Bradykinesia detection in Parkinson’s disease using smartwatches’ inertial sensors and deep learning methods

Citation: Sigcha, L.; Domínguez, B.; Borzì, L.; Costa, N.; Costa, S.; Arezes, P.; López, J.M.; De Arcas, G.; Pavón, I. Bradykinesia Detection in Parkinson’s Disease Using Smartwatches’ Inertial Sensors and Deep Learning Methods. Electronics 2022,11, 3879. https://doi.org/ 10.3390/electronics11233879 Academic Editor: Nicola Francesco Lopomo Received: 31 October 2022 Accepted: 21 November 2022 Published: 24 November 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). electronics Article Bradykinesia Detection in Parkinson’s Disease Using Smartwatches’ Inertial Sensors and Deep Learning Methods Luis Sigcha 1,2 , Beatriz Domínguez 1, Luigi Borzì 3, Nélson Costa 2, Susana Costa 2, Pedro Arezes 2, Juan Manuel López 1, Guillermo De Arcas 1and Ignacio Pavón 1,* 1 Instrumentation and Applied Acoustics Research Group (I2A2), ETSI Industriales, Universidad Politécnica de Madrid, Campus Sur UPM, Ctra. Valencia, Km 7, 28031 Madrid, Spain 2ALGORITMI Research Center, School of Engineering, University of Minho, 4800-058 Guimarães, Portugal 3Department of Control and Computer Engineering, Politecnico di Torino, 10129 Turin, Italy *Correspondence: [email protected]; Tel.: +34-91-067-7222 Abstract: Bradykinesia is the defining motor symptom of Parkinson’s disease (PD) and is reflected as a progressive reduction in speed and range of motion. The evaluation of bradykinesia severity is important for assessing disease progression, daily motor fluctuations, and therapy response. However, the clinical evaluation of PD motor signs is affected by subjectivity, leading to intraand inter-rater variability. Moreover, the clinical assessment is performed a few times a year during prescheduled follow-up visits. To overcome these limitations, objective and unobtrusive methods based on wearable motion sensors and machine learning (ML) have been proposed, providing promising results. In this study, the combination of inertial sensors embedded in consumer smartwatches and different ML models is exploited to detect bradykinesia in the upper extremities and evaluate its severity. Six PD subjects and seven age-matched healthy controls were equipped with a consumer smartwatch and asked to perform a set of motor exercises for at least 6 weeks. Different feature sets, data representations, data augmentation methods, and ML models were implemented and combined. Data recorded from smartwatches’ motion sensors, properly augmented and fed to a combination of Convolutional Neural Network and Random Forest model, provided the best results, with an accuracy of 0.86 and an area under the curve (AUC) of 0.94. Results suggest that the combination of consumer smartwatches and ML classification methods represents an unobtrusive solution for the detection of bradykinesia and the evaluation of its severity. Keywords: Parkinson’s disease; bradykinesia; wearables; inertial sensors; artificial intelligence; deep learning 1. Introduction Parkinson’s disease (PD) is one of the most common neurodegenerative diseases worldwide [ 1 ], affecting millions of people and impacting their quality of life (QoL) [ 2 ]. PD is a progressive disease with a slow and variable evolution. In the early stages, the symptoms are weak, and they increase in intensity as the disease progresses [ 3 ]. PD involves both motor and non-motor symptoms, with some of the latter (i.e., speech impairment and sleep disorders) manifesting up to 20 years before the clinical diagnosis [ 4 ]. Being primarily a movement disorder, several motor signs are associated with PD, including bradykinesia, tremor, and rigidity. As the disease progresses, postural instability and freezing of gait (FOG) manifest, increasing the risk of falls [ 5 ] and contributing to decreased mobility [ 6 ]. As the main biochemical abnormality in PD is dopamine deficiency [ 7 ], current treatments are mainly based on dopamine replacement, with Levodopa representing the most effective drug treatment for PD [ 8 , 9 ]. However, current treatments do not prevent disease progression, their effectiveness decreases with disease progression [ 10 ], and long-term therapy frequently leads to severe side effects [ 11 ]. Moreover, as the disease progresses and Electronics 2022,11, 3879. https://doi.org/10.3390/electronics11233879 https://www.mdpi.com/journal/electronics Electronics 2022,11, 3879 2 of 19 drug therapy is administrated, patients may experience fluctuations in the state of their motor system, between the so-called ON state, where symptoms are under control and the patient can move fluidly, and an OFF state, in which a lack of dopamine predominates and symptoms reappear when the effect of the medication vanishes. Bradykinesia represents one of the earliest motor signs of PD, and it is one of the main aspects that specialists try to quantify to diagnose PD and optimize therapy. It is defined by the slowness and decrease in the amplitude or speed of movement in a body part [ 12 ]. Akinesia and hypokinesia refer respectively to poor spontaneous movements (i.e., in facial expression) or associated movement (i.e., arm swing during walking) and the low amplitude of movement [ 2 , 13 ]. Bradykinesia can vary throughout the day and its severity also vary depending on the timing and amount of the last medication. In addition, the symptom’s severity also depends on the patient’s emotional state and environment [ 2 ]. Bradykinesia is one of the key signs in the evaluation of PD, it is directly related to dopamine deficiency [ 14 ], and it shows an exceptional response to treatment [ 15 ]. Thus, objectively quantifying this symptom would provide relevant information for treatment adjustments and early diagnosis. Following the movement disorder society revised version of the unified Parkinson’s disease rating scale (MDS-UPDRS), neurologists assess bradykinesia severity through the execution of rapid, repetitive, alternating hand and heel movements, and they observe the amplitude and slowness of the movement [ 16 ]. However, the assessment is performed sporadically, during brief follow-up visits, often without considering the effect of medication. Moreover, intra-rater and inter-rater variability affect the evaluation of the patient’s motor performance [17,18]. Subjectivity and late diagnosis highlight the need for new, more objective methodologies allowing the early diagnosis of the disease, the continuous monitoring of its evolution, and the evaluation of the response to therapy [ 19 ]. In this context, digital technologies have demonstrated their potential to change the disease paradigm, providing unobtrusive yet efficient solutions for the diagnosis, assessment, monitoring, and treatment planning of PD patients [ 20 , 21 ]. Indeed, an objective measure of PD symptoms can help improve disease management and