Design of bpm modeling on skin cancer: dynamics actuate modeling business process management on skin cancer screening and optimization

i DESIGN OF BPM MODELING ON SKIN CANCER Zhuo. JING DYNAMICS ACTUATE MODELING BUSINESS PROCESS MANAGEMENT ON SKIN CANCER SCREENING AND OPTIMIZATION Dissertation for obtaining the Master’s degree in Information Management i NOVA Information Management School Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa DYNAMICS ACTUATE MODELING BUSINESS PROCESS MANAGEMENT ON SKIN CANCER SCREENING AND OPTIMIZATION by Zhuo. JING Dissertation for obtaining the master’s degree in Information Management/ Master’s degree in Information Management, with a specialization in Information System and Technology Management Advisor: Professor Dr. Vitor Duarte dos Santos (NOVA IMS) Co Advisor: Professor Dr. Carolina Santos (NOVA ENSP) January 2022 ii ACKNOWLEDGEMENTS Many thanks to my thesis advisors, who are extraordinary patience and consistent encouragement, provide me great help by necessary materials and advice. It is the professor's professional guidance and more creative thinking that inspired my thesis and model creation. Thank you very much for the research and knowledge transfer in the past two years. Special acknowledge is given to my respectable Director of IPO Operation Room-- Professor Luís d’Orey and Director of the dermatology department – Professor Cecilia Moura. Thanks to participating in the review and validation of the paper from Weiwei.Liu, Rodrigo Almaraz, Bharat Hooda. From Liga Portuguesa Contra o Cancro, Administrative and Financial Coordinator, UDPV, Portuguese Union against Cancer--Hugo Tavares, that we have acknowledged that currently, skin cancer screening programs have not been launched due to technical problems. iii Abstract With the wide application of big data platform today, many industries are eager to find a breakthrough in solving business process and information transformation. The application experience in the traditional industrial field has drawn a conclusion that there is no perfect process, and only by trying to innovate and transform the sluggish and complicated working mode can cloud data give full play to its characteristics of fast, accurate and forecast. Due to the special working mechanism of Business Plan Model (BPM) technology application in Skin Cancer Screen and Optimization, the customer group is not an isolated consumer group ---but a strong dependent candidate. Both patient uncertainty and data collection are confined, which restricts quantitative analysis by reference to many simulated data. However, it is also where we seek to improve effectiveness such as addressing cancer screening issue. Through the analysis of universal healthcare systems and cancer diagnosis processes, we hope to create new BPM solutions that combine dynamic data with dynamic business platforms. This also provides a theoretical reference for the multi-function and expansion space of the medical system in industry 4.0. Keywords BPM; Governance; Skin Cancer; Screen and Optimization; Skin Inspection; Framework iv INDEX 1 INTRODUCTION .............................................................................................................. 1 1.1 Background and Problem Identification ................................................................. 1 2 LITERATURE REVIEW ...................................................................................................... 5 2.1 Healthcare Research .............................................................................................. 5 2.2 Screening And Optimization ................................................................................... 7 2.3 Feasibility of Screening ........................................................................................... 8 2.4 Current Challenges ................................................................................................. 8 2.5 Governance and Decision Definition ...................................................................... 9 2.5.1 Public Governance on Primary Health Care ........................................................ 9 2.5.2 Clinic Governance ................................................................................................ 9 2.5.3 Data Governance Impact Decision .................................................................... 10 2.5.4 Enforceability by AI and Governance Decisions ................................................ 11 2.6 Data Security and Risk .......................................................................................... 11 2.7 Data Transaction Strategy .................................................................................... 13 2.8 Case Study ............................................................................................................ 14 2.9 Business Process Management ............................................................................ 15 2.9.1 BPM Application Importance Research ............................................................. 15 2.9.2 BPMN 2.0 Introduction ...................................................................................... 15 2.10 Medicine-Meta-Analysis Importance Research ................................................. 16 2.10.1 Process Ming in Medical Diagnosis Application Research .............................. 18 3 METHODOLOGY ........................................................................................................... 21 3.1 Research Strategy ................................................................................................. 21 3.2 Actions Identification ............................................................................................ 22 3.3 DMAIC Method ..................................................................................................... 25 3.4 National Healthcare Strategy ............................................................................... 26 3.4.1 Electronic Medical Recording System ............................................................... 26 3.4.2 Recognition of Entities ....................................................................................... 27 3.5 Skin Cancer Classification ..................................................................................... 28 3.6 Interview As Method for Qualitative Survey and Validation .............................. 29 4 STUDY ........................................................................................................................... 33 4.1 Identification of AS-IS Processes .......................................................................... 33 4.2 AS-IS Model........................................................................................................... 33 4.2.1 AS-IS Self-Examination Process ......................................................................... 34 v 4.2.2 AS-IS Dermoscopy Detection Process ............................................................... 36 4.2.3 AS-IS Image Assessment Process ....................................................................... 38 4.2.4 AS-IS Practice Check-Up Appointment .............................................................. 38 4.2.5 AS-IS Radiology Appointment ............................................................................ 39 5 CRITICAL ANALYSIS ....................................................................................................... 41 5.1 Simulation Analysis ............................................................................................... 44 5.2 Recognition of Entities .......................................................................................... 46 6 DESIGN TO-BE DIAGRAMS ............................................................................................ 49 6.1 TO-BE Skin Cancer Screening Decision Structure Flow ........................................ 49 6.2 New Macro Process in TO-BE Model .................................................................... 51 6.3 New Modules in TO-BE ......................................................................................... 54 6.4 TO-BE Appointment Service Process .................................................................... 54 6.5 EMRS Trigger Activate in TO-BE Design ................................................................ 56 6.6 Skin Cancer Resource Assignment Process (TO-BE) ............................................. 60 6.7 Data Requirement Service Process ....................................................................... 60 6.8 TO-BE Data Service Framework Design ................................................................ 62 7 TO-BE MODEL ANALYSIS .............................................................................................. 63 7.1 Simulation Analysis ............................................................................................... 63 7.2 Issues Risk Awareness .......................................................................................... 64 8 VALIDATIONS ................................................................................................................ 65 9 DISCUSSIONS ................................................................................................................ 69 10 CONCLUSIONS ............................................................................................................ 71 11 LIMITATIONS AND RECOMMENDATIONS FOR FUTURE WORKS ................................ 72 BIBLIOGRAPHY ................................................................................................................. 73 ANNEXES .......................................................................................................................... 84 ANNEXES -1 IPO SELF-EXAMINATION PROCESS ................................................................. 84 ANNEXES -2 SAMPLE OF E-REFERRAL DATA FORM ............................................................. 85 vi LIST OF FIGURES Figure 2-1--BPMN Application Example ................................................................................... 16 Figure 3-1--Methodology Scheme (prepared by author)......................................................... 21 Figure 4-1--Skin Cancer Screening Status Diagram (AS-IS) ...................................................... 34 Figure 4-2--Self-Examination Process Diagram (AS-IS) ............................................................ 35 Figure 4-3--Dermoscopy Detection Process Diagram (AS-IS) .................................................. 37 Figure 4-4--Image Assessment Process Diagram (AS-IS) .......................................................... 38 Figure 4-5--Practice Check-Up Appointment Diagram (AS-IS) ................................................. 39 Figure 4-6--Radiologist Appointment Diagram (AS-IS) ............................................................ 40 Figure 6-1--Skin Cancer Screening Decision Structure Flow (TO-BE) ....................................... 50 Figure 6-2--Health Center Task-tree Diagram (TO-BE) ............................................................ 51 Figure 6-3-- Skin Cancer Screen Flow (Macroprocess) Diagram (TO-BE) ................................. 52 Figure 6-4--Task-Driven Skin Cancer Screening Flow Diagram (TO-BE) ................................... 53 Figure 7-1--Appointment Trigger Resource Utilization (TO-BE) .............................................. 63 Figure 7-2--Appointment Trigger Simulation Analysis (TO-BE) ................................................ 63 vii LIST OF TABLES Table 3-1--Actions Descriptions Identification ........................................................................ 23 Table 3-2--Tumor Staging Symptoms....................................................................................... 26 Table 3-3--System and Data Forms Engaged Process Introduction ......................................... 27 Table 3-4--Data Flow Recognition ............................................................................................ 28 Table 3-5--Skin Cancer Classification (Example) ...................................................................... 28 Table 4-1--Identified Process & Module Name ....................................................................... 33 Table 5-1--Critical Analysis ....................................................................................................... 42 Table 5-2--Skin Cancer-Self-Examination What-IF-Analysis Utilization Rate ........................... 44 Table 5-3--Entity Recognition .................................................................................................. 46 Table 5-4--Assessment Measure Matrix .................................................................................. 47 Table 6-1--New Process Design (TO-BE) .................................................................................. 51 Table 6-2--New Modules Tasks and Responsibilities (TO-BE) ................................................. 54 Table 6-3--EMRS Trigger Data Flow Introduction .................................................................... 56 Table 7-1--Major Risk (TO-BE) .................................................................................................. 