesiol. 2023 Jun;70:102772. https://doi.org/10.1016/j.jelekin.2023.102772 62 Patel S, Park H, Bonato P, Chan L, Rodgers M. A review of wearable sensors and systems with application in rehabilitation. J Neuroeng Rehabil. 2012 Apr 20;9:21. https://doi.org/10.1186/1743-0003-9-21 63 Gravina R, Alinia P, Ghasemzadeh H, & Fortino G. Multi-sensor fusion in body sensor networks: State-of-the-art and research challenges. Information Fusion, 2017; 35, 1339-1351. https://doi.org/10.1016/j.inffus.2016.09.005 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 66 specific joints on the skin which can be fused for monitoring and visualizing them in order to support by physicians or patients to assess the quality of movement and eventually the effectiveness of exercise. Lastly, Adans-Dester et al. 64 2020  concentrated on importing data collected from wearable sensors to visualization tools to give instant feedback of movement patterns, muscle activation and joint angles for the subjects, while they are performing rehabilitation exercises so that they could be well-appraised by regarding their postures and movements as incorrect. This expansive work elucidates the importance of sensor basedvisualization in contemporaneous rehabilitation and has fuelled a new generation of patient-centred, livefeedback point-of-care approaches to optimize outcomes. Sensor-based technologies in the last few years, revolutionized rehabilitation as advanced tools for monitoring analysing and visualizing patient movement in real-time were introduced. These inventions have become key support in the road of recovery itself, helping to gain importance for the therapy performance and deliver some clarifications needed for both physiotherapists as well as patients. Wearable sensors, like IMUs, EMG and accelerometers are becoming extremely popular to track body movements, muscle activity and biomechanics. Meanwhile, optical systems such as the Microsoft Kinect and depth cameras have made movement analysis a reality in non-invasive settings which is important for detailed kinematic analysis without need for wearables. Multisensor systems have also made it even more accurate and reliable in the capturing of different motor patterns thanks to this fusion between different sensors, which allows for adding and improving the quality of movement tracking introduced. There are some real-time visualization tools that use raw sensor data streams to provide intuitive, interactive feedback for patients and therapists so they can keep tabs on progress and modulate therapy as needed. Beyond increasing the effectiveness of rehabilitation, these tools also increase patient motivation by providing patients with immediate performance feedback and allow opportunities for customized therapy. These sensor-based visualization tools are transforming rehabilitation from postural correction to gait analysis and neuromuscular recovery in an innovative way for dynamic, datadriven solutions that improve patient outcomes. Here we can see the most relevant platforms that are currently used in the healthcare industry. Figure 17. Indicative images of the Motek Medical – CAREN65, with a user interacting with the platform. − Motek Medical - CAREN (Computer Assisted Rehabilitation Environment) 65 : The CAREN (Cardiovascular Aquatic Rehabilitation & Exercise Networks) is a next-generation rehabilitation platform, which combines 64 Adans-Dester C, Hankov N, O'Brien A, Vergara-Diaz G, Black-Schaffer R, Zafonte R, Dy J, Lee SI, Bonato P. Enabling precision rehabilitation interventions using wearable sensors and machine learning to track motor recovery. NPJ Digit Med. 2020 Sep 21;3:121. https://doi.org/10.1038/s41746-020-00328-w 65 The World’s Most Advanced Biomechanics Lab. Study all aspects of balance and locomotion. [Online]. Available: https://www.motekmedical.com/solution/caren/ D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 73 a reduction of venous capacity 91 . According to Varaki et al.92, plethysmographic methods allow the extraction of hemodynamic parameters 92 of the limbs, such as: venous filling index, ejection fraction, residual volume fraction, and multiple venous blood outflow parameters. Figure 23. Typical VOP signal tracings by three different plethysmography techniques 93 . 3.2.1. Strain-gauge plethysmography (SGP) The SPG method is based on mercury-or gallium-filled flexible tubes surrounding the limb acting as a straingauge91,92. Variations in strain-gauge length caused by changes in the limb volume result in changes in the electrical resistance. When combined, various parameters derived from the SGP measurements have an overall sensitivity of 96% to determine venous pathology, without the ability to distinguish between specific disorders, thus leaving it with a diagnostic uncertainty92. Moreover, the SGP method is sensitive to temperature and has a chemical hazard, the latter can be reduced when using indium gallium strain gauges92. The ideal gauge length is calculated by taking 90% of the limb’s circumference, with a 10% stretch when applied91, this can pose limitations for routine screening due to individual differences. Another variation of strain-gauge plethysmography - ambulatory strain gauge plethysmography (ASGP) is used to measure calf volume changes in the upright position 94 . Strain gauges are applied to both ankles above the malleolus to avoid artifacts related to calf muscle contraction. According to A. N. Nicolaides the ASGP testing is done in the following way: the usual test exercise includes 20 knee bends at a specific rate of 30 per minute, after which the patient is asked to stand completely still until the full volume return is reached. This return usually lasts between 1-2 minutes for a normal person. This allows the calculation of venous refilling time (RT) and expelled volume (EV). The test is followed by the application of a below-knee compression cuff that is 2.5 cm wide and inflated to a pressure of 70 mmHg. This normalizes the venous return time in patients with isolated superficial venous reflux. Reference values for normal controls are RT of 42 to 96 seconds and EV of 0.7 to 3.1 mL/100 mL. In summary, ASGP measures the performance of the venous muscle pump by assessing venous reflux and expelled volume. While it can help distinguish between superficial and deep venous insufficiency, it does not provide detailed information about the location or 91 Neumann H.A.M., & Maessen-Visch, M. B. (1999). Plethysmography. Current problems in dermatology, 27, 114– 123. https://doi.org/10.1159/000060635 92 Varaki et al. (2018). Peripheral vascular disease assessment in the lower limb: a review of current and emerging non-invasive diagnostic methods. Biomedical engineering online, 17(1), 61. https://doi.org/10.1186/s12938-0180494-4 93 Anderson F.A. Impedance plethysmography in the diagnosis of arterial and venous disease. Ann Biomed Eng 12, 79–102 (1984). https://doi.org/10.1007/BF02410293 94 Nicolaides A. N. Investigation of Chronic Venous Insufficiency. Circulation, vol. 102, no. 20, pp. e126–e163, 2000. https://doi.org/10.1161/01.cir.102.20.e126 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 74 extent of reflux or obstruction within the venous system. Duplex scanning with colour flow imaging is necessary for accurate localization and assessment94. 3.2.2. Air plethysmography (APG) The APG relies on an air-filled chamber that encloses the lower limb, and air displacement is used to measure blood volume changes92. Venous hypertension arises from impaired venous return, typically caused by one or a combination of factors including venous reflux, obstruction, and inadequate calf muscle pump function. Air plethysmography can measure each of these three components, enhancing the understanding of venous pathophysiology94. APG is performed with a 5-liter air chamber encompassing the leg from the ankle to the knee, which is then connected to the pressure sensor reading changes. The pressure change is calibrated with 100 ml of air 95 . According to A. N. Nicolaides the air-plethysmograph consists of a 35cm-long tubular air chamber made from polyurethane, which surrounds the whole leg and is inflated to a 6mmHg pressure. For venous outflow (obstruction) testing not only an air chamber but also a tourniquet (10-12cm wide) is placed on the thigh as proximally as possible and is inflated to 80 mmHg (Figure 24)94. Rapid deflation of the tourniquet allows the recording of a venous outflow signal. According to A. N. Nicolaides, the outflow fraction at 1 second represents the venous outflow as a percentage of the total venous volume. This measurement is repeated after occluding the long saphenous vein at the knee. Clinically, an outflow fraction at 1 second greater than 38% suggests no significant functional obstruction, a range of 30% to 38% indicates moderate obstruction and values below 28% suggest severe obstruction94. In another study by N. Labropoulos et al., APG was found to be 95% sensitive, 95% specific, and 95% accurate in detecting proximal (thigh) DVT. However, the method was found to be 45% sensitive, 95% specific, and 81% accurate for distal (calf) DVT 96 . DVT was confirmed by venography or other gold standard procedures. Despite relatively good results these methods are not widely used and accepted. One of the key factors preventing adoption could be the cumbersomeness of the method and its application while also being operator dependent. Figure 24. Typical experimental setup of APG test. Outflow fraction (OF) (median and 90% range) without and with occlusion of superficial veins (s) in limbs of 50 normal volunteers (N), 157 limbs with primary varicose veins, 70 limbs with deep venous incompetence (DVI), and 68 limbs with venographic evidence of deep venous occlusion (DVO)94. 3.2.3. Electrical Impedance Plethysmography (EIP) Impedance is an electrical parameter that describes how much an electrical circuit impedes (opposes) alternating current (AC), and it is calculated according to Ohm’s law 𝑈 = 𝑉/𝐼. In the measurement of human impedance over the skin, the typical measurement setup involves a tetrapolar electrode setup, where two electrodes supply current and two other electrodes detect voltage. The four-electrode circuit will allow the effects of electrode-skin contact to be negligible due to minimal amount of current flowing into the voltage 95 Dezotti et al. The clinical importance of air plethysmography in the assessment of chronic venous disease. J Vasc Bras. 2016 Oct-Dec;15(4):287-292. https://doi.org/10.1590/1677-5449.002116 96 Labropoulos et al. Air Plethysmography in the Detection of Suspected Acute Deep Vein Thrombosis. Phlebology. 1995;10(1):28-31. https://doi:10.1177/026835559501000107 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 75 detection circuit. In EIP, the electrical current is both high frequency and low amplitude such that it is imperceptible to humans, and therefore, the method is completely non-invasive and safe. Different materials have different impedance values according to their capability to conduct current, and blood is a good electrical conductor due to a high amount of ions 97 . EIP measurement is typically used for the measurement of blood flow or volume through arteries. When blood flows through the arterial system and blood volumes increases, into a limb for example, the conductivity increases, and impedance gets lower. As blood leaves the measurement area, the reverse happens. The main impedance changes in a limb during a period of no activity are due to flow of blood because no other large changes in tissue happen that would affect electrical conductivity. Therefore, the impedance measurement in such a setting accurately reflects blood volume change in the limb, Δ𝑍 ∝ Δ𝑉. This relation of impedance value to blood volume is used by venous occlusion plethysmography (VOP) to measure the amount of blood the venous system can hold (venous capacity) as well as the drainage (venous outflow). Both are limited and affected by thrombi occlusion. 3.2.4. Approach and extracted parameters. The impedance measurement for VOP is based on a similar approach to the other plethysmographic methods. The legs are first raised to drain the venous system. A cuff is inflated to stop the flow of venous, but not arterial blood flow. This causes filling of the venous system in the lower legs, which decreases impedance due to increased conductivity from blood pooling. The cuff is released after set amount of time or when blood pooling saturates, and the outflow is computed together with the capacity. The function of blood outflow to stored blood (venous capacity) is plotted graphically and the diagnosis is based on a determinant line or an index value that is determined experimentally from a population with those two parameters. It is, therefore, an index-based diagnosis. The two parameters measured from the VOP impedance signal are: 1. Venous capacity (this is the inflow of blood after cuff inflation. It is the difference in impedance value after emptying the veins and the value where impedance is lowest due to inflation of the cuff), 2. Venous outflow at 2 s and 3 s (this is the outflow of blood shortly after the cuff deflation. It is the difference in impedance from the lowest impedance value before deflation to the value 2 or 3 seconds after deflation). The units for the two values are typically arbitrary impedance units. With assumptions about leg geometry and electrical model 98 , they are also sometimes given as the normalized volume change of the amount of blood to the amount of surrounding tissue (ml/100ml or %) and outflow is sometimes reported as speed (%/min). However, as correct use of these units would require measuring the limb as well as knowledge of the haematocrit of blood, the arbitrary unit is used in most literature. Some systems may have population indices they use to convert the impedance units to volume units by assuming the anatomical dimensions of the measurand. Although the explained two values are the basis of DVT diagnosis in majority of the EIP VOP literature, some other parameters are also considered. These include flow resistance, outflow time constant, venous tone, arterial inflow, base impedance and others that are derived from the signal. More hemodynamical parameters are available if also the arterial pulse signal is analysed. 97 Miklavčič et al. Electric Properties of Tissues. in Wiley Encyclopedia of Biomedical Engineering, John Wiley & Sons, Ltd, 2006. https://doi.org/10.1002/9780471740360.ebs0403 98 Jindal et al. Corrected formula for estimating peripheral blood flow by impedance plethysmography. Med. Biol. Eng. Comput., vol. 32, no. 6, pp. 625–631, Nov. 1994, https://doi.org/10.1007/BF02524237 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 76 The changes in venous capacity and venous outflow in DVT are due to large acute thrombi in popliteal, femoral or iliac veins that obstruct normal pressure/volume relationship in the venous system 99 . This causes especially the outflow to be much slower but also for the capacity to be dampened. Bilateral testing is especially revealing but thrombi can also occlude both legs at the same time though probability of this to happen is low. The main advantage of the impedance method is the low cost 100 and non-invasiveness as well as its simplicity. It was widely reported in the literature beginning in the 1970s 101 to the 1990s 102 but generally replaced by other methods such as compression ultrasound. 