64 viii LIST OF ABBREVIATIONS AND ACRONYMS CT Computerized Tomography MRI Magnetic Resonance Imaging LNM Lymph Node Metastasis CG Computer Graphics TUS Tele-Ultrasound Scan PaaS Platform as a Service BPM Business Process Management IBPM Intelligence Business Process Management CPS 5C CPS 5Componetns Architecture EMR Electronics Medical Records IoT Internet of things IaaS Infrastructure-as-a-Service BPR Business Plan Reengineering BPA Business Plan Analysis BPMMM Business Process Management Maturity Models BCC Basal Cell Carcinoma SCC Squamous Cell Carcinoma MC Melanoma Carcinoma MCC Merkel Cell Carcinoma CBCC Cutaneous squamous Basal Cell Carcinomas CSCC Cutaneous Squamous Cell Carcinoma KS Kaposi Sarcoma KA Keratoacanthoma AK Actinic Keratosis 6 Table2-1--Healthcare Levels Introduction in Portugal Healthcare Levels Roles Activity Features Admission National Healthcare Service (NHS) Encompass range scope of healthcare including health surveillance, prevention, diagnosis, and patient treatment. Characterized as being national, universal, general, and free service to all citizens and foreign residents with public permissions. Central Administration of the Health System (ACSS) and by the five regional health administrations (North, Center, Lisbon, and Tagus Valley, Alentejo and Algarve) Primary Healthcare General doctors and nurses at a physician (Doctors, Pediatricians, Geriatricians, Nurses, Nurse Practioners, Physician assistants) Emergency Care; Urgent Care; Chronic Disease Care without self-inability; Mental Health care; Hospice Care Community Hub Management Health centers groups including Home and Community care (agrupamentos de centros de saúde, ACES) Second Healthcare Medical professionals (Allergists, Infectious Disease Doctors, Ophthalmologists, Endocrinologists, Dermatologist) Recommended by Primary care for suggestive treatment plans in prescribing medication through triage of nurse; Shortterm Treatment Hospital establishments (Hospital Center) Tertiary Healthcare Involved complex and advanced equipment, treatment, or procedures Oncological Treatment; surgical interventions Local health unitsunidades locais de saúde, ULS Regional Healthcare Center Independent Healthcare Service (Managed by respective regional administrations) Regional Healthcare Risk Assessment, Prevention autonomous regions health services (serviços regionais de saúde, SRS) in Azores and Madeira Private Healthcare Independent Healthcare Service for specific groups or organizations Special schemes parallel to SNS for the providing of healthcare. Some subsystems are financed by State or appointed medical service providers Like National Republican Guard, and Public Security Police and military, Ministry of Justice e-Health Network Terminals Applied into Healthcare Practices and Information Communications Electronic Health Record. Computerized physician operation. Telemedicine. 7 Tele-surgery. Tele-examination Consumer Health Informatics National Network of Continuing and Integrated Care-Rede Nacional de Cuidados Continuados Integrados (RNCCI) National Healthcare Network For long term maintenance of chronic illness nursing and care Convalescent units. Medium duration and rehabilitation units. Long-term and maintenance units. Integrated continuing Care Team-Home. Mental health unit Long term and maintenance units-Unidade de Média Duração e Reabilitação (UMDR) Integrated continuing care team-Equipas de Cuidados Continuados Integrados (ECCI) 2.2 SCREENING AND OPTIMIZATION From the basic living environment, human genes, race, to work environment, the sensitivity and difference of the application of ABCDE screening principles (Lucas et al., 2016) as the standard of technical means and a series of screening factors to obtain the screening discrimination criteria and decision-making methods affecting human infection with various skin cancer diseases. To improve the screening effectiveness through dermoscopy and cloud-based services in comparing individual history data with other relevant periodic data of suspected cases the core significance of this paper as a reference is to provide a business plan model (Hossain & Muhammad, 2016) (Muthukumaran et al., 2021). Mnemonic “ABCDE” description is as follows (Lucas et al., 2016) Table2-2--ABCDE Guidance Table Index Sign Description A Asymmetry The two parts from one side to the other side are not matchable same in shape and size B Border Has irregular (uneven, crusty, or notched) edge to define the melanoma lesions borders with skin C Color In a variety of colors, White, Blue, Brown D Diameter Larger than 6 mm (1/4 inch) in diameter of melanoma E Evolving Change in growth size over time or bleeding, or scabbing (E) Elevated Protuberance above skin surface (F) Firm Hard texture to touch (G) Growing Vary irregularly 8 As a dynamic decision model, it aims to break through the barriers of loose organization and resource communication and provide the latest and reliable clinical and non-clinical evidence to support innovative clinical decision-making. Skin cancer screening for the dynamic model of the building will not only depend on the potential patient's information screening, cases of transverse analogy, the illness progress report, and forecast data simulation, seamlessly between patients with dermatologist and data analysis service including simplified unnecessary file handling process, low time-consuming and hospital medical personnel screening costs. 2.3 FEASIBILITY OF SCREENING From the evaluation result of the standard parameters for melanoma screening especially on tumor thickness comparison, there is evidence of a clear relationship between mortality and early detection of diagnosed skin cancer (Shellenberger et al., 2016)(PDQ Adult Treatment Editorial Board, 2020) The digital dermoscopy has proved greatly improve the detection operation effectiveness, but not yet reached out of totally depending on its current accuracy and specificity of remote screening heretofore (Thomas & Puig, 2017). In addition, optical technology basis on artificial intelligence application (Image Deep Learning) (Kermany et al., 2018) can be developed to explain how much capacity there is still to support the analysis of skin cancer lesions, multispectral mapping, and diagnostic confidence required for joint solutions in related fields. Common skin cancer screening techniques and measures are always mentioned as following (Situm et al., 2014)(Esteva et al., 2017)(PDQ Adult Treatment Editorial Board, 2020): ⚫ Skin observation – magnifying lens; ABCDE guidelines. ⚫ Skin biopsy -- local anesthesia; Pathological excision analysis. ⚫ Microbiological Test -- skin or nail testing; A fungus or bacterium. ⚫ Photosensitivity Test -- light source exposure test (ultraviolet, visible, infrared); Report of skin sensitivity reaction. ⚫ Contact Test -- Test for contact dermatitis caused by allergens. ⚫ Dermoscopy -- manual or digital dermoscopy; Timeline of state cell changes. ⚫ Confocal laser reflection microscopy testing -- depth focus optical stratification three latitude capture, depth scanning acquisition, plane mapping, and stereo mapping, focus information collection biological sample detection. 2.4 CURRENT CHALLENGES ⚫ Lack of adequate imaging screening results in low accuracy and efficiency in identifying cancer (Dlamini et al., 2020). ⚫ Lack of core processing capability in data processing for skin cancer's multiple complex factors and conditions (Iyer et al., 2020). 9 ⚫ The relative imbalance of medical resources is mainly reflected in the lack of skin cancer doctors and excessive concentration of resource allocation waiting in Primary Healthcare Center (PHC) (Helfand et al., 2001). Meanwhile, available resources in secondary clinics are not fully activated and utilized, leading to a long waiting time for patients. ⚫ The differentiation of skin cancer types and the single dependence of non-adult skin cancer screening methods require further refinement of screening procedures as a majority of pediatric skin cancer (Maguire-Eisen, 2013) (Pérez & Bashline, 2019). ⚫ At present, although skin cancer screening process for regional cultural differences, and by the laws and regulations, ethics and social factors such as restriction of science and technology, has not yet fully consensus on the process of the official certified framework basis for lacking relevant security definition (Eleanor Bird et al., 2020). 2.5 GOVERNANCE AND DECISION DEFINITION Decisions on Skin Cancer Screening and Optimization are emitted request of the Ministry of Healthcare, which grants authority to health centers and service unit authorities. In addition, the Ethics Committee and Data Security Oversight Board have protection and interpretation power over all procedures and Data processing related to the process (Fabiano, 2019). 2.5.1 PUBLIC GOVERNANCE ON PRIMARY PUBLIC HEALTH CARE Primary Health Care (Cueto, 2004) is essential health care made universally accessible to individuals and families in the community by means acceptable to them. The detection and collection of such data will be sent to national public health centers and regional governance centers promptly when public safety and health and environmental issues are involved. This series of programs aim to improve the quality of social public services, prevent and diagnose cancer in the early stage, and accelerate the construction of a sustainable technology platform (Rosemann & Vom Brocke, 2015)(Low et al., 2019) for cross-sector and cross-functional big data transformation. Another major feature of the project is the introduction of individual experience collation and feedback combined with data safety and feasibility, as the transition of the next stage objective of evaluating the construction of dynamic cancer database feedback mechanism, to finally achieve the effect of improving the quality of life of cancer. 2.5.2 CLINIC GOVERNANCE By 2000, there is not yet specified on “how the health service organization framework for Australian Clinic System, but it guided defining the system implementation needs, scope, service environment, and roles and responsibilities of relevant personnel” (reported by “National Model Clinical Governance Framework, 2000). Till 2010, they have developed out standard descriptive clinical governance models commonly activating jointed among the United Kingdom, Australia, and New Zealand as guidance, where infinitely clarified “National Level, Regional Level, Service Level, Multilevel Approaches” with fewer discussion on “Quality” session involved into that framework(Phillips et al., 2010)(Buja et al., 2018). 10 The emergency services mechanism(Catalano et al., 2003) is also subject to the basic principles and assessment criteria of the framework. The openness of resources, centered on supporting environmental improvement and applied governance and security of services, has gradually expanded into a framework for building collaborative public sector and non-profit organizations to work on primary health care clinical governance. Currently, we are unable to confirm the personal individual root cause of skin cancer, but the clues indicate malignant skin cancer (including melanoma, and non-melanoma) contracts risks are affiliated to human factors, ethnicity, excessive ultraviolet sunlight or instruments sunburn, skin protected area exposed to the chemical carcinogen of the environment medium factors may increase the risk. However, how to use physical media (mobile and portable devices) to obtain personal real-time data statistics and feedback to the dermatology department for diagnosis is restricted by personal survival and living environment and status, as well as limited, acknowledge the prevalence of finding the root causes of skin cancer in patients at the primary stage. For example, statistics on how long some potential patients have worked with chemicals, the proportion of vitamin-D in their bodies, the length of time they have been exposed to direct and indirect ultraviolet radiation, and the detection of their skin conditions basic on “ABCDE” guidance skin self-examination. Normally, the results of visual detection and professional instrument detection combined with the actual conclusion process including pathological picture recognition, time series analysis and comparison of tumor changes, analysis, and comparison of similar pathological pictures and other factors will morph into the main influencing components for the dynamic acquisition of individual skin cancer screening results. 2.5.3 DATA GOVERNANCE IMPACT DECISION After the introduction of the logistic regression (LR) model based on traditional risk quoted judge only learning model based on X-ray image depth and blend to rebuild based on the above two methods of model contrast test unceasingly, scientists have been driving up AI algorithms for some auxiliary judgment accuracy and clinical use. However, in the primary screening of skin cancer (including melanoma and non-melanoma tumors), the rationality and effectiveness of the data process still need to be deduced through valid data and analysis, to provide suggestions for improvement of the process for relevant medical physicians and experts. In 2011, Governance Definition (Elango et al., 2010) was discussed in the organization's poor management of information, such as ambiguous goals, lack of resources, and inconsistent processes, which caused conflicts among people in the organization and conflicting decisions. To avoid unreasonable confusion, the governance framework is more advocated to strategize for interactions with balancing each side of stakeholders that is essential to be proved scope of acceptance first. RACI models definition map (Rose, 2013) figured out who would be responsible for a decision including datacollection, options-identifications, and recommendations, etc.,. Conflicts from different perspectives may lead naturally to impacting security, performance, and quality. Without the clinicians, managers, and members of governing bodies will not have individual and collective responsibility. For example, Model-Driven Design (MDD) application approach has upgraded traditional methods of challenging with intrinsic complexity of integrating large-scale business processes and existing applications (Perez-Castillo et al., 2019), that is, introducing data warehouse metamodel and platformspecific transform models can present scenarios from a pilot project in technological solutions. The 11 Behavior-Driven Design (BDD) (Freet et al., 2015)(Moe, 2019) will be goal-oriented and record test process and test results. With the assistance of deep learning tools and remote intelligent devices, skin cancer experts can automatically cross-compare similar skin cancer pictures and diagnosis results and preliminarily infer conclusions. At the same time, the BDD development mode will also provide feedback to the error processing center to summarize the specific reasons for the process error and feed back to the system development staff. 