3.2.5. EIP implementation details The implementation details vary in literature though the methodology is commonly the same. The end of the bed is elevated between 25-45° to facilitate the venous emptying by elevating the legs above the heart level. Knees are flexed 10-35° to prevent posterior border of the tibia from compressing the popliteal vein in the knee. The hip is often externally rotated. The cuff is inflated to a pressure that is between 35 to 80 mmHg. It is inflated from 45 seconds to 3 minutes. In some implementations, it is inflated and deflated in cycles up to five times 103 . Repeating the test increases the venous capacity in subsequent tests. In healthy people the outflow will equally increase, while for those occluded with thrombi the outflow does not increase. In the commercial Medis equipment it is first inflated for 40 mmHg for one minute, followed by one minute of 60 mmHg and finally one minute of 80 mmHg for a total of three minutes before deflation. The cuff is between 15-21 cm in width in the reported values. Venous capacity is measured as the difference in impedance from the baseline before cuff inflation to the point with lowest impedance during occlusion seen in Figure 23, where the impedance value is inverted. The outflow is typically measured from the peak value to the 3-s point after the deflation of the cuff although other values such as 2-s are also used. The derivative of the downslope is also reported in some research and additionally outflow time constant is calculated in others. The diagnosis is made based on the discriminant line presented in Figure 25. Sometimes, a grey zone is used to indicate results that are ambiguous. The electrodes used in EIP VOP literature are generally band-type metal electrodes used together with paste to improve conductivity. The electrodes are not always reported and were considered secondary to other methodological concerns. Typically, tetrapolar measurement is used where four electrodes are applied to the calf. The current electrodes are below the knee and near the ankle, while the voltage electrodes are between these with fixed length between them. The most important parameters regarding the device are the frequency of the current, amplitude of the current and electrodes used as well as electrical details of the circuit. The current is a sinusoidal signal. Higher amplitude has a better signal-to-noise ratio. All studies that reported their amplitude had a safe amplitude as defined by the IEC-60601-1 standard. The frequency has had a high variance with a lower limit of around 99 Wheeler et al. Diagnostic Methods for Deep Vein Thrombosis. Pathophysiology of Haemostasis and Thrombosis, vol. 25, no. 1–2, pp. 6–26, Apr. 1995, https://doi.org/10.1159/000217140 100 Goodacre et al. Measurement of the clinical and cost-effectiveness of non-invasive diagnostic testing strategies for deep vein thrombosis. Health Technology Assessment, vol. 10, no. 15, May 2006, https://doi.org/10.3310/hta10150 101 Wheeler et al. Occlusive impedance phlebography: A diagnostic procedure for venous thrombosis and pulmonary embolism. Progress in Cardiovascular Diseases, vol. 17, no. 3, pp. 199–205, Nov. 1974, https://doi.org/10.1016/0033-0620(74)90044-9 102 Baker W. F. Diagnosis of deep venous thrombosis and pulmonary embolism. Medical Clinics of North America, vol. 82, no. 3, pp. 459–476, May 1998, https://doi.org/10.1016/S0025-7125(05)70005-5 103 Hull et al. Impedance plethysmography: The relationship between venous filling and sensitivity and specificity for proximal vein thrombosis. Circulation, vol. 58, no. 5, pp. 898–902, Nov. 1978, https://doi.org/10.1161/01.CIR.58.5.898 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 77 10 kHz and upper limit of around 100 kHz. The lower limit is likely due to the effects of skin contact impedance at lower frequencies and the higher is due to implementation challenges of the circuit. The circuit injects voltage or current and measures the drop in voltage between the two voltage electrodes and transfer this to impedance value based on the injected current. Figure 25. Discriminant line examples used for diagnosis of DVT with plethysmographic methods. It is calculated as the best fit when outflow is the function of venous capacity. Above the line are healthy values, while values below the line are those likely to have DVT occlusion. The version on the left includes a “grey zone” with unsure results. The version on the right includes patient data with recent proximal DVT to determine the discriminant line. Images from107. 3.2.6. Diagnostic accuracy A meta-analysis with 42 cohorts 104 reported sensitivity and specificity of 88% and 90% for proximal DVT. For distal DVT, the sensitivity was 28%. The report also noted a very high heterogeneity in the results. In another study 105 the sensitivity and specificity of proximal DVT were 77% and 93%. With the exclusion of certain small thrombi, the sensitivity increased to 91%. Applying a clinical model with clinical history to subgroup patients before the EIP result assessment also increased the sensitivity on the same cohort105. Generally, EIP is considered unreliable in the detection of calf thrombi and also proximal DVT in the case of asymptomatic patients. Other sources share higher values for occlusive proximal DVT in the range of 94% for both sensitivity and specificity 106 . The higher values were in the initial literature. The subsequent reporting of lower values was investigated, and it was reported to be possibly due to changes in patient referrals as well as differences in methodology 107 . In the case of symptomatic proximal DVT, the sensitivity and accuracy are considered good, but slightly worse than compression ultrasound. 104 Locker et al. Meta-analysis of plethysmography and rheography in the diagnosis of deep vein thrombosis. Emerg Med J, vol. 23, no. 8, pp. 630–635, Aug. 2006, https://doi.org/10.1136/emj.2005.033381 105 Wells et al. A simple clinical model for the diagnosis of deep-vein thrombosis combined with impedance plethysmography: Potential for an improvement in the diagnostic process. Journal of Internal Medicine, vol. 243, no. 1, pp. 15–23, 1998, https://doi.org/10.1046/j.1365-2796.1998.00249.x 106 H. B. Wheeler & F. A. Anderson Impedance Plethysmography. in Practical Noninvasive Vascular Diagnosis, 2nd ed. Year Book Medical Publishers, 1987, pp. 407–437. [Online]. Available: https://archive.org/details/practicalnoninva0000unse_w8y2/ 107 M. A. Mansour and D. S. Sumner Overview: Plethysmographic Techniques in the Diagnosis of Venous Disease,” in Noninvasive Vascular Diagnosis: A Practical Guide to Therapy, A. F. AbuRahma and J. J. Bergan, Eds., London: Springer, 2007, pp. 375–384. https://doi.org/10.1007/978-1-84628-450-2_34 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 78 3.2.7. Challenges, errors, and limitations Firstly, the Society for Phlebology has stated in its guidelines on diagnostics that air-cuff plethysmography is not suitable for thrombosis diagnostics 108 . According to two decades of data, the EIP remained largely a research test while other methods for DVT detection took over 109 . EIP equipment was generally unavailable in hospitals besides those that used it for research109, 110 . The hospitals that used it mostly used the same devices, and those that used new equipment reported worse results. Those reports were also criticized for their selection bias and differences in methodology 111 . In 2024, the availability of commercial EIP equipment for DVT detection is very limited. There is a lack of consistency in reporting on the equipment used. Details of electrodes used as well as specifics of the methodology such as the position of the knee and hip and cuff inflation details are not always reported and vary between reports. This causes some rate of error on the true accuracy of the method. In addition, it is also clear that operator experience had some impact as the two main researcher Wheeler101 and Hull 112 consistently reported the best results in their clinics. The application of the electrodes and cuff may cause errors. Some patient groups such as pregnant women or those having respiratory difficulties may also need some special considerations with the method106 causing inexperience of the operator to also lower the accuracy of the method. Wheeler considered the reported results with worse sensitivities to have significant methodological deficiencies, use of nonstandard instrumentation, small sample sizes and problems in retrospective study design99. The methodology is simple but differences in the parameters such as the leg elevation cause errors in diagnostic accuracy. The method is limited to certain types of DVT as well as certain types of patients. Asymptomatic patients generally have a much lower sensitivity due to lack of occlusion with a small thrombus. The outflow is also not always reduced due to, e.g., collateralization of the veins and these cause false negatives. False positives are caused by any other disorder that may cause outflow impairment such as tumor or other external compression. Certain patient groups may also reduce accuracy such as those with obesity, arterial insufficiency, Baker’s cyst, calf or thigh hematoma or abscess, inguinal adenopathy, peripheral vasoconstriction and hypovolemia among others102 and should be considered when using the method. 3.2.8. Future considerations The EIP method for DVT detection is especially useful for screening certain patient populations due to its noninvasiveness and ease of use. For wearable applications, the electrodes must be designed to be comfortable and non-intrusive to everyday life. Some modern studies exist that employ a cuff-less method by raising the 108 Dörr D. Die Luftplethysmographie : und ihre Bedeutung für die Diagnostik der chronisch venösen Insuffizienz. Air plethysmography. Doctoral Thesis, 2000. https://publikationen.ub.unifrankfurt.de/frontdoor/index/index/year/2005/docId/3928 109 Stein et al. Tracking the Uptake of Evidence: Two Decades of Hospital Practice Trends for Diagnosing Deep Vein Thrombosis and Pulmonary Embolism. Archives of Internal Medicine, vol. 163, no. 10, pp. 1213–1219, May 2003, https://doi.org/10.1001/archinte.163.10.1213 110 Wilbur J. & Shian B. Diagnosis of Deep Venous Thrombosis and Pulmonary Embolism. afp, vol. 86, no. 10, pp. 913– 919, Nov. 2012, Accessed: May 31, 2024. [Online]. Available: https://www.aafp.org/pubs/afp/issues/2012/1115/p913.html 111 Akers et al. Impedance plethysmography. It’s the clinical outcome that counts. Chest, vol. 106, no. 5, pp. 1317– 1318, Nov. 1994, https://doi.org/10.1378/chest.106.5.1317 112 Hull et al. Impedance plethysmography using the occlusive cuff technique in the diagnosis of venous thrombosis. Circulation, vol. 53, no. 4, pp. 696–700, Apr. 1976, https://doi.org/10.1161/01.CIR.53.4.696 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 79 leg instead 113 , 114 . A method before the occlusion method employed high respiratory effort using the Valsalva maneuver 115 . Both methods are untested on larger patient populations and could have possible use. Especially with modern microprocessing tools, the respiratory efforts from the impedance signal are easier to analyze, and even passive respiratory effects may be considered with signal processing tools. Similarly, the simple discriminant line was used mostly due to ease of calculation. Machine learning may be employed for better use of parameters and patient information 116 . The signal waveform morphology is also not used in the old literature, and it may be used for further information either together with the occlusion method or without it for information about the venous and arterial system 117 , 118 . Harmonic analysis, spectral analysis, and unilateral changes to wave morphology are largely unexplored possibilities. Continuous monitoring of the base impedance may also reveal especially the onset of edema relevant to DVT. The already clinically validated method can therefore be complemented with modern approaches to signal processing. 3.2.9. Commercially available VOP systems The modular system combining VasoScreen4000 and VasoScreen5000 (medis Medizinische Messtechnik GmbH, Germany) is designed for arterial and venous vascular diagnosis ( Figure 26a). Together with “Cardiovascular Lab” software it provides 7 main parameters (based on EIP) 119 : Base impedance (Ω), Arterial inflow (%/min), Venous capacity (%), Venous outflow at 2s rate (%/min), Outflow volume at 3s (%), Outflow volume at 5s (%), Outflow time constant (s). Additionally, it provides LRR and arterial occlusion plethysmography measurement capabilities. The AngE-System (SOT Medical Systems, Austria) is conceived as a modular solution for vascular and cardiovascular diagnostics, too ( Figure 26b). The AngE-Complete is a vascular diagnostics lab capable of performing VOP tests 120 . The sensing is performed using an air cuff, and a total of 7 parameters are provided (based on APG): Venous capacity (ml/100ml), Venous outflow (ml/100ml/min), OV1 (ml/100ml) (OF 1%), OV2 (ml/100ml) (OF 2%), OV3 (ml/100ml) (OF 3%), OV4 (ml/100ml) (OF 4%), Arterial flow (ml/100ml/min). 113 Pittella et al. Combined impedance plethysmography and spectroscopy for the diagnosis of diseases of peripheral vascular system. in 2017 IEEE International Symposium on Medical Measurements and Applications (MeMeA), May 2017, pp. 367–372. https://doi.org/10.1109/MeMeA.2017.7985904 114 Uday et al. Design and Development of an EIP System Without an Occlusive Cuff to Detect Deep Vein Thrombosis. presented at the Second International Conference on Emerging Trends in Engineering (ICETE 2023), Atlantis Press, Nov. 2023, pp. 26–35, https://doi.org/10.2991/978-94-6463-252-1_5 115 Wheeler et al. Impedance Phlebography: Technique, Interpretation, and Results. Archives of Surgery, vol. 104, no. 2, pp. 164–169, Feb. 1972, https://doi.org/10.1001/archsurg.1972.04180020044008 116 Tumey D. M. & Randolph L. T. Apparatus and method for detecting deep vein thrombosis. 5991654A, Nov. 23, 1999 Accessed: Jun. 11, 2024. [Online]. Available: https://patents.google.com/patent/US5991654A/en 117 De Macedo et al. A novel model to simulate venous occlusion plethysmography data and to estimate arterial and venous parameters. Res. Biomed. Eng., vol. 36, no. 4, pp. 463–473, Dec. 2020, https://doi.org/10.1007/s42600020-00087-3 118 Bejarano Monroy M. G. A novel approach to bioelectrical impedance plethysmography for the assessment of arterial and venous circulatory problems in the forearm. doctoral, City, University of London, 2019. Accessed: Jan. 25, 2023. [Online]. Available: https://openaccess.city.ac.uk/id/eprint/24064/ 119 VasoScreen 5000 - 4000. [Online]. Available: https://www.medis.company/en/products/vasoscreen-5000-4000 120 AngETM COMPLETE. [Online]. Available: https://www.sot-medical.com/solution/ange-complete/ D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 80 Figure 26. Commercially available VOP systems. (a) VasoScreen4000-5000 (medis GmbH, Germany), and (b) AngE-System (SOT Medical Systems, Austria). 3.2.10. Summary of knowledge gaps and potential solutions with implications for requirements − Although plethysmographic methods have existed for decades, there is a lack of studies with large sample sizes that provide results on different types of thrombosis and use cases. − There is a lack of consistency in reporting on the equipment and methodology used. Details of electrodes used and specifics of the methodology such as the patient’s leg position and cuff inflation details are not always reported and vary between reports. This causes differences between studies and the true diagnostic accuracy is hard to determine. − Among the plethysmographic methods presented, EIP appears to be the most user-friendly and comfortable due to its non-invasiveness, simplicity, and ease of use. The APG method requires an air bladder, and SGP relies on mercuryor gallium-filled flexible tubes, making these methods cumbersome and less reliable compared to the EIP method. It should be noted that for continuous or frequent EIP testing, textile electrodes should be used instead of typical Ag/AgCl gel-based electrodes to ensure comfort and minimize disruption to everyday life. 3.3. Light reflection rheography (LRR) and photoplethysmography (PPG) The search engine PubMed found 31 sources for the following query (DBQ-03-2024): (DVT OR “deep vein thrombosis”) AND (LRR OR “light reflection rheography” OR photoplethysmography). Of these, 29 sources appeared relevant to the project. 