2.5.4 ENFORCEABILITY BY AI AND GOVERANCE DECISIONS AI tools connect different data sources to support the screening of skin cancer patients for potential risk and collection of skin cancer-like symptoms. The algorithm was used to predict the difference between symptoms and conclusions. However, how to solve the transfer and identification problems of different file formats across systems like graph-zip in High-Definition presentation and dynamicreplay, filter effective data connection points, and import the efficiency of flow data transfer has become one of the challenges that affect the rapid response of organizational decisions (Ting et al., 2019). Whereas the concern raised from standing on making it easier enforceable on organizational decision and protocol standardization within the process of data transformation and migration, is to present the information transparency and flexibility without any loss from protecting primitiveness of information. To explore the reasons of lagging and delaying against feedback in decision-making due to bias, it will provide the evidence for reducing complexity for involving skin cancer screening decisions and precision from AI reliability. The input of a large amount of data that can only be introduced manually, and the complexity of the algorithm will introduce and generate a large amount of garbage data, which will lead to insufficient and incomplete data, and the formed AI model will generate data bias. It is morally necessary to design unbiased artificial intelligence models by removing unrepresentative data and reducing the error of fuzzy comparisons (Bennett & Hauser, 2013)(Jetzek et al., 2019). 2.6 DATA SECURITY AND RISK By introducing remote digital application services in the medical Internet of things and intelligent applications, data leakage has become the focus and concern of security experts. When Data Exfiltration and Data Tampering fail customers to use normally, the illegal theft of personal privacy information, and the great loss of the reputation of the organization and relevant stakeholders. Fundamentally this may not save the protected product agreement to reach an understanding. Data Snooping (Jazdi, 2014) may also have a psychological impact on the patient. In the course of disease treatment, data can be altered randomly and lead to serious medical accidents. The main reasons for data leakage are as follows: ⚫ Phishing emails and websites --users click on links with Trojan viruses without any purpose and are directly implanted into the client devices, resulting in the contamination of the computer storage devices. 12 ⚫ The communication port will be modified due to the large-scale and complex hacker attack on the medical health care data, failing in the normal transmission of network data. ⚫ End-to-end data loss directly leads to telemedicine diagnosis failure, photo resolution distortion, inability to compare other relevant images, machine language inability to recognize images sent by the client, and other problems. In the cloud data platform solution, the basic principles of data governance are Effectiveness, Performance, Security, Accountability, and Risk Awareness. How to transform concurrent and continuous processing of cyber security and data elements (Identify, Protect, Detect, Respond, Recover) (Lee et al., 2015) into dynamic data services(I-scoop, 2017)(Jetzek et al., 2019). To consider the core problems of data consider its current risk management practices, threat environment, legal and regulatory requirements, business/mission objectives, and organizational constraints, such as network security enforcement levels. Share Decision Making Flexibility--Shared Decision Making (SDM)-- a non-compulsive medical solution for sharing diagnosis and treatment process decisions between patients and physicians. In this way, the medical staff can directly and transparently communicate the treatment methods and procedures to the patient without any bias, which could decrease the patient's concern about the risks and burdens. The diagnosis of each clinical case beyond the patient's personal case information will collect key disease detection evidence nodes, for accumulating decision files in the public medical care resource exploration and similar symptoms in the medical record archive. The SDM-Model (SDMN) -- provides a selection mode for decision-making, which can be either a multiparty conversation or a professional mode. The eco-sharing mode promoted can provide patients with automatic answers and visits for consultation on disease problems and can also make suggestions and expressions with their understanding and troubles about the disease artificially. At the psychological tracking level, this approach can provide a reference for the follow-up investigation of patients' emotional expression factors. Consequently, Multi-Dimensional Analysis (MDA) and MultiDimensional Decision (MDD) can be applied to extract discretionary solutions optimally (Zionts, 1979). The development and application of the Metaheuristic Algorithm (Robbins & Monro, 1951) provide mathematical analysis for finding different solutions (Genova & Guliashki, 2011) (Rieck et al., 2012)(Bomhof-Roordink et al., 2019). Table 2-3--Metaheuristic Methods No. Metaheuristic Methods Descriptions 1 Genetic Algorithms (GA) strategies for intensification of the search use offspring 2 Self-adaptation (SA) search process in some regions of the search space with cooperation research 3 Scatter search (SS) generates systematically a set of dispersed points and convex or non-convex combinations of subsets 13 4 Ant systems (ASY) Ant Colony Optimization (ACO) a constructive procedure that enables generate a new solution of the emergency 5 Particle Swarm Optimization (PSO) makes steps from its current position to a new position and this motion is determined as a sum of three vectors: inertia, competition, and cooperation 6 Memetic Algorithms (MeAg) a hybridization of evolutionary or any population-based approach with separate individual learning or local improvement procedures to solve a given problem 7 Particle Swarm Optimizer (PSO) Is an s solution that computational methods optimize a problem by iteratively regarding given measure or candidate solutions, different gradient descent, and quasi newton methods 8 Resource Leveling problem (RLP) Calculate the effective path in which start, and finish dates are adjusted based on resource limitation to balance the demand for resources with the available supply 2.7 DATA TRANSACTION STRATEGY The organizational governance model of public welfare organizations includes five basic models: advisory committee model, cooperative governance model, policy committee model, sponsor governance model, and management team model (Elango et al., 2010)(Jetzek et al., 2019). At the same time, due to the continuous promotion and adoption of community participation as a new intermediate responsibility service center, the view of establishing a mutual trust channel and cooperative repeater between the community and non-profit organizations is gradually derived into a mixed governance model. The structure of the organization is divided into policy committees and administrative committees(Stinchcombe, 2000)(Lunenburg, 2012)(“Governance by Committee: The Role of Committees in European Policy Making and Policy”, 2000). Policy formulation and implementation are the responsibility of the Policy Council. Under the policy committee model, the relationship between the board and the staff is a partnership. It can be divided into a management committee, working committee, policy governance committee, and collective committee. According to different social needs and environmental constraints, public health centers in each region will adopt different governance models. Table 2-4--Governance Dimensions Governance Structure Dimension Features Centralized or largely centralized structure Single authority Less Transparency Poor Accessibility Local health units are primarily led by employees of the state 14 Institution-based Trust Fewer incentives on law A decentralized or largely decentralized structure Community Partial Transparency &Trust Community-based Local health units are primarily led by employees of local governments. Mixed structure Local Transparency Targeted Accessibility Some local health units are led by employees of the state, and some are led by employees of local government. No single structure predominates. Shared or largely shared structure Complete Transparency based on codes Strong Accessibility Interactive Local health units might be led by employees of the state or by employees of local government. If they are led by state employees, then local government has the authority to make financial decisions and/or issue public health orders; if they are led by local employees, then the state has the authority 2.8 CASE STUDY For personal health information, patients' privacy belongs to sensitive content, and both technical processing and non-technical application in storage, transmission, and sharing are protected legally. ⚫ UK Public Health England (PHE) plans to implement Cancer Taskforce Cancer Strategy for England 2015-2020. Implement coordinated and exploit Cancer prevention and detection governance measures and strategies across organizations. Sharing expertise, and creating, which emphasizes synergy and cost savings, are the three areas. ⚫ NRCARS provides clear data support for the establishment of a professional framework for cancer assessment and health impact that will be useful for the commissioning of the patient, professional, media, communications, and professional facilities. ⚫ The “Privacy Act of 1974-The Privacy Act, 1996-Health Insurance Portability and Accountability Act, HIPAA, 2003Privacy Rule & Security Rule in HIPAA” came into effect gradually shaping the personal Health Privacy protection laws and regulations in The United States. The privacy rules (Protected Health Information, PHI) are involved. From the database file sorting and division, the personal Health information archived includes two parts, (Electronic Medical Record, EMR, and Patient-Generated Health Data, PGHD). From the perspective of privacy rules, identification Health Information (IIHI) can be defined as a subset of Health Information, which can assist in providing individuals with sustainable personal Health care records, and physiological conditions. Even though many countries have enacted laws regarding data security and the scope of data use, the personal information of 25 million AMCA patients was hacked in 2019. The breach of experimental data, personal and financial data, social security accounts, and medical information led to the bankruptcy of AMCA's parent company and the suspension of Quest Laboratories. 15 ⚫ The serious consequence caused by network attacks makes the system collapse, the data flood, the use of false data, and duplicate data directly affect the credibility of data. Dynamic ecosystem health to assist decision-making, like the "bank" operation mode "CHRIS Dynamic Consent (DC)" method, Oxford University "SPRAINED" project, the University of Manchester "GenBank" project, Japan's "RUDY" project, and the patient groups genetic alliance "PEER" platform, and so on to upgrade the original low efficiency "E-Consent" version is safe and effective for health data sharing as the core idea of open a new window. 2.9 BUSINESS PROCESS MANAGEMENT 2.9.1 BPM APPLICATION IMPORTANCE RESEARCH The Design for Six Sigma (DFSS), discussed how to improve customer satisfaction (Cudney & Furterer, 2016). A maturity BPM model is persistently optimizing and reorganizing the resources to undertake the upstream and downstream of the business through sorting and screening(van der Aalst et al., 2007)(Perez-Castillo et al., 2019). Simply covering business process actions cannot activate specific tasks stakeholders and data transmissibility in attenuated, congested, or duplicative functional performance, when burdensome and intricate BPM models in many enterprises and organizations come to draggle operation. Redundant and unnecessary entity setup, as well as the complexity and inefficiency of the execution process due to the vague planning of functional boundaries, are often the two main hurdles to the successful implementation of business processes. Reducing the coupling degree between units and role layers always requires the seamless connection of highly integrated intelligent dynamic modules, and the processing machine that can deal with various emergencies flexibly strengthens against the priority of digital transformation and the logical processing sequence of event processing. Confronting at this point, Big-data, Data-Cloud collection, and sharing are available to be auxiliary data-driven boost into evoking intelligent model, the balance of fault-tolerant capability, and improve the request response speed. However, whether event-driven(Sun et al., 2021), function-driven, or task-driven method, the technical requirements for the key elements of executable capability are extremely trenchant. That expresses in reducing the pressure on the server-side means that each drive module has a corresponding data control response. For the urgent demand of the intelligent transformation of medical institutions, more units containing superb medical skills and can sophomore technological servers be essential the physicians, and communication channel utility (Dimarogonas et al., 2012)(Ji et al., 2021). Excavating data mapping plasticity through optimizing cancer workflow will pursue establishing a BPM model that conforms to medical care. Given the purpose of decision-making to promote task efficiency and effect the high-level management goal of BPM is the interaction that demands to be achieved through the introduction of low-level management goals to generate decision values. It is essential to reevaluate vulnerabilities and also transform successful experiences into strategizing outline of the business plan (Hoerbst et al., 2011)(Barbazza et al., 2019). 