3.3.1. Background Photoplethysmography (PPG) is a non-invasive optical technique used to measure heart rate, blood oxygen levels, and other cardiovascular parameters by detecting blood volume changes in tissue. It works on the principle that blood absorbs light. Α light source (LED) emits light into the skin, and a photodetector (PD) measures the reflected or transmitted light. As the heart pumps blood, changes in capillary blood volume affect light absorption, generating a PPG signal. This signal has two main components: the alternating component (AC), which reflects pulsatile blood flow corresponding to each heartbeat, and the DC component, which represents the baseline blood volume (venous blood) and remains relatively stable over time (Figure 27). a) b) D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 81 Figure 27. PPG signal components 121 . Light Reflection Rheography (LRR), sometimes referred to as Digital Photoplethysmography (D-PPG, digital vs. analog version of LRR), is a specialized mode of PPG that captures low-frequency (<0.5 Hz) variations in blood volume. It was hypothesized 122 , 123 that this mode is suitable for detecting DVT as it primarily detects changes in venous blood volume using LRR sensors positioned on both lower legs. Blood flow in the peripheral venous system can be recorded quickly and easily. 3.3.2. Approach LRR measurements are taken from a selected skin area by emitting non-visible infrared (IR, ~940 nm) light into the skin. Depending on the blood volume in that area, the light is absorbed, and the proportion of reflected light corresponds inversely to the blood volume. The light intensity of the sensors can automatically adapt to the conditions at the measurement location, making LRR measurements possible regardless of skin colour and even through targeted textiles. The measurement should be performed in a well-tempered room (20-24°C) and after a rest period of at least several minutes before the start of the examination is recommended. LRR examination for DVT uses venous blood flow changes and observes changes in LRR signals. Several methods are used to induce venous blood flow changes: 1) passive execution of postural movements of the limbs in supine position, 2) foot pumping in sitting position, i.e.., the patient performs the dorsiflexion at the ankle to push the blood in the vein of the calf back to the heart, 3) VOP technique used in venous occlusion plethysmography similar to EIP case. 3.3.3. Implementation details and extracted parameters The LRR probe is usually placed on the calf 10 cm from the internal malleolus 124 over the long saphenous vein near the ankle. LRR signals represent DC trend variation due to three protocol phases of the examination: 1) resting baseline, 2) venous emptying manoeuvre, and 3) refiling phase. As mentioned above, there are three types of venous emptying maneuvers or tests. LRR measurement takes about 2 minutes to perform. 121 Photoplethysmography Textbook. [Online]. Available: https://peterhcharlton.github.io/post/ppg_book/ 122 Pearce et al. Hemodynamic Assessment of Venous Problems. Surgery vol. 93,5 (1983): 715-21. 123 Shepard et al. Light reflection rheography: A new non-invasive test of venous function. Bruit 1984, 8, 266–270. https://doi.org/10.1016/S0196-0644(05)82516-8 124 Antignani et al. Light-reflection rheography and acute deep vein thrombosis. Angiology vol. 44,7 (1993): 523-6. https://doi.org/10.1177/000331979304400703 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 82 In the “passive postural movements of the limb’s” test, the limbs are heightened individually. Figure 28 shows characteristic patterns of LRR signal. (a) (b) (c) (d) Figure 28. Illustrations of LRR in health and pathology124: (a) a normal subject, (b) iliofemoral thrombosis, (c) severe iliofemoral thrombosis, and (d) thrombosis of the vena cava and iliac veins. In the “foot pumping” test, the patient should sit comfortably on a chair with the foot resting flat on the floor, and the angle between the calf and thigh should be approximately 110 degrees. This angle prevents the restriction of venous blood flow in the popliteal vein and the saphenous-popliteal junction. Visual or audio guidance is provided to the patient when the foot needs to be raised and lowered, involving movement at the ankle joint (Figure 29). This action activates the calf pump, allowing blood to be pumped out of the vein. After 8-10 beeps, the leg is kept stationary, and the device measures how long it takes for the vein to refill. Motion sensors, such as accelerometers 125 , can help automatically track the voluntary movement of the foot. The study 126 proposed a robot for passive ankle movement to activate the calf pump (Figure 29). Figure 29. Monitoring and control of the foot in foot pumping mode: by accelerometer (left and middle125), robot-based foot movement (right126). There are clear differences among LRR signal patterns and venous system states as registered by foot pumping test (Figure 30). 125 Liu et al. An Examination System to Detect Deep Vein Thrombosis of a Lower Limb Using Light Reflection Rheography. Sensors 21, no. 7 (2 April 2021): 2446. https://doi.org/10.3390/s21072446 126 Wnuk et al. Effect of passive ankle movement in the sitting position on the symptoms of chronic venous insufficiency with long-term observation. Advances in clinical and experimental medicine: official organ Wroclaw Medical University vol. 33,2 (2024): 135-141. https://doi.org/10.17219/acem/166046 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 89 Figure 33. Color‐coded elastogram, with stiff tissue areas coloured red, intermediate tissue areas coloured green, and soft tissue areas coloured blue 159 Figure 34. Myoton digital palpation device for soft tissue assessment 160 , 161 The Evoked response of the muscle to the mechanical impulse or vibration can be measured as well by electromyography (EMG) 162 , 163 . Kang JW et al used this method to classify DVT stages using CNN of EMG with vibrotactile stimulation on a pig model (Figure 35). They achieved the total average classification accuracy before and after DVT stages of 72.0 ± 11.9%. 159 Mumoli et al. Ultrasound elastography is useful to distinguish acute and chronic deep vein thrombosis.” Journal of thrombosis and haemostasis: JTH vol. 16,12 (2018): 2482-2491. https://doi.org/10.1111/jth.14297 160 Agoriwo, Mary W et al. “Feasibility and reliability of measuring muscle stiffness in Parkinson's Disease using MyotonPRO device in a clinical setting in Ghana.” Ghana medical journal vol. 56,2 (2022): 78-85. doi:10.4314/gmj.v56i2.4 161 Myoton digital palpation device https://www.myoton.com/ 162 Ritzmann et al. EMG activity during whole body vibration: motion artifacts or stretch reflexes? European journal of applied physiology. 