2.9.2 BPMN 2.0 INTRODUCTION The Business Process Model and Notation (BPMN) is a set of graphical notations that can express complex Business Process languages. To realize the standardization of business process modeling and 22 3.2 ACTIONS IDENTIFICATION Through background investigation and question sorting for skin cancer screening, we have clearly obtained the main actions and descriptions involved in the current process. Here, they are summarized into the process description and displayed in the table, so as to analyze the problems existing in the current screening process (AS-IS 23 Table 3-1--Actions Descriptions Identification Activity Main Steps Descriptions Processes Invoke Skin Cancer Diagnosis Face-to-Face Diagnosis Main Detection for Skin Cancer particularly to those who are not trusting the Teledermoscopy or Non-Remote-EquipmentHolders. Need to take exam in Clinics Self-Examination Process Remote Diagnosis dermatologist guides the procedure through a dermatoscope in Semi-RemoteDetection / Tele-dermatology or Fully Remote Detection Equipment Training Self-Examination Process Treatment Surgical Treatment Biopsy Surgery for pathological examination by obtaining the suspicious skin tissues Radiologist Appointment Process Preoperative Observation (Prepare for check-up) Regular Periodic Checking against high-risk groups including genes screening Self-Examination Treatment Advice Advise by dermatologist, pathologist, GP, radiologist, and other relative examinees after examination in summarized report or images. Self-Examination. Dermoscope Detection Process. Image Assessment Process Remote Skills Dermascopes Digital Dermascopes Non-Digital Dermascopes Portable Skin Cancer Examination Application Dermoscope Detection Process. Self-Examination Assessment Self-risk estimation and suspected skin cancer screening Self-Examination Process Dermatology Confirmation Dermatologists draw conclusions based on experience, comparison of images, pathology, and genetic examinations Dermoscope Detection Process. Devices Interaction CT Scan Test result of soft tissues variations in details and if exists any internal organs were spread of melanoma in suspicious spots Image Assessment Process MRI Detect the degree of structures nearby particularly non-melanoma skin cancer affected places on eyes or salivary glands. Image Assessment Process TUS (Tele-Ultrasound Scan) High resolution images of tissue structure of skin application surface were obtained by ultrasound non-invasive imaging technology, and the diagnosis was compared with clinical and dermatoscopy images of suspected skin cancer Image Assessment Process FDG-PET Adjuvant detection of tumor stage with skin cell changes and lymph node metastasis Image Assessment Process 24 X-Ray Support detecting in the images of the lymph nodes enlarging to bones caused by melanoma spreading Image Assessment Process Nursing Quality Triage Assignment by skin cancer disease detection degree and priority in the risk of treatment based on resources and inspection needs Practice Check-up Appointment Process Radiologist Appointment Process Schedule Arrangement Request a diagnosis from a doctor, clinic, dermatologist, imaging specialist, or pathologist Dermoscope Detection Process Practice Check-up Appointment Process. Radiologist Appointment Process Administration Policy and Decision Regulate and publish the examination methods, instrument implementation methods and parameter standards, screening and examination period, diagnostic treatment, case analysis approach, non-technical participation screening route, etc. Administrator or Technician Skills Screening and Optimization Statistics Report based on individual screening and community screening results are aggregated Administrator or Technician Skills Electronic Medical Records (EMR) Anamnesis Personal skin cancer screening experience and screening results Practice Check-up Appointment Process Radiologist Appointment Process Dermoscope Detection Process Dermoscopy Detection Dermoscopy tutorial steps Differentiate between Melanocytic and Non-melanocytic Lesions. Differentiate between Benign Melanocytic Lesions and Melanoma. Propose a data-driven oriented BPM model scheme that can be adapted to screen human completion tasks and RPA automatically triggering business flows. Dermoscope Detection Process Patient Psychology Training Personal skin cancer basic knowledge and related equipment use counseling Self-Examination 25 3.3 DMAIC METHOD Quality Management aims to eliminate deviations in business processes, reduce defects and work-inprogress (WIP), and increase manufacturing capacity and speed of operation. In the transformation design of business processes, the introduction of Design for Six Sigma's (DFSS) strategic philosophical thinking is helpful to improve customer satisfaction and promote sustainability of service quality competitiveness (Ruben et al., 2018). Extended from DMAIC, one of data-driven management conception mode, business process problem discovering and process optimization designing in healthcare management framework can be elicited as following steps: ⚫ 1st step-Define — To Define the problem in the Medical Care field particularly in Skin Cancer Screening. So far as now we have discovered a lot of issues among medical sources utilization, disease treatment, and patients from research. All the exposure issues will be clarified into the current-- AS-IS definition during BPM model designation. ⚫ 2nd step-Measure —To Collect Data on the current process (AS-IS). The relative data originating from Skin Cancer Screen and Optimization will refer to all Structural and NonStructural format files, documents, scanned images, etc. which will be introduced into the engagement of the whole process. ⚫ 3rd step-Analyze — To enhance Interpret Data and exclaim suggestions. To escalate more ponderable information that could be executable for transformation and transactions for the future (TO-BE), we also need to quantify the result of each main and sub-process workflow in metrics for delivery. ⚫ 4th step-Clarify — To Clarify BPM objectives of the outcome and benefits regarding current BPM activities measurement, we will set the momentum and intense purpose with specific goals towards performance improvement and changing improvement, which will help us to understand intelligent BPM implementation importance by data-drive conceptual design. ⚫ 5th step-Assume — To Assume process and sub-process implementation in scalability and agility. In this step, we will identify each enhancement on changing to process along with expecting to improve the performance of our operations and to reduce risks of bottlenecks by resolving explorative issues, it is necessary to create integration scenarios for productivity improvement upon using automation and process standardization in the scope of cost confined. Besides, the complexities in our business processing, we will consider how to make more integrations and a new repository of data to meet needs with agility. ⚫ 6th step-Improve — Building up one reliable BPM model will depend on what we will have explored with technological skills and analytical methods in our business process. The new model will be declared in optimized design to screen out how much efficiency has been improved in actualization. ⚫ 7th step-Control —To Confine negative impact on improvement by deviation verification and reduction. Undesired deviation threshold setting in the BPM model will cause stressful tasks loading or incompact activities streamline, which will seriously impact seamlessly connecting to actualization between business and technology strategy. 26 3.4 NATIONAL HEALTHCARE STRATEGY National Healthcare Strategy (NHS) is an integrated implementation of a national-level health service, where the medical care system under this system covers the medical care benefits of the country and legal aliens residing in the territory of the country (Barbazza et al., 2019). The national health institutions are the implementors of national health protection entrusted by national law. Most public Health center physicians are affiliated with the Primary Healthcare Service (PHS), which means that citizens are entitled to make an appointment with a PHS physician through Electric Medical Record System (EMRS) Registration to engage those diagnoses and screening of chronic and terminal illnesses such as skin cancer identification and treatment (Tanbeer & Sykes, 2021). Under the service of the EMRS framework, both the request from participation and the response by physicians are designated as the responsible role protected by the public health department for the appointment, referral, consultation, and payment of a series of a disease services engagement. Table 3-2--Tumor Staging Symptoms Stage Melanoma Symptom Features Stage 0 Exist only in the outer layer of the skin without invading the inner tissue Stage 1 Exist only in the outer layer of the skin without invading the inner tissue The outer layer (epidermis) may be ulcerated Tumor’s thickness may be between 1mm and 2 mm or less than 1 mm. No ulcers. Not spreading to adjacent lymph nodes. Stage 2 Tumor thickness greater than 2 mm. Happens ulcers. Not spreading to adjacent lymph nodes. Stage 3 Have spread to surrounding tissue. Stage 4 Have metastasized to other organs, lymph nodes, or areas of the skin far from the original tumor Recur Reappear in its original position or in another organs after treatment 3.4.1 ELECTRONIC MEDICAL RECORDING SYSTEM Electronic Medical Recording System (EMRS) is an operational medical administration and clinical data management system commonly utilized in the Primary Healthcare System Framework (PHSF) (Buja et al., 2018) (Low et al., 2019), which covers the functionality including patient appointment, medical records, image capture, and process maintenance, and effectively supports administrative decisions, medical decisions, diagnostic processes and data classification, and summary, etc.. Cloud EMRS system (Muthukumaran et al., 2021) provides direction and channel for scientific largescale data processing. Behind processing, real-time data and analysis Associates cloud data sources with internal database storage systems to simulate and generate reliable data while maintaining longterm fault tolerance and self-learning capabilities. It also shares datasets with dermatologists, oncologists, imaging, and pathologists’ access to databases of similar anamnesis, improving the 27 granularity of accurate predictions by machine algorithms. However, due to the high level of technical personnel required for a fully server-side EMRS deployment, the increased operational costs will bring the burden of medical diagnostic costs, which has been conscious one of the bottlenecks in the current transformation of medical systems to medical data transformation (Khan et al., 2014). 3.4.2 RECOGNITION OF ENTITIES An entity is signified by a symbol in a business process and, as part of process standardization, an individual or attribute to a process that can operate, invoke, commit, store, record, etc. When entities are represented in the form of files, they will be specified the triggering of information processing, whether structured or unstructured. Entities are as originators and receivers of system task execution, either can be exposed for internal or implicit attributes if they are data or documents. Currently under the roles of involved system and data form in description as in the following table: Table 3-3--System and Data Forms Engaged Process Introduction System Description Engaged Processes Electronic Health Record Patient Basic Information Anamnesis Clinical History Self-Examination Practice Check-up Appointment Process Radiologist Appointment Process Pre-appointment checklist Check for suspicious lesions and patient family history Self-Examination Report Quick check-up report Complete check-up report Dermoscope Detection Process Practice Check-up Appointment Process. Radiologist Appointment Process Email/Mail By email – sending all the information in electronic format, except the imaging exams By postal mail – sending all the information in physical format Administrator or Technician Skills Image Assessment Process Telecall For appointment, training, and consulting Self-Examination Practice Check-up Appointment Process Derm.AI A framework for proceeding Dermatological Images Analysis and Skin Cancer Screening Image Assessment Process Euromelanoma System Europe Melanoma Patients Management Database External Communication Service 28 Here is the Entities enrolled system service with Input and Output present as following: Table 3-4--Data Flow Recognition Entity (Healthcare Unit/System Service) Input/Output (Report/Database) Stage & Grade of Carcinoma Stages and grades of skin cancer Report (Stage 0, I, II, III, IVA, IVB, T1-T4, N0-N3, M0) Radiology & Imageology (HC Unit) FDG-PET Report, CT Report, X-Ray Report, Dermoscopy Report, Biopsy Report, MRI Report, etc. Biopsy (HC Unit) Patient Information, Gross description, Microscopic description, Diagnosis (Tumor Type, Tumor Size) Other Information Public Source (Service) HAM10000 Images (Public multi-source dermatoscopes) Computation Center (Unit) Algorithme List and Rules 3.5 SKIN CANCER CLASSIFICATION Skin Cancer classification in research (PDQ Adult Treatment Editorial Board, 2020),(ENCR, 2015)： Table 3-5--Skin Cancer Classification (Example) Neoplasm Classification Subtype Histopathological Feature Recognition Risk Factors Melanoma (PDQ Adult Treatment Editorial Board, 2020) Superficial spreading Melanoma. (Longo & Pellacani, 2016) Skin surface raised, irregularly growing, dark brown, black, or pink. (ENCR, 2015) Female: legs. Men: Head, neck, chest, abdomen, back (ENCR, 2015) BRAF mutation (Rabbie et al., 2019) Nodular Melanoma (Situm et al., 2014) Asymmetrical, greater than 6mm; (ENCR, 2015) (Gershenwald et al., 2017) Legs, torso, head, (ENCR, 2015) PD-L1 expression (GiavinaBianchi et al., 2020) Lentigo Maligna melanoma (Occidental et al., 2020) There may be dents (ENCR, 2015)(Scope et al., 2012) Legs, torso, head, (ENCR, 2015) sun-exposed skin. elderly white population (Situm et al., 2014) Amelanotic melanoma (Gong et al., 2019) Nodules, mostly black projections, or papules(ENCR, 2015) (Cheung et al., 2012) -- Not easy to see (Gong et al., 2019) 29 Acral lentiginous Melanoma (Goydos & Shoen, 2016) Observe on palms, soles, and nail beds. All Races (ENCR, 2015) Asian or African (Goydos & Shoen, 2016) Mucosal melanoma (Olla & Neumeister, 2021) Occurs on the mucosal surface. no pigmentation, and the position is not obvious (ENCR, 2015) (Jovanovic et al., 2013) The gastrointestinal tract, anus, vagina. Women (ENCR, 2015) -- 3.6 INTERVIEW AS METHOD FOR QUALITATIVE SURVEY AND VALIDATION Qualitative research is an analysis process that relies on the continuous acquisition of corresponding data from observation, interview and participant description and recording. Such non-digital data will assist in understanding objectively existing social phenomena and individual event experiences(Liang, 2019). As the scope of the survey is very abstract, collecting the experience of experts based on semistructured interview mode to guide and find the defects and key points caused by the lack of experience, as one of the bases for feasibility analysis(Adams, 2015). This model survey can be combined with the rich experience of participants, and at least 4 or 5 experts can explain in detail and elaborate on the open thinking of the problems in joining or separate group. The questions are aimed to investigate the security implications of data governance strategies and the feasibility of remote skin cancer screening through expert surveys. Furthermore, Open-ended in-depth interview Qualitative Validation is a method conducive to objective evaluation of technical conclusions from the perspective of experts' experience. In order to respect the validity of narrative details and setting, interviewees explain and clarify emerging facts and supporting positions through several rounds of multi-dimensional exploratory questions. Question design without leading hints or answers and simply and clearly informing interviewees is the main strategy of this expert interview. Table 3-6--Research Questions and Answers Research Questions Answers by Experts RQ1: How do you concern about the open expansion of data accessibility and admission that various users own during executing the remote skin cancer screening process? EP1: Data must be secure and only available to whom the patient allows. EP2: With the world becoming digital, the privacy protection is a growing issue in the modern medical environment, which includes the issues of skin cancer screening area. 30 EP3: I think it’s very important to consider the confidentiality and security in every information process. EP4: Data can be accessed from a variety of sources ranging from hospitals to the Department of Health. Electronic medical records systems and giving patients control over permissions to view their record, sometime become incompatible. Personal finance information bank accounts info can create security issue. RQ2: How do you concern the privacy protection during the data opening transition to cloud service framework when executing Skin Cancer Screening? EP1: Doctors have an obligation to protect patients' privacy, regardless of the type of examination performed EP2: Any information given by a patient must not be passed to a third party either intentionally or unintentionally. As the consequence, it is essential that the highest level of security and digital protection is used when storing patient data in hospital or clinic. For example, hospital/clinics should uphold strict confidentiality procedures to ensure staff member is familiar with the confidentiality content and understands what information should not be shared outside of the working environment. EP3: Personally, I wouldn’t concern about the privacy protection as long as the information is properly encrypted. EP4: During Skin cancer screening, data transition to cloud service is no easy feat. To protect patient privacy while transition must balance delivering standards of patient care and meet the strict regulatory requirements set by HIPAA and other regulations, like as the EU (GDPR) General Data Protection Regulation. RQ3: Do you think digital dermatoscopy can help with remote skin cancer screening? EP1: Yes, it can. But remote screening for skin cancer requires the mobilization of a large number of medical staff. EP2: It is a fantastic modern method for monitoring skin cancer, especially in the country with ununiform population distribution. It is no doubt that dermatoscopy with proper training can help with diagnosis of skin cancer. especially for the residents in remote area. It is 31 essential for government to set up a vital objective when developing digital health in terms of cancer diagnosis/care in remote care. For example, providing adequate training to radiographer/photographer to manage mobile dermatoscopy unit which sends all of the photographs electronically to the city dermatologists and oncologists, who evaluate the images, and give diagnosis. EP3: I think it’s a great improvement opportunity in the process. EP4: Digital dermatoscopy is performed by a microscopic examination of the entire surface of the skin, locating lesions, photographing them, and representing them on complete body. It will help detection of skin cancers in the early stages of development, which helps to mitigate it. RQ4: What do you think about the role of community clinics in skin cancer screening? EP1: Depends on the clinical expertise of the clinician. The best trained doctors are dermatologists. EP2: To implement an effective communitybased programme in remote area, an issue of staff shortage needs to be solved. Extending nurses' code of conduct may take on some functions of doctors, so that to reduce the cost and to improve the service provision. It is a common use of specialist nurses to help with diagnosis and treatment in remote setting hospital. However, the skin cancer screening protocol should be followed. EP3: The role of this type of clinics is undoubtedly very important considering the high rate of patient reception. EP4: Pharmacists can detect people with skin cancer risk factors amongst their users. This intervention can be considered in multidisciplinary strategies of skin cancer screening. RQ5: What do you think about the main obstacles to implementing remote screening? EP1: We use remote screening in Portugal with no problems. EP2: Building a team remotely (a virtual team to implement screen service) could have the major obstacles to it. First, the process needs a leader 38 4.2.3 AS-IS IMAGE ASSESSMENT PROCESS In Image Assessment process, after the tumor center will classify images beyond different types of skin cancer till the location of tumor sections has been determined, and then compare them with the original images of the same type of different skin cancer types declared in HAM10000 public images space. Results produced from the oncology department will be archived whether the results detect a tumor-like map that matches a certain type of skin cancer. The health center’s task is regularly updating the archives of new pathological images it receives to analyze their clarity and accuracy and to describe their specificity. Figure 4-4--Image Assessment Process Diagram (AS-IS) 4.2.4 AS-IS PRACTICE CHECK-UP APPOINTMENT In the “Practice Check-Up Appointment” process (Figure 3.4), by following informing Cancer Exam advice from Oncology Center, the patient will make a reservation of appointment for arranging exam with Clinic and Dermatology Department in real date of advance. The helpdesk located at Clinic is responsible for assisting the triage service line when the new schedule booking abrupted is in the discussion. 39 Figure 4-5--Practice Check-Up Appointment Diagram (AS-IS) 4.2.5 AS-IS RADIOLOGY APPOINTMENT In the Radiologist Appointment process, Clinic will double-check the appointment if the existing reschedule has been updated with the patient in the EMR system. The dermatologist will review medical records to further determine whether there will have to stand in queue for waiting for X-Ray, CT/MRI, or PCGPET scanning arrangement with Radiology Department operations. Not all the patients or cases are essential to accept all kinds of Imageology Examinations, while is conditional or optional. In the terminal, anamnesis will also be updated with the latest images report. 40 Figure 4-6--Radiologist Appointment Diagram (AS-IS) 41 5 CRITICAL ANALYSIS The design of this model is an attempt to meet the transition of healthcare information technology and platforms to increasingly digital and automated applications. It also provides a theoretical reference for the development and optimization of a similar model’s process design to the cancer screening process. At present, in most of the digital transformation processes, there are many constraints in the process of module process and parameterization, and security risks and hardware problems brought by algorithm research and development and big data storage still need to be solved. ⚫ Regarding different scenarios adaptation, such as “Cancer Classification Definition”, a special group of patients and other rare skin cancer will have to be formulated a large picture of image database with intelligent algorithm application to optimize technology-aided prediction. ⚫ Moreover, the “Appointment” function of patients, diversified technical means of remote skin cancer screening will be explored by medical terminals and non-medical technician’s communications without any data security interruptions, in aiming to solving problem of patients unable to seek medical treatment nearby due to difficulty in an appointment 42 Table 5-1--Critical Analysis Issue Name Description Assumption Define screening process principles Assess Self-examination accuracy and effectiveness Popularize basic skin cancer testing with community healthcare hub ABCDE guideline; 7-point checklist (aka Glasgow checklist); Monitor Healthcare and Clinics screening process and procedure Collect effective technical data and transfer the parameters into dynamic meta-reference and service with analogy including like: PRISMA-DTA Checklist; Source of patient samples (fixed or randomized); Adaptability matching and patient sensitivity of skin cancer detection technology; The detection indexes include whether to introduce deviation calculation and threshold assumption; Testing technology implementation and standard deviation; Self -Examination The network is not available to support submission for generating self-exam estimation results by the format into EMR Timely online corresponding is required to assist patients in installing client software Self-examinations basis on dermoscopy operation is in not in line with the procedure Standardize remote self-testing instructions and procedures; Statistics on disabled participations status; Provide multimedia and community publicity with training Patients may not be able to get a whole-body test for bias, existing disease, or living conditions Provide selectional remote examinations like combine dermoscopy and distancedermatologist mixed mode Appointment and Triage Unregistered residents may not be screened for full-body examination in clinics Multiple appointment services in the agency will release the inadequate resource pressure The patient cancels the appointment for a communication gap or other unconfirmed reasons. Refunction the scheduling system by investigating the main reasons for more than two failed appointments and remedying them with technical measures. For example， EMI may occur in patients who have CIED implanted so that a Dermoscopy with magnets is used for testing Practice detection Screening for the perspective case when rare skin cancer classification is detected To assess the risk of screening for rare skin cancer subtypes regarding different stages for example, when it is unclear if the same safety margins should be applied for the respectable stage III on primary tumors Radiology is not possible as inadequate equipment in remote areas Construct remotely image exam analysis and monitoring service upon the structuring and exchanging diagnostic parameters in the protected medical channel of intermediate access authorization and permission The range of non-invasive tissue that may be excised by resection of tumors with indistinct boundaries is uncertain during surgery Compare various techniques for detecting skin resection edges to ensure minimal damage through the solutions candidates Data of diagnosis and treatment systems at different levels are not updated synchronously, leading to information asymmetry Strengthen internal communication and coordination 43 Image recognition and prediction from melanoma various tracking Image formats exist in a variety of formats, and the style is not clear in the process of storage and conversion, which makes it impossible to be used for algorithm analysis and crossdepartment reference Need to unify the image format and production produced in specification to avoid secondary image processing Diagnosis of dysplastic nevi may result in an inaccuracy calculation on melanoma lesions estimation Through the development of AI and gene mapping, light source screening, 3D technology detection, and other joint applications to locate and predict dysplastic navies early development status Expert diagnosis and advice Single diagnosis cannot be applied as a complete screening result for insufficient professional experience Arrange more professional training to expedite motivation on physician's diagnosis and treatment ability Due to insufficient authority or software incompatibility, remoted diagnosis cannot be implemented. Clarify the different level of authorities and cross-platform configurations 44 5.1 SIMULATION ANALYSIS Simulation Analysis is a result verification method using partial virtual reality data after model construction(Damij et al., 2008). This approach parameterizes and displays views for local models and events, with monitoring the completion of each module and event executed in the BPM model in duration. What-IF-Analysis is one of the techniques used by hypothesis analysis to measure the impact of conditional fluctuations. According to the condition setting of different performance indicators, the data lost during initialization is repeatedly tested and the overload or insufficiency is analyzed to find the best solution. Initial Parameter Setting in Self-Examination Model (AS-IS) of different resources including General Practitioner (GP), Nurse and Technician. (Example): Assuming 200 instances of Self-Examination for the patient to be proceeding in 30 days, each daily work loading duration last 8 hours from 9 am to 6 pm, the process, tasks, events, gateway, and completed status will be presented in figuring “Event End” result with instances fulfilled as accepted from start events acceptance. 1st Scenario: SelfExam (AS-IS) process is set 60 minutes waiting time to start. 