2010; 110(1):143–51. https://doi.org/10.1007/s00421-010-1483-x 163 Barnes W. S. & Williams J. H. Effects of ischemia on myo-electrical signal characteristics during rest and recovery from static work. Am J Phys Med. 1987; 66(5):249–63. PMID: 3434627 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 90 Figure 35. The EMG sensor and vibration motor on a pig model of DVT 164 Finaly, in Tensiomyography 165 , the evoked response to the electrical stimulation of the muscle is measured by displacement sensor (Figure 36). Figure 36. Tensomyiography principle 166 , 167 Mechanomyography (MMG) is a technique, enabling registration of the radial displacement of a muscle during a contraction 168 . Using alone or with neuromuscular electrical stimulation (NMES), MMG is used as an alternative to EMG signal for the monitoring of muscle function, in terms of fatigue, muscle force, and its derivative (torque) as well as for prosthesis control and the detection of myopathies. Sensors used for MMG 164 Kang et al. Classification of deep vein thrombosis stages using convolutional neural network of electromyogram with vibrotactile stimulation toward developing an early diagnostic tool: A preliminary study on a pig model. PLoS ONE 18(2). 2023. https://doi.org/10.1371/journal.pone.0281219 165 Tensiomyography. Quantifying muscle function. [Online]. Available: https://www.tensiomyography.net/ 166 https://www.tensiomyography.net/ 167 Wilson, Hannah V et al. “Repeated stimulation, inter-stimulus interval and inter-electrode distance alters muscle contractile properties as measured by Tensiomyography.” PloS one vol. 13,2 e0191965. 16 Feb. 2018, doi:10.1371/journal.pone.0191965 168 Uwamahoro et al. Assessment of muscle activity using electrical stimulation and mechanomyography: a systematic review. BioMed Eng OnLine 20, 1 (2021). https://doi.org/10.1186/s12938-020-00840-w D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 91 signal acquisitions are based on piezoelectrics, microphones, accelerometers or capacitive displacements sensors 169 (Figure 37). Figure 37. Surface MMG sensors and their signals compared to EMG 170 , 171 3.4.2. Summary of Knowledge Gaps and Potential Solutions with Implications for Requirements − The most common symptoms of DVT, though lacking sensitivity and specificity, are calf pain and swelling. While these symptoms are nonspecific, they are associated with changes in muscle stiffness, which may provide additional information for monitoring DVT. − Tissue stiffness can be assessed actively (with external stimulation causing muscle movement or compression) or passively. Feasible methods within the scope of the project include ultrasonography or analyzing mechanical muscle response through EMG or mechanical sensors. − Ultrasound, a commonly used technology for DVT assessment, offers two elastography methods: (a) strain elastography, which analyses tissue displacement in response to compression provided by the operator, and (b) shear wave elastography, which uses specialized transducers. Both methods require further development: operator-independent solutions for strain elastography and specialized transducers for shear wave elastography. − Recording the evoked mechanical response after stimulation, which causes muscle movement, measures muscle properties as a purely mechanical system. This is highly correlated with an individual’s body composition. These methods are related to the PPG/LRR/IMU applications (analysis of responses from these modalities to stimuli for DVT detection) and can be further explored for adding additional data to DVT monitoring. 4. Actuators for tissue compression and US transducer positioning Tissue compression actuator (TCA) will serve three purposes: (1) it will aid in the detection of thrombus via automatic, operator-independent compression ultrasonography (CUS), (2) it will be used in the venous 169 Scarborough et al. Quantifying muscle contraction with a conductive electroactive polymer sensor: introduction to a novel surface mechanomyography device. International Biomechanics, 10(1), 37–46. 2023. https://doi.org/10.1080/23335432.2024.2319068 170 Scarborough, D. M. et al. ‘Quantifying muscle contraction with a conductive electroactive polymer sensor: introduction to a novel surface mechanomyography device’, International Biomechanics, 10(1), pp. 37–46, (2023). doi: 10.1080/23335432.2024.2319068. 171 https://www.figur8tech.com D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 92 occlusion plethysmography (VOP) method, as described in Chapters 3.2 and 3.3., and (3) it will be capable of automatically steering the US transducer along the vein during prolonged clinical monitoring. 4.1. Background In 1982, technologist Steve Talbot discovered that it was possible to distinguish between normal veins and those containing clots by using a combination of pulsed Doppler and real-time B-mode ultrasonography 172 . He found that an obstructed vein would not change in diameter during respiration (Valsalva maneuver) or when light pressure was applied to the skin. Today, CUS is the primary imaging technique for diagnosing DVT in routine clinical practice. The standard CUS procedure utilizes a 5to 10-MHz linear probe, and the examination is performed using compression, interpreted according to established methods 173 . Venous duplex ultrasound (VDUS), incorporating B-mode compression maneuvers and Doppler evaluation, has been shown to provide additional diagnostic value 174 , 175 , 176 . Colour-flow Doppler helps detect residual flow within a thrombosed venous segment (i.e., non-occlusive DVT) and in confirming the patency of venous segments that cannot be accessed by compression maneuvers (e.g., the iliocaval veins). Along with assessing respirophasicity, distal augmentation maneuvers, such as calf compression, are performed during spectral Doppler evaluation to further confirm venous patency. A sharp "spike" of augmented anterograde venous flow should be observed during the distal augmentation maneuver. Blunted or absent flow augmentation suggests venous obstruction distal to the site being assessed174. Another method requiring venous compression (the VOP method) is already analysed in Section 3.2. 4.2. Approaches No operator-independent compression ultrasonography TCAs have been described in the literature yet. Most of the literature focuses on compression cuffs used for therapeutic purposes to facilitate venous blood return by utilizing arrays of bladders that transmit pressure to tissues. For effective treatment, pressures between 15 and 50 mmHg are required. However, a study by Hedigalla et al.177 (Figure 38) concludes that compression sleeves designed with miniaturized air bladders should be supplied with over 250% of the targeted pressure due to propagated pressure loss through the skin, fat, and muscle layers. Golgouneh et al.178 reviewed soft actuators (pneumatic-based, electrothermal, electrical) and sensors (pneumatic-based, piezoresistive, capacitive, piezoelectric) that could potentially be used in on-body compression applications (Figure 38). The challenges in actuation include dynamics modelling, material properties, interand intra-personal variabilities, biocompatibility and safety concerns, and a lack of methods for in-vivo validation. Sensing challenges, such as response drift, hysteresis, nonlinearity, sensitivity to curvature, sensor multiplexing, and manufacturing, are also discussed. The authors concluded that that future developments in soft systems may benefit from more precise pressure sensing, better predictability of applied pressure levels, and smarter closed-loop control strategies. The only limitation of pneumatic-based components is discomfort caused by ventilation issues with air-tight bladders. 172 Martens et al. The History of Diagnosing Venous Thromboembolism. Semin Thromb Hemost. 2024;50(5):739-750. https://doi.org/10.1055/s-0044-1779484 173 Lensing et al. Detection of deep-vein thrombosis by real-time B-mode ultrasonography. N Engl J Med 1989;320(06):342–345. https://doi.org/10.1056/NEJM198902093200602 174 Gornik H. L. & Sharma A. M. Duplex ultrasound in the diagnosis of lower-extremity deep venous thrombosis. Circulation. 2014;129(8):917-921. https://doi.org/10.1161/CIRCULATIONAHA.113.002966 175 Adam et al. Duplex ultrasound for evaluation of deep venous blood flow in fractured lower extremities. Polish journal of radiology, 83, e47–e53. 2018. https://doi.org/10.5114/pjr.2018.73291 176 Necas M. (2010). Duplex ultrasound in the assessment of lower extremity venous insufficiency. Australasian journal of ultrasound in medicine, 13(4), 37–45. https://doi.org/10.1002/j.2205-0140.2010.tb00178.x D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 93 Figure 38. Compression therapy device with array of mini-bladders 177 (left), array of cuffs for controllable compression 178 on tissues (right). In VOP method, the objective is to occlude the vein lumen. The bladders are implemented to act as veinoccluding actuators179 (Figure 39). Bioimpedance-based VOP measuring systems provide leg-overall (integral arterial and venous) features of blood flow. Here, the air cuff on the lower thigh is stepwise inflated to 4060-80 mmHg to softly close venous outflow from the lower leg. The cuff was also used in a more sophisticated system for basic research180 (Figure 39) on peripheral blood flow, where only foot-skin blood flow is measured with a laser while the cuff is applied to the calf. The study aimed to gain insights into the origins of blood flow surges in the skin during intermittent pneumatic compression (IPC) therapy. Figure 39. Impedance venous occlusion plethysmography 179 (left) and a skin blood flow measuring laser system with IPC 180 (right). Here, the cuffs are used to exert occlusion on all venous vessels in the thigh or calf. Several requirements can be deduced from the literature mentioned above. The TCA must be capable of applying a precise, adjustable pressure range (e.g., 0-200 mmHg) to ensure proper vein compression without causing patient discomfort. Compression should be applied uniformly across the targeted tissue area to avoid uneven imaging results, and the actuator should allow for quick release and re-application of pressure to facilitate various diagnostic maneuvers. In addition, a feedback mechanism should be incorporated to monitor tissue deformation in real time, ensuring consistent and accurate compression, such as a predefined reduction in vein lumen by 50% or 75%. Fail-safe mechanisms are essential to prevent excessive pressure that could damage tissue, including a dedicated safety valve designed to handle pressures of 300-350 mmHg. 177 Hedigalla et al. Numerical Study to Investigate the Pressure Propagation Patterns by a Compression Sleeve with Miniaturised Air-Bladders. 2022 Moratuwa Engineering Research Conference (MERCon), Moratuwa, Sri Lanka, 2022, pp. 1-5, https://doi.org/10.1109/MERCon55799.2022.9906258 178 Golgouneh A. & Dunne L. E. A Review in On-Body Compression Using Soft Actuators and Sensors: Applications, Mechanisms, and Challenges. in IEEE Reviews in Biomedical Engineering, vol. 17, pp. 166-179, 2024, https://doi.org/10.1109/RBME.2022.3220505 179 Heinrich et al. Bridging vascular physiology to vascular medicine: an integrative laboratory class. Advances in physiology education, 47(1), 97–116, 2023. https://doi.org/10.1152/advan.00170.2022 180 Wang et al. Multiple blood flow surges during intermittent pneumatic compression: The origins and their implications, Journal of Biomechanics, Volume 143, 2022, 111264, https://doi.org/10.1016/j.jbiomech.2022.111264 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 94 Furthermore, emergency release features must be in place to quickly relieve pressure in case of patient discomfort or equipment failure. The TCA must be integrated with the ultrasound imaging system to synchronize compression with image capture, ensuring accurate data input for AI-based diagnostic support. Lastly, the system should support automation, enabling the TCA to adjust pressure dynamically based on imaging feedback. Additionally, the operation of the actuator should not cause noise discomfort for the patient during prolonged monitoring. Steering the ultrasound transducer is another task of the same TCA. It must be designed to adjust the roll angle of the ultrasound transducer to optimize the imaging plane. This adjustment is critical for locating veins and improving diagnostic accuracy for DVT. In addition, the TCA plays a crucial role in Doppler ultrasound imaging by enabling precise assessments of blood flow dynamics, making it an essential tool for the accurate and efficient diagnosis of DVT. Hailu181 described modelling and experiments with fabricated prototypes of micro electrostatic actuators (Figure 40). Theoretical modelling showed that the midpoint of the moving plate could be raised by 22 micrometers with 145V. The actuator prototype was fabricated as a micromechanical system with a multi-layered structure. The actuator measured 5 mm in length and 3 mm in width. With a driving voltage of 150V, the hybrid actuator achieved a measured displacement of 6.48 µm in the vertical direction. However, this small displacement is insufficient in the practical context of imaging human tissues. Figure 40. A micro electrostatic hybrid actuator 181 . Hao182 focused on a simple soft gripper, a four-fingered pneumatic elastomeric robot mimicking a biological finger with infinite degrees of freedom. The robot operates by deflating the soft actuators to open the gripper and approach objects, then inflating to grip them. The soft fingers were made from silicone elastomer using a multi-step molding process with 3D-printed molds. Tests were conducted on finger lengths ranging from 30 mm to 100 mm. Under air pressures of -40 kPa, 0 kPa, and +40 kPa, the fingertip moved from -70 mm to +30 mm, and a maximum pull force of 13.5N was observed (Figure 41). The gripper safely handled soft objects like cacti and milk bags. 181 Hailu et al. Hybrid micro electrostatic actuator. Microsyst Technol 22, 319–327. 2016. https://doi.org/10.1007/s00542-015-2424-8 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 95 Figure 41. Kinematics of single soft finger for soft gripper in pneumatic robot 182 . The hand-held venipuncture robotic device developed by Leipheimer et al.183 combines ultrasonic imaging with miniaturized robotics for vein cannulation in the forearm. Currently, the operator uses ultrasonic guidance to manually align the needle trajectory with the vein, while the robotic system handles needle interaction with tissues and measures the required force (up to 2 N). The device has demonstrated results comparable to clinical standards, achieving an 87% success rate among 31 participants, with an average procedure time of 93 ± 30 seconds. In trials, manual needle placement accuracy was measured at 0.23 ± 0.17 mm (n = 25). However, the current design does not ensure the needle remains centred with the ultrasoundimaged vessel during user or patient movement. Future developments will include a third degree of freedom (DOF) to automatically align the needle trajectory with the underlying vessel as detected by US imaging. Figure 42. A robotic device for intravenous interventions 183 . 182 Hao et al. Universal soft pneumatic robotic gripper with variable effective length. 2016 35th Chinese Control Conference (CCC), Chengdu, China, 2016, pp. 6109-6114, https://doi.org/10.1109/ChiCC.2016.7554316 183 Leipheimer et al. First-in-human evaluation of a hand-held automated venipuncture device for rapid venous blood draws. TECHNOLOGY. 7. 1-10. 