2nd Scenario: Save waiting time for each execution step, like Waiting time is reduced to 45 minutes during each patient self-examination implementation separately. 3rd Scenario: Improve Utilization working proportion like one technician can be a candidate as an alternative service in the “Estimate Self-Examination “step. 4th Scenario: Attempt to add new resource with one nurse and one General Practitioner (GP) for serving the whole process but remaining only one technician as still. Table 5-2--Skin Cancer-Self-Examination What-IF-Analysis Utilization Rate Resource Scenario Utilization GP SelfExam（AS-IS） 66.25% GP SelfExam（TO-BE）-SaveWaiting 81.90% GP SelfExam（TO-BE）-ImproveUtilization 86.98% GP SelfExam（TO-BE）-AddNew 76.97% Nurse SelfExam（AS-IS） 69.92% Nurse SelfExam（TO-BE）-SaveWaiting 86.43% Nurse SelfExam（TO-BE）-ImproveUtilization 96.35% Nurse SelfExam（TO-BE）-AddNew 94.74% Technician SelfExam（AS-IS） 75.52% Technician SelfExam（TO-BE）-SaveWaiting 77.40% Technician SelfExam（TO-BE）-ImproveUtilization 64.16% Technician SelfExam（TO-BE）-AddNew 63.49% 45 The resource utilization by tasks proceeding contrast in the chart. Figure 5-1--Resource Utilization Task Proceeding 5th Scenario Given 40% of patients decided to accept Dermoscopy for skin-cancer examination in a remote way, which means more nurses and GP are expected to enroll as training guides or entirely consultants. Figure 5-2--Stress Prediction Analysis of Resource Utilization The two nurse operation of 99.99% standing in the 5th Scenario is previewed in saturated working without any assistant. Comparatively, monotonous variations in the stable index of 79.66% (5th Scenario) around 76.97% (4th Scenario）and 63.61% (5th Scenario) vs around 63.49% (4th Scenario） Therefore, both technician and GP are more urgent for requiring remote detection to contribute their participance in intensity or proficiency for improving efficiency. 46 5.2 RECOGNITION OF ENTITIES In static business planning model, Entities as the elementary symbol activate the roles of defining admission, various participating process services and directives, including groups, users, processes, and tasks. Although participate in the formulation of the basic attributes of data and the rules of the process route, the task scope between entities is sometimes vague or the communication channels flexible, resulting in the poor quality at transformation and transportation of data. Having the time delay in the waiting process would lead to the unreasonable allocation of resources alignment and the failure of functions implementations. However, in the introduction process of dynamic data flow, the attributes of the entity are triggered by events of being into a node transformed into participating in the completion of the target of the event around the same or similar purpose activities continuously. That will prompt new requirements for the security of business models. Here is the Entity Roles, Tasks and Responsibility enrolled: Table 5-3--Entity Recognition Entity Roles, Tasks & Responsibility Administration Unit Administration Authorization, Control, Audit Healthcare Center Primary Healthcare (HC) Service Clinic Schedule Patients Appointment and Update Anamnesis Dermatologist Skin Cancer Diagnosis Patient Suspicious or Confirming person or group with dermatoma Nurse Assist doctor to commit schedule appointments and confirm reservations, patient care, etc. Pathologist Research and produce skin pathological section results. Confirm pathology surgery operation Radiologist Radiologist to support CT/MRI/X-Ray Scan, etc. Researcher Researcher existing technical and medical care process issues. Data Engineer/Scientist Summarize process, service, medical science data into producing applicable Algorithm Help Desk Support Patients triage task Oncologist Analyze and Define Skin Cancer Categorization Trigger Service Appointment trigger EMR System to adjust medical resource Image Archive Image resources from outside (HAM 10000) and inside produced by Radiology and Pathology Anamnesis DB Patient cases examination report, files in the database Radiological Report Radiological Examination Result, Report, and Files in DB Pathological Report Pathological Examination Result, Report, and Files in DB Dermoscopy A kind of device to support detecting dermatoma and lesions. Also, assist observers in assessing lesions with little to no pigment 47 As direct engaging into medical care process, screening and then formatting in comparison of a large number of images are going to provide physicians with accurate prediction and pixel positioning above capturing quark factors out the path for algorithm optimization, which is the key to embrace into Robotic Business Process Automation. According to the recognition of entity application and subprocess, the process response assessment with measure guidance matrix is described as follows: Table 5-4--Assessment Measure Matrix Enrollment Response Measure Factor Self-Examination 1. Request for Remote check 2. Understand self-check criteria 3. Confirm Remote/Home self-test 4. Evaluate health status 5. Assess to primary healthcare service or clinic or family doctor 1. Accept Training for preparation 2. Network conditions and instruments are available 3. Barrier-free communication 4. Grasp ABCDE principles or 7 points 5. Time of data entry for patient selfexamination Device/APP 1. Permission to the software application on the client 2. Patients can make an appointment with EMR to read the case 3. Device and equipment configuration records 4. Network security and patient information sharing 1. Install the client software 2. Adequate medical equipment, waiting for the appointment time 3. Failure rate of software equipment 4. Remote detection reservation success rate Dermatology 1. Training on basic usage standards of Dermoscopy 2. Evaluation of techniques for expert detection on skin cancer 3. Hospitals and clinics accept skin cancer screening and detection 4. Description of known skin cancer symptoms and risks 5. Instructions for skin cancer diagnosis 1. Configure remote experts 2. Preliminary diagnostic accuracy 3. Number of tests per unit time 4. Time of medical record data entry 5. Expert review time Radiology 1. Description of basic testing technology process 2. Assess disease types based on images 3. Determine disease progression 4. Participate in disease consultation 5. Case report compilation 6. Coordinate testing technology and non-technical problem solving 7. Disease detection and analysis are presented 1. Waiting time for an appointment in the imaging department 2. Detection time of imaging department 3. Number of tests per unit time 4. Success rate of diagnosis in the radiology department 5. Data entry time of imaging Department 6. Expert review time Biopsy 1. Biopsy operation arrangement 2. Implementation of biopsy surgery 1. Number of tumor biopsy operations 54 6.3 NEW MODULES IN TO-BE New Modules has been redesigned as substitute modules appearing in TO-BE: Table 6-2--New Modules Tasks and Responsibilities (TO-BE) New Modules in TO-BE Task/Responsibilities Define Screening Scope In Primary Health Center, “Define Screening Scope” will be resettled for inclusively targeting types of carcinomas (SCC, BCC, Melanoma, Pediatric Skin Cancer, Rare Skin Cancer, etc.). Estimate Capacity As a resource transformer, real-time evaluation of existing available resources such as hospitals, secondary clinics, Physicians, Nurses, Equipment and Instruments, Dermatologists, Pathologists, Radiologists, algorithm application status, database, and cloud platform situation and EMRS running status. List Task Queues To reset upon different levels of priorities tasks including emergencies and non-emergencies events interruptions. Adjust Physicians Resources Arrange nearby medical treatment according to the activity status of doctors and nurses at different levels Trigger EMRS Alert Service Trigger automatically release and evoke physicians and nurse resource for appointment reservation if in a condition of the not-busy line 6.4 TO-BE APPOINTMENT SERVICE PROCESS The new Appointment Service Process (Figure 3.10) not only configures different levels of EMRS permissions but also can observe the latest reservation status in real-time and feedback to the central management system. That enables to alleviate the problem of over-reliance on central deployment caused by insufficient resource capacity. However, this configuration requires that both secondary clinics and tertiary health care hubs have the same capacity to terminate emergencies in addressing contingency flexibility with increasable risk of technology fund investment. The Appointment Service diagram represented as following: 55 Figure 6-5--Appointment Service Diagram (TO-BE) 56 6.5 EMRS TRIGGER ACTIVATE IN TO-BE DESIGN EMRS Trigger Service is going to simply design flexibilities in which any level of Secondary or Tertiary Health Hub level to reflect the appointment queries. EMRS Trigger. EMRS Trigger Module introduction are designed in the following three processes. Table 6-3--EMRS Trigger Data Flow Introduction Input Output Entity Enrollment Patient Self-Registration Request Self-Exam Result Assessment Patient Patient Self-Exam Estimation Dermatologist Appointment Appointment Notice Anamnesis Dermatologist Request for Radiological Exam Report Permission on reading Radiological Exam Report Radiologist Request for Radiology Exam Agreement/Decline on Radiology Exam Radiological Images and Result Radiological Exam Report Radiologist Advice Radiological Treatment Solutions EMRS Trigger is not only the respond of medical service of resource balance, but also as the main integral service of medical data collector and distributor through automatically resource estimation and reconfigure the elements in format or non-format as request from clinics, radiologists, patients, or dermatologist, etc. Moreover, EMRS Trigger will be also the corresponding analyzer to extract the relative skin cancer images backgrounds from Anamnesis and historical Pathological Archives to transfer forward Healthcare Hub, Oncology Analyzer, and Computation Center. Here is the Radiology Appointment Diagram design in TO-BE), Self-Examination Diagram, Practice Check-up Appointment Process in Diagrams: 57 Figure 6-6--Radiology Appointment Diagram (TO-BE) 58 Figure 6-7--Self-Examination Diagram (TO-BE) 59 Figure 6-8--Practice Check-up Appointment Process Diagram (TO-BE) 60 6.6 SKIN CANCER RESOURCE ASSIGNMENT PROCESS (TO-BE) Optional Remote-Screening operations will be designed in the TO-BE model. One choice is full expert guidance on the operation of the equipment, and the other is semi-independent but requires the patient to consult the doctor for some dermatological problems and technical problems. The Health Center will coordinate Health care resources timely to make early judgments on patients with remote skin cancer screening, and even organize the community about dermoscopy operations introduction and basic screening criteria for skin cancer screening. Over the main responsibility, the health center monitors and reviews existing screening technical and non-technical problems according to the EMRS automated operation, tracking, feedback, and summary until problems are resolved. Figure 6-9--Skin Cancer Resource Assignment Process Diagram (TO-BE) 6.7 DATA REQUIREMENT SERVICE PROCESS Considering the development and utilization of data node technology, the design pattern of the healthcare cloud platform will cause cost differences of the Service layer. With the development of Omni bearing private cloud with high technical demand, long cycle, time cost, and uncountable cost requirements this choice is always off the docket rather than others. Comparatively, Public Cloud has wide availability and high demand satisfaction. Hence, Hybrid cloud models can be determined based on existing resources, technology, and cost considerations. The open skin cancer Image library and open-source algorithm such as HAM10000 can be obtained from the outside without any financial pressure, and medical records of hospitals and cross-department can be shared by private cloud to solve the problem of huge Image memory issue as well. 61 Figure 6-10--Skin Cancer Screening Data Service Process Diagram (TO-BE) 62 6.8 TO-BE DATA SERVICE FRAMEWORK DESIGN In Data Service Framework Design, “Health Center Adapter”, “Process Controller, Query Collector”, “Event Management Service”,” Data Analysis Service” and “Security Management Service” are embedded in the main box of the Data Service processing chip. One example to explain here, the “Event” and “Rules” triggers in the module of “Event Management Service” will enhance concurrent process interaction ability behand of the event response and request status spotlight in the inventory log, as “Process Controller” service is not in deadlock to the demands. Figure 6-11--Data Service Framework Diagram 63 7 TO-BE MODEL ANALYSIS 7.1 SIMULATION ANALYSIS Primary Healthcare level (Line-1) as the resource service commander, appointment trigger capability has arrived to the 98.27% significantly while balancing the GP activities, with the service monitoring operation line 82.86% in the number of instances of appointment request event. To the Secondary level (Line-2) of Healthcare, 48.43% GP work loading without any stressure in average as handling of capacity, however, Tertiary Healthcare (Line 3) would be glad to volume of target appointment events in 81.69%. Figure 7-1--Appointment Trigger Resource Utilization (TO-BE) Figure 7-2--Appointment Trigger Simulation Analysis (TO-BE) 70 At present, face-to-face consultation is the main pathway to diagnose skin cancer, while remote expert dermatoscopy screening provides convenience for early targeted prevention and reliable detection of preventing deep skin cancer lesions for special people. The accuracy of remote screening depends on the expert and the patient's dermatoscopy technique, but a pathology report is no substitute for a definitive diagnosis of skin cancer. The accessibility and equity of personal medical records are activated under the framework of EMRS personal appointment process. Services as dynamic nodes are provided to users with different levels of validity through encryption and decryption security technology. However, in the process of remote personal screening, patients are still restricted by hacker attacks, information loss, interruption of transmission, imperfect data, and many other factors. Patients' participation in remote skin cancer screening is not very intense in that the risk of personal privacy disclosure is not removed. Indeed, the introduction of different options for new remote skin cancer screening raises the potential to change the relative path of screening. The sustainability and scalability of two diagnostic processes are considered, including one is noted as Person-Centered Integrative Diagnosis (PID)，and the other one accounting with Meta-Centred Clinical Diagnosis involvement. Currently, the screening methods available to individuals are mainly face-to-face detection, while in the TO-BE, the full-distance screening or semi-distance mixed skin cancer screening will be persistent to expand the treatment scope coverage through expert guidance and community training. That will be subjected to the accountability on chronic diseases including cancers for Primary and Secondary Healthcare Service non-emergency precontrol against risk-resistance. The current process (As-IS) single mandate with less transparency and permissions, the activities are specified in processing independence without generating a large number of multi-dimensional nodes, yet even did not have cross-level data platform interaction capabilities to release. Therefore, a vertical authorization screening resource balance mechanism is an alternative. According to the simulation data calculation of the model, it can estimate the emergency capacity and risk reduction possibility of the three levels of Primary, Secondary and Tertiary healthcare service under the processing of health and medical resources. The SDM decision-making framework provides a reference for solving the doctor-patient screening and diagnosis plan for chronic cancer diseases. Dermatologist and scientist compared AI image recognition results, classification in stages under different types of skin cancer screening process to obtain the clues to each sub-type of melanoma and other non-melanoma carcinoma. 