2020. https://doi.org/10.1142/S2339547819500067 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 96 Schimmoeller’s184 force sensing device measures the interaction between an imaging transducer and soft tissue while orienting the transducer during freehand imaging (Figure 43). The system includes a 6-axis load cell, an inertial measurement unit, and an optional camera-based motion capture sensor. Tested with a standard ultrasound system, it effectively characterizes the indentation forces applied by the ultrasound probe. In in-vivo testing on the upper leg, the maximum indentation force reached 11 N, resulting in 18% tissue compression. The 6-axis load cell fully characterizes the mechanical interaction with an uncertainty of less than 1 N, allowing for precise quantification of unusual loads and ensuring repeatability and accuracy in measurements. In addition, the specialized hydrogel clip provides a dry interface free of residue, enabling high-quality ultrasound imaging without liquid gels on the patient’s skin, using an Aquaflex ultrasound gel pad, which is flexible and disposable. Figure 43. Ultrasonic imaging probe encased with force sensors (6-axis load cell) 184 . Figure 44. Overhead Origami-based Collapsible Mount (Over-COM) 185 and 186 with inner shell’s have 4 DOFs in translational and rotational motion (the 5th DOF). 184 Schimmoeller et al. Instrumentation of off-the-shelf ultrasound system for measurement of probe forces during freehand imaging, Journal of Biomechanics, Vol. 83, 2019, pp. 117-124, https://doi.org/10.1016/j.jbiomech.2018.11.032 185 Li et al. An Overhead Collapsible Origami-Based Mount for Medical Applications. Robotics. 2023; 12(1):21. https://doi.org/10.3390/robotics12010021 186 Lailu et al. Review on Wearable System for Positioning Ultrasound Scanner. Machines. 11. 325. 2023. https://doi.org/10.3390/machines11030325 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 97 The actuation prototype with the highest degree of freedom (a total of five DOF) was researched by Li et al.185 (Figure 43). The actuator is proposed for overhead human mounting, designed as a compact, lightweight, and portable supplement to automate ultrasonic imaging. It serves as a mount or holder for a transcranial Doppler ultrasonic sensor, intended to meet the requirements of ambulatory monitoring without the need for a skilled operator. The ultrasonic sensor, with the prototype actuator, is angled to acquire the maximum signal from the Doppler ultrasound. Currently, this non-automated procedure relies heavily on the physician’s intuition and requires significant practice to master. The overhead collapsible origami-based mount (Over-COM) features a lightweight, portable, and compact design achieved through its origami structure. It collapses from 10 cm to 2 cm (an 80% reduction), enhancing portability while ensuring sufficient contact force between the medical sensor and the patient’s skin. In Li’s 2023 tests, the Over-COM prototype moved the medical sensor at a velocity of up to 3.14 cm/s over a scan area of 7 cm², adequately locating the region of interest on the human body. Additionally, the ultrasound wave direction could be adjusted to form a normal angle with the skin surface within a range of 0° to 45°, optimizing Doppler signal tracking. The prototype’s complexity arises from its actuation control, utilizing eight motor-actuated strings to achieve 5 degrees of freedom (DOF) within the origami structure. A very interesting solution for controlling the contact pressure of wrist-worn PPG sensors was proposed by Jai et al.187 (Figure 45). They researched a thermo-pneumatic regulator designed to maintain consistent contact force between the PPG probe and the measurement site. A watch-type PPG platform was fabricated with an overall size of 35 mm × 19 mm. In PPG measurements on the radial artery at the wrist, while the wrist posture was changed to extension, neutral, or flexion, the regulation of contact force ensured consistent PPG readings, minimizing variations in the PPG amplitude (PPGA). A thermo-pneumatic actuator, which converts thermal energy into mechanical energy through fluid expansion, was selected as the force regulator. Figure 45. Human wrist wearing the force-regulator integrated 187 PPG sensor. This actuator has the advantage of generating a force of up to 2 N and can be easily integrated into the portable PPG platform. The total size of the current PPG platform is 35 mm in diameter and 19 mm in thickness. The contact force regulation provided a constant force of 0.6 N between the PPG module and the skin. Jai and colleagues187 concluded that by using contact force regulation based on a feedback algorithm, the proposed PPG platform significantly improved PPG measurements, reducing variations in the PPGA from 43.2% to 7.2%, despite changes in wrist posture. The TSA must meet several key requirements to ensure optimal performance in vein imaging. It should enable precise roll angle adjustments with a resolution of at least 0.1 degrees and support a roll range sufficient to cover all necessary angles for complete vein imaging. Real-time adjustments should be possible, either automatically based on imaging feedback or manually controlled by the clinician. Once the optimal angle is found, the TSA must stabilize the transducer to maintain consistent imaging during compression and other 187 Jai et al. A contact-force regulated photoplethysmography (PPG) platform. AIP Advances 1 April 2018; 8 (4): 045210. https://doi.org/10.1063/1.5020914 D2.3|Technical and functional requirements v2.0 | 12 Nov 2025 98 diagnostic maneuvers. Additionally, the TSA should be integrated with the TCA, ensuring that transducer adjustments are synchronized with tissue compression for optimal imaging results. Safety and precision are paramount, requiring the incorporation of sensors to monitor and correct any drift or unintended movement of the transducer. Furthermore, all movements should be smooth and precise to avoid disrupting the imaging process or causing patient discomfort. A main component of the TCA and TSA is a pump. The ultrasonic piezo disc pump is a relatively recent innovation 188 in precision gas pumping for medical applications. Because it operates in the ultrasonic region (>20 kHz), it doesn’t produce noise and therefore could increase usability of the DVT monitoring device. There are various available commercial devices available on the market 189 , 190 , 191 , 192 . The Pump189 (Figure 46) can achieve a flow rate of up to 2 L/min and is lightweight, weighing only 11 g. The ThrombUS+ project will determine whether such a small piezo pump is capable of providing enough air pressure, with sufficient speed characteristics, to drive TCA and TSA for CDUS and VOP in DVT monitoring methods. Figure 46. Ultrasonic piezo disc pump with an electronic driver189. 4.3. Summary of knowledge gaps and potential solutions with implications for requirements − Several approaches, characteristics, and parameters were analysed for the realization of a precise tissue compression actuator (TCA) to be used in compression duplex ultrasound (CDUS) and venous occlusion plethysmography (VOP) methods for DVT monitoring. − The analysed actuators are based on different operating principles. Industrial developments, in particular, offer a wide range of enabling possibilities. Pneumatic actuation is the preferred method for biomedical applications, especially for in vivo use. Mechanical robotic principles are still not sufficiently advanced for implementation in on-body-mounted applications. − Imaging transducer interactions with soft tissues are currently monitored using load cells, which aim to characterize the subtle and subjective maneuvers performed by sonographers with the US imaging 188 Li et al. “A review of recent studies on valve-less piezoelectric pumps.” The Review of scientific instruments vol. 94,3 (2023): 031502. https://doi.org/10.1063/5.0135700 189 Smart pump module. [Online]. Available: https://www.theleeco.com/product/smart-pump-module/# 190 Piezoelectric Micro Pump Testing Kit. [Online]. Available: https://www.auroraprosci.com/custom-oemfluidics/oem-miniature-pump/piezoelectric-micro-pump-testing-kit-for-liquid-or-gas-maximum-flow-rate-25-mlmin-@-60-hz-or-120-hz 191 MPM5 piezo pump. [Online]. Available: https://www.maxcleversz.com/html_products/MPM5-piezo-pump120.html 192 muRata. [Online]. Available: https://www.murata.com/products/mechatronics/fluid