71 10 CONCLUSIONS The first objective of this dissertation is to figure out the existing main methods and processes of skin cancer screening (AS-IS Model), and then to present a design model for the future (TO-BE Model) as comparative reference. Throughout analyzing the importance of public decision-making, some evidence for source-based screening and remote screening are attempted to track for exploring new skin cancer screening solutions. The models and processes were validated during experts’ interviews and discussions. In the research, we have investigated what factors influence public health decisions - self and public decisions - in determining skin cancer screening. Public decision making provides reference for the possibility of standardized operation and legal guarantee of screening procedures and targets, but many challenges have to be mentioned, whether it is remote skin cancer screening patients' dependence on face-to-face screening and data security vigilance, or digital dermatoscopy, teleradiology, AI technology support for high resolution imaging faces costs and technical risks. In the existing screening model process, patient self-screening and prevention dominate, so that insufficient screening and imaging data for different sub-skin cancer types collected by public decision making are not adequate to provide digital resources for completely enrolling AI analysis detection. At this stage, especially within the scope of the existing resources defined engaged, the level of authority is insufficient, and the event processing capability of different levels is predicted according to the model. In data-driven and task-driven modes, both static and dynamic objects will be referential research into practice. Entity attributes are no longer stared at data input and output request, additional attribute with explorable extensibilities and driving capabilities will also occur as part of the dynamic data flow. Process nodes drive tasks. Although the promotion of cloud technology improves the activity and intelligent radiation range of data, the large-scale expansion of module carrying capacity may not adapt to the configuration of between each region and medical center in crossinglayer institutions. Balancing medical resources and cost control is the direction of sustainable medical center digital services to improve disease control. In this way, a deductive map for the algorithm to predict future changes in suspected tumors that will be provided to the next step for machine learning in possibility. 72 11 LIMITATIONS AND RECOMMENDATIONS FOR FUTURE WORKS The limitations of telemedicine implementation are constricted by legislations, ethics, health insurance policies and the success rate of innovative technologies that patients have to consider during skin cancer screening. For example, in the Intelligent Business Performance Management (IBPM) era, in the event that personal privacy issues in the data bank are protected under the data security protocol agreement, does the portable equipment or device implanted in the use of the sensor affect the credibility of the implementation of a remote skin cancer diagnosis? Decision-makers need to provide more evidence on what models to implement for cloud data services, in this model not yet concluded. Different subtypes of skin cancer require a large number of data parameters and professional evaluation as to which time series analysis model to use for analysis due to the different proportions of dermoscopy. With conjunct risk measure models, especially costsharing estimation models, to identify medical decisions that affect cost changes due to the introduction of medical insurance and medical technology, assuming that process analysis and state operation priorities have been determined. Attending on data transformation and execution effectiveness brought by cost decision-making will enable the public decision center to propose a factor analysis system for the socialized operation of disease control. About intelligent medical decision-making system architecture thinking, how to capture the sensitive factors affecting the screening process of the system and reflect them in the intelligent terminal through the algorithm can assist doctors to participate in the cancer image capture quickly and accurately, which is flexible assignment to practice for various reasons. Skin cancer screening scenario simulation conditions are diverse, and it is hoped that more availability data can be obtained through real simulation experiments to provide dynamic data support for data-cloud-driven decision intelligence. In view of complexity of the skin cancer pathogenesis remains to be further researched through more sophisticated optical device, image recognition algorithm and genetic genealogy data forecast and validation, hope that this model by the dermatology detecting results with different variables and image simulation data, will be expedited to perceive the clues to the rare cancer root in earlier stage of groups. These risk factors need to be investigated until when is introduced into the model to further explore and predict the incidence of skin cancer in a variety of conditions. More than that as so far, it is necessary to verify which quality factors drive the variation of carrying capacity and conceptional deviation fluctuation besides priority and cost under different decision-making and multi-task in parallelized conditions. 73 BIBLIOGRAPHY Adams, W. C. (2015). Conducting Semi-Structured Interviews. In Handbook of Practical Program Evaluation (pp. 492–505). John Wiley & Sons, Inc. https://doi.org/10.1002/9781119171386.ch19 Agarwal Gupta, N., & Kabra, M. (2014). Diagnosis and management of Down syndrome. Indian Journal of Pediatrics, 81(6), 560–567. https://doi.org/10.1007/s12098-013-1249-7 Ahmed jan, N., & Masood, S. (2021). Poroma. In StatPearls. http://www.ncbi.nlm.nih.gov/pubmed/32809744 Andellini, M., Fernandez Riesgo, S., Morolli, F., Ritrovato, M., Cosoli, P., Petruzzellis, S., & Rosso, N. (2019). Experimental application of Business Process Management technology to manage clinical pathways: a pediatric kidney transplantation follow up case. https://doi.org/10.1186/s12911-017-0546-x Andrews, R., Wynn, M., Vallmuur, K., ter Hofstede, A., Bosley, E., Elcock, M., & Rashford, S. (2019). Leveraging Data Quality to Better Prepare for Process Mining: An Approach Illustrated Through Analysing Road Trauma Pre-Hospital Retrieval and Transport Processes in Queensland. International Journal of Environmental Research and Public Health, 16(7), 1138. https://doi.org/10.3390/ijerph16071138 Avhad, G., Ghuge, P., & Jerajani, H. R. (n.d.). Generalized eruptive syringoma. Indian Journal of Dermatology, 60(2), 214. https://doi.org/10.4103/0019-5154.152586 Barbazza, E., Kringos, D., Kruse, I., Klazinga, N. S., & Tello, J. E. (2019). Creating performance intelligence for primary health care strengthening in Europe. BMC Health Services Research. https://doi.org/10.1186/s12913-019-4853-z Becker, J. C., Stang, A., DeCaprio, J. A., Cerroni, L., Lebbé, C., Veness, M., & Nghiem, P. (2017). Merkel cell carcinoma. Nature Reviews Disease Primers, 3(1), 17077. https://doi.org/10.1038/nrdp.2017.77 Bennett, C. C., & Hauser, K. (2013). Artificial intelligence framework for simulating clinical decisionmaking: A Markov decision process approach. Artificial Intelligence in Medicine. https://doi.org/10.1016/j.artmed.2012.12.003 Bernardes Filho, F., Martins, G., Luchi, G. S., Kac, B. K., Nery, J. A. da C., Azulay-Abulafia, L., & Azulay, D. R. (2014). Multiple lesions by vampire bat bites in a patient in Niterói, Brazil - Case report. Anais Brasileiros de Dermatologia, 89(2), 340–343. https://doi.org/10.1590/abd18064841.20142996 Bhambri, A., & Del Rosso, J. Q. (2009). Solitary sclerotic fibroma. The Journal of Clinical and Aesthetic Dermatology, 2(6), 36–38. http://www.ncbi.nlm.nih.gov/pubmed/20729948 Bomhof-Roordink, H., Gärtner, F. R., Stiggelbout, A. M., & Pieterse, A. H. (2019). Key components of shared decision making models: a systematic review. BMJ Open, 9(12), e031763. https://doi.org/10.1136/bmjopen-2019-031763 Buja, A., Toffanin, R., Claus, M., Ricciardi, W., Damiani, G., Baldo, V., & Ebell, M. H. (2018). Developing a new clinical governance framework for chronic diseases in primary care: An umbrella review. BMJ Open. https://doi.org/10.1136/bmjopen-2017-020626 74 Carroll, C. T. (2013). Six Sigma for Powerful Improvement. Productivity Press. https://doi.org/10.1201/b14806 Casteleyn, L., Dumez, B., Van Damme, K., & Anwar, W. A. (2013). Ethics and data protection in human biomarker studies in environmental health. International Journal of Hygiene and Environmental Health. https://doi.org/10.1016/j.ijheh.2013.03.016 Castori, M., Ruggieri, S., Giannetti, L., Annessi, G., & Zambruno, G. (2008). Schöpf-Schulz-Passarge syndrome: further delineation of the phenotype and genetic considerations. Acta DermatoVenereologica, 88(6), 607–612. https://doi.org/10.2340/00015555-0547 Trigeminal trophic syndrome simulating rodent ulcer basal cell carcinoma: a new clinico-dermoscopic approach, 92 Anais Brasileiros de Dermatologia 148 (2017). https://doi.org/10.1590/abd18064841.20175762 Catalano, R., McConnell, W., Forster, P., McFarland, B., & Thornton, D. (2003). Psychiatric emergency services and the system of care. Psychiatric Services. https://doi.org/10.1176/appi.ps.54.3.351 Charifa, A., Volner, K., & Oakley, A. M. (2021). Intraepidermal Carcinoma. In StatPearls. http://www.ncbi.nlm.nih.gov/pubmed/29494081 Cheng Low, P. K., & Ang, S. L. (2013). Confucian Ethics, Governance and Corporate Social Responsibility. International Journal of Business and Management. https://doi.org/10.5539/ijbm.v8n4p30 Cheung, W. L., Patel, R. R., Leonard, A., Firoz, B., & Meehan, S. A. (2012). Amelanotic melanoma: a detailed morphologic analysis with clinicopathologic correlation of 75 cases. Journal of Cutaneous Pathology, 39(1), 33–39. https://doi.org/10.1111/j.1600-0560.2011.01808.x Chimenos-Küstner, E., Pascual, M., Blanco, I., & Finestres, F. (n.d.). Hereditary familial polyposis and Gardner’s syndrome: contribution of the odonto-stomatology examination in its diagnosis and a case description. Medicina Oral, Patologia Oral y Cirugia Bucal, 10(5), 402–409. http://www.ncbi.nlm.nih.gov/pubmed/16264375 Coskey, R. J., & Pinkus, H. (1976). Trichodiscoma. International Journal of Dermatology, 15(8), 600– 601. https://doi.org/10.1111/j.1365-4362.1976.tb04898.x Costello, C. M., Han, M. Y., Severson, K. J., Maly, C. J., Yonan, Y., Nelson, S. A., Swanson, D. L., & Mangold, A. R. (2021). Dermoscopic characteristics of microcystic adnexal carcinoma, desmoplastic trichoepithelioma, and morpheaform basal cell carcinoma. International Journal of Dermatology, 60(3). https://doi.org/10.1111/ijd.15040 Crane, J. S., Rutt, V., & Oakley, A. M. (2021). Birt Hogg Dube Syndrome. In StatPearls. http://www.ncbi.nlm.nih.gov/pubmed/28846260 Cudney, E. A., & Furterer, S. L. (Eds.). (2016). Design for Six Sigma in Product and Service Development. CRC Press. https://doi.org/10.1201/b12090 Cueto, M. (2004). The origins of primary health care and selective primary health care. In American Journal of Public Health. https://doi.org/10.2105/AJPH.94.11.1864 Dahl, N. A., Sheil, A., Knapke, S., & Geller, J. I. (2016). Gardner Fibroma: Clinical and Histopathologic Implications of Germline APC Mutation Association. Journal of Pediatric Hematology/Oncology, 38(5), e154-7. https://doi.org/10.1097/MPH.0000000000000493 75 Damij, N., Damij, T., Grad, J., & Jelenc, F. (2008). A methodology for business process improvement and IS development. Information and Software Technology. https://doi.org/10.1016/j.infsof.2007.11.004 Dimarogonas, D. V., Frazzoli, E., & Johansson, K. H. (2012). Distributed event-triggered control for multi-agent systems. IEEE Transactions on Automatic Control. https://doi.org/10.1109/TAC.2011.2174666 Dinnes, J., Bamber, J., Chuchu, N., Bayliss, S. E., Takwoingi, Y., Davenport, C., Godfrey, K., O’sullivan, C., Matin, R. N., Deeks, J. J., & Williams, H. C. (2018). High-frequency ultrasound for diagnosing skin cancer in adults. In Cochrane Database of Systematic Reviews. https://doi.org/10.1002/14651858.CD013188 Dlamini, Z., Francies, F. Z., Hull, R., & Marima, R. (2020). Artificial intelligence (AI) and big data in cancer and precision oncology. In Computational and Structural Biotechnology Journal. https://doi.org/10.1016/j.csbj.2020.08.019 Eddie G. Montgomery, U. of M. U. C. (2014). TALENT MANAGEMENT VULNERABILITY IN GLOBAL HEALTHCARE VALUE CHAINS: A GENERAL SYSTEMS THEORY PERSPECTIVE. Journal of Business Studies Quarterly, Volume 5,. Elango, B., Paul, K., Kundu, S. K., & Paudel, S. K. (2010). Organizational Ethics, Individual Ethics, and Ethical Intentions in International Decision-Making. Journal of Business Ethics. https://doi.org/10.1007/s10551-010-0524-z Eleanor Bird, Fox-Skelly, J., Jenner, N., Larbey, R., Weitkamp, E., & Winfield, A. (2020). The ethics of artificial intelligence Issues and initiatives. In STOA | Panel for the Future of Science and Technology AUTHORS. ENCR. (2015). Malignant Melanoma of the Skin (MM) Factsheet. In European Network of Cancer Registries. Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature. https://doi.org/10.1038/nature21056 Fabiano, N. (2019). Robotics, intelligent systems, ethics and data protection. PervasiveHealth: Pervasive Computing Technologies for Healthcare. https://doi.org/10.1145/3309772.3309787 Ferreira, G. S. A., Silva, U. R., Costa, A. L., & Pádua, S. I. D. de D. (2018). The promotion of BPM and lean in the health sector: main results. Business Process Management Journal, 24(2), 400–424. https://doi.org/10.1108/BPMJ-06-2016-0115 Freet, D., Agrawal, R., John, S., & Walker, J. J. (2015). Cloud forensics challenges from a service model standpoint: IaaS, PaaS and SaaS. 7th International ACM Conference on Management of Computational and CollEctive Intelligence in Digital EcoSystems, MEDES 2015. https://doi.org/10.1145/2857218.2857253 Gay, J. T., Troxell, T., & Gross, G. P. (2021). Muir-Torre Syndrome. In StatPearls. https://doi.org/30020643 Genova, K., & Guliashki, V. (2011). Linear integer programming methods and approaches - A survey. Cybernetics and Information Technologies. 76 Gershenwald, J. E., Scolyer, R. A., Hess, K. R., Sondak, V. K., Long, G. V., Ross, M. I., Lazar, A. J., Faries, M. B., Kirkwood, J. M., McArthur, G. A., Haydu, L. E., Eggermont, A. M. M., Flaherty, K. T., Balch, C. M., & Thompson, J. F. (2017). Melanoma staging: Evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA: A Cancer Journal for Clinicians, 67(6), 472–492. https://doi.org/10.3322/caac.21409 Giavina-Bianchi, M., Giavina-Bianchi, P., Sotto, M. N., Rodig, S., Mihm, M., Festa Neto, C., Duncan, L. M., & Kalil, J. (2020). Nodular primary cutaneous melanoma is associated with PD-L1 expression. European Journal of Dermatology : EJD, 30(4), 352–357. https://doi.org/10.1684/ejd.2020.3846 Globle Coalition. (2021). 2O2O Melanoma Skin Cancer Report Stemming the global epidemic. https://melanomapatients.org.au/wp-content/uploads/2020/04/2020-campaign-report-GCversion-MPA_1.pdf Goeteyn, M., Geerts, M. L., Kint, A., & De Weert, J. (1994). The Bazex-Dupré-Christol syndrome. Archives of Dermatology, 130(3), 337–342. http://www.ncbi.nlm.nih.gov/pubmed/8129412 Gong, H.-Z., Zheng, H.-Y., & Li, J. (2019). Amelanotic melanoma. Melanoma Research, 29(3), 221–230. https://doi.org/10.1097/CMR.0000000000000571 Goydos, J. S., & Shoen, S. L. (2016). Acral Lentiginous Melanoma. Cancer Treatment and Research, 167, 321–329. https://doi.org/10.1007/978-3-319-22539-5_14 Groups, I. T. M. B.-. (n.d.). ISO GUIDE 73:2009 (1st ed.). https://www.iso.org/standard/44651.html Guidelines for Hazard Evaluation Procedures, 3rd Edition. (n.d.). Gupta, N., Sunwoo, B. Y., & Kotloff, R. M. (2016). Birt-Hogg-Dubé Syndrome. Clinics in Chest Medicine, 37(3), 475–486. https://doi.org/10.1016/j.ccm.2016.04.010 Hafsi, W., Badri, T., & Shah, F. (2021). Apocrine Hidrocystoma. In StatPearls. http://www.ncbi.nlm.nih.gov/pubmed/28846245 Hanssen, A. M., & Fryns, J. P. (1995). Cowden syndrome. Journal of Medical Genetics, 32(2), 117–119. https://doi.org/10.1136/jmg.32.2.117 Hashim, P. W., Chen, T., Rigel, D., Bhatia, N., & Kircik, L. H. (2019). Actinic Keratosis: Current Therapies and Insights Into New Treatments. Journal of Drugs in Dermatology : JDD, 18(5), s161-166. http://www.ncbi.nlm.nih.gov/pubmed/31141862 Helfand, M., Mahon, S. M., Eden, K. B., Frame, P. S., & Orleans, C. T. (2001). Screening for skin cancer. American Journal of Preventive Medicine. https://doi.org/10.1016/S0749-3797(01)00258-6 Herrera Sánchez, M., Suánez Fernández, R., del Cerro Heredero, M., Rueda Gómez-Calcerrada, M., Vigaray Conde, J., & Sánchez Yus, E. (1998). Sclerotic fibroma. Dermatology (Basel, Switzerland), 196(4), 429–430. https://doi.org/10.1159/000017939 Hoerbst, A., Hackl, W. O., Blomer, R., & Ammenwerth, E. (2011). The status of IT service management in health care - ITIL ® in selected European countries. BMC Medical Informatics and Decision Making. https://doi.org/10.1186/1472-6947-11-76 77 Hopkins, C., Moyer, A., & Huttenbach, Y. T. (2020). Spiradenoma With Adenoid Cystic Carcinoma-Like Features. The American Journal of Dermatopathology, 42(6), 442–445. https://doi.org/10.1097/DAD.0000000000001594 Hossain, M. S., & Muhammad, G. (2016). Cloud-assisted Industrial Internet of Things (IIoT) - Enabled framework for health monitoring. Computer Networks. https://doi.org/10.1016/j.comnet.2016.01.009 I-scoop. (2017). Industry 4.0: the fourth industrial revolution – guide to Industrie 4.0. I-Scoop. Iacobelli, J., Harvey, N. T., & Wood, B. A. (2017). Sebaceous lesions of the skin. Pathology, 49(7), 688– 697. https://doi.org/10.1016/j.pathol.2017.08.012 Iyer, V., Ganti, B., Hima Vyshnavi, A. M., Krishnan Namboori, P. K., & Iyer, S. (2020). Hybrid quantum computing based early detection of skin cancer. Journal of Interdisciplinary Mathematics. https://doi.org/10.1080/09720502.2020.1731948 Jazdi, N. (2014). Cyber physical systems in the context of Industry 4.0. Proceedings of 2014 IEEE International Conference on Automation, Quality and Testing, Robotics, AQTR 2014. https://doi.org/10.1109/AQTR.2014.6857843 Jetzek, T., Avital, M., & Bjørn-Andersen, N. (2019). The sustainable value of open government data. Journal of the Association for Information Systems. https://doi.org/10.17705/1jais.00549 Ji, X., Yin, Z., Zhang, Y., Wang, M., Zhang, X., Zhang, C., & Wang, D. (2021). Real-time robust forecasting-aided state estimation of power system based on data-driven models. International Journal of Electrical Power and Energy Systems. https://doi.org/10.1016/j.ijepes.2020.106412 Jordão, C., de Magalhães, T. C., Cuzzi, T., & Ramos-e-Silva, M. (2015). Cylindroma: an update. International Journal of Dermatology, 54(3), 275–278. https://doi.org/10.1111/ijd.12600 Jovanovic, P., Mihajlovic, M., Djordjevic-Jocic, J., Vlajkovic, S., Cekic, S., & Stefanovic, V. (2013). Ocular melanoma: an overview of the current status. International Journal of Clinical and Experimental Pathology, 6(7), 1230–1244. https://doi.org/10.1007/s11912-012-0244-x Kazakov, D. V. (2016). Brooke-Spiegler Syndrome and Phenotypic Variants: An Update. Head and Neck Pathology, 10(2), 125–130. https://doi.org/10.1007/s12105-016-0705-x Kermany, D. S., Goldbaum, M., Cai, W., Valentim, C. C. S., Liang, H., Baxter, S. L., McKeown, A., Yang, G., Wu, X., Yan, F., Dong, J., Prasadha, M. K., Pei, J., Ting, M., Zhu, J., Li, C., Hewett, S., Dong, J., Ziyar, I., … Zhang, K. (2018). Identifying Medical Diagnoses and Treatable Diseases by ImageBased Deep Learning. Cell. https://doi.org/10.1016/j.cell.2018.02.010 Khan, F. A., Ali, A., Abbas, H., & Haldar, N. A. H. (2014). A cloud-based healthcare framework for security and patients’ data privacy using wireless body area networks. Procedia Computer Science. https://doi.org/10.1016/j.procs.2014.07.058 Kibbi, N., Worley, B., Owen, J. L., Kelm, R. C., Bichakjian, C. K., Chandra, S., Demirci, H., Kim, J., Nehal, K. S., Thomas, J. R., Poon, E., & Alam, M. (2020). Sebaceous carcinoma: controversies and their evidence for clinical practice. Archives of Dermatological Research, 312(1), 25–31. https://doi.org/10.1007/s00403-019-01971-4 Kim, K. S., & Weinberg, P. E. (1981). Dermoid tumor. Surgical Neurology, 15(5), 375–376. https://doi.org/10.1016/0090-3019(81)90175-0 78 Landry, M., & Winkelmann, R. K. (1972). Multiple clear-cell acanthoma and ichthyosis. Archives of Dermatology, 105(3), 371–383. http://www.ncbi.nlm.nih.gov/pubmed/4334961 Laver, N. V, McLaughlin, M. E., & Duker, J. S. (2010). Ocular melanoma. Archives of Pathology & Laboratory Medicine, 134(12), 1778–1784. https://doi.org/10.5858/2009-0441-RAR.1 Lee, J., Bagheri, B., & Kao, H. A. (2015). A Cyber-Physical Systems architecture for Industry 4.0-based manufacturing systems. Manufacturing Letters. https://doi.org/10.1016/j.mfglet.2014.12.001 Leshob, A., Bourgouin, A., & Renard, L. (2018). Towards a Process Analysis Approach to Adopt Robotic Process Automation. Proceedings - 2018 IEEE 15th International Conference on eBusiness Engineering, ICEBE 2018. https://doi.org/10.1109/ICEBE.2018.00018 Liang, J. (2019). Qualitative research methods: collecting evidence, crafting analysis, communicating impact (2 nd Edition). Communication Research and Practice, 5(4), 408–409. https://doi.org/10.1080/22041451.2019.1688620 Longo, C., & Pellacani, G. (2016). Melanomas. Dermatologic Clinics, 34(4), 411–419. https://doi.org/10.1016/j.det.2016.05.004 LONGO, C., PELLACANI, G., TOMASI, A., MANDEL, V. D., & PONTI, G. (2016). Fibroepithelioma of Pinkus: Solitary tumor or sign of a complex gastrointestinal syndrome. Molecular and Clinical Oncology, 4(5), 797–800. https://doi.org/10.3892/mco.2016.794 Low, L. L., Ab Rahim, F. I., Johari, M. Z., Abdullah, Z., Abdul Aziz, S. H., Suhaimi, N. A., Jaafar, N., Mohd Hanafiah, A. N., Kong, Y. L., Mahmud, S. H., Zulkepli, M. Z., Perialathan, K., Muharam, N., Zainudin, N. H., Mohd Zin, Z., Mohd Roslan, N., Aris, T., & Murad, S. (2019). Assessing receptiveness to change among primary healthcare providers by adopting the consolidated framework for implementation research (CFIR). BMC Health Services Research. https://doi.org/10.1186/s12913-019-4312-x Lucas, A. S., Chung, E., Marchetti, M. A., & Marghoob, A. A. (2016). A guide for dermatology nurses to assist in the early detection of skin cancer. Journal of Nursing Education and Practice. https://doi.org/10.5430/jnep.v6n10p71 Lunenburg, F. C. (2012). Organizational Structure: Mintzberg’s Framework. International Journal for Scholarly, Academic, Intellectual Diversity. Maguire-Eisen, M. (2013). Skin cancer: A growing health problem for children. Seminars in Oncology Nursing. https://doi.org/10.1016/j.soncn.2013.06.006 Mahbod, A., Schaefer, G., Wang, C., Ecker, R., & Ellinge, I. (2019). Skin Lesion Classification Using Hybrid Deep Neural Networks. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. https://doi.org/10.1109/ICASSP.2019.8683352 Maher, E. E., & Vidal, C. I. (2015). Trichilemmoma. Cutis, 96(2), 81, 104–106. http://www.ncbi.nlm.nih.gov/pubmed/26367753 Marhavilas, P., Koulouriotis, D., & Mitrakas, C. (2014). FAULT AND EVENT-TREE TECHNIQUES IN OCCUPATIONAL HEALTH-SAFETY SYSTEMS - PART I: INTEGRATED RISK-EVALUATION SCHEME. Environmental Engineering and Management Journal, 13(8), 2097–2108. https://doi.org/10.30638/eemj.2014.232 79 Martin, N., Martinez-Millana, A., Valdivieso, B., & Fernández-Llatas, C. (2019a). Interactive Data Cleaning for Process Mining: A Case Study of an Outpatient Clinic’s Appointment System. Undefined, 362 LNBIP, 532–544. https://doi.org/10.1007/978-3-030-37453-2_43 Martin, N., Martinez-Millana, A., Valdivieso, B., & Fernández-Llatas, C. (2019b). Interactive Data Cleaning for Process Mining: A Case Study of an Outpatient Clinic’s Appointment System (pp. 532–544). https://doi.org/10.1007/978-3-030-37453-2_43 MEGAHED. (1999). Polypoid basal cell carcinoma: a new clinicopathological variant. British Journal of Dermatology, 140(4), 701–703. https://doi.org/10.1046/j.1365-2133.1999.02774.x Moe, M. M. (2019). Comparative Study of Test-Driven Development TDD, Behavior-Driven Development BDD and Acceptance Test–Driven Development ATDD. International Journal of Trend in Scientific Research and Development, Volume-3(Issue-4), 231–234. https://doi.org/10.31142/ijtsrd23698 Mohiuddin, W., Laun, J., & Cruse, W. (2018). Brooke-Spiegler Syndrome. Eplasty, 18, ic14. http://www.ncbi.nlm.nih.gov/pubmed/30093932 Muthukumaran, D., Umapathy, K., & Omkumar, S. (2021). Health Cloud—Health Care as a Service. In Lecture Notes in Networks and Systems. https://doi.org/10.1007/978-981-16-0666-3_39 Navarrete-Dechent, C., Bajaj, S., Marghoob, A., Gonzalez, S., & Munoz, D. (2016). Multiple familial trichoepithelioma: confirmation via dermoscop. Dermatology Practical & Conceptual, 6(3). https://doi.org/10.5826/dpc.0603a10 Nicolson, N. G., & Han, D. (2019). Desmoplastic melanoma. Journal of Surgical Oncology, 119(2), 208– 215. https://doi.org/10.1002/jso.25317 North, J. P., Lo, J., & Landers, M. (2012). Poromatosis in pregnancy: a case of 8 eruptive poromas in the third trimester. Cutis, 89(2), 81–83. http://www.ncbi.nlm.nih.gov/pubmed/22474731 Occidental, M., Shapiro, R., & Jour, G. (2020). Lentigo maligna melanoma in situ with neurotropism. Journal of Cutaneous Pathology, 47(12), 1155–1158. https://doi.org/10.1111/cup.13778 Olla, D., & Neumeister, M. W. (2021). Mucosal Melanoma. Clinics in Plastic Surgery, 48(4), 707–711. https://doi.org/10.1016/j.cps.2021.05.010 Owen, J. L., Kibbi, N., Worley, B., Kelm, R. C., Wang, J. V, Barker, C. A., Behshad, R., Bichakjian, C. K., Bolotin, D., Bordeaux, J. S., Bradshaw, S. H., Cartee, T. V, Chandra, S., Cho, N. L., Choi, J. N., Council, M. L., Demirci, H., Eisen, D. B., Esmaeli, B., … Alam, M. (2019). Sebaceous carcinoma: evidence-based clinical practice guidelines. The Lancet Oncology, 20(12), e699–e714. https://doi.org/10.1016/S1470-2045(19)30673-4 Parekh, V., & Seykora, J. T. (2017). Cutaneous Squamous Cell Carcinoma. Clinics in Laboratory Medicine, 37(3), 503–525. https://doi.org/10.1016/j.cll.2017.06.003 Parren, L. J. M. T., & Frank, J. (2011). Hereditary tumour syndromes featuring basal cell carcinomas. British Journal of Dermatology, 165(1), 30–34. https://doi.org/10.1111/j.13652133.2011.10334.x PDQ Adult Treatment Editorial Board. (2020). Skin Cancer Treatment (PDQ®). In National Cancer Institute (US). 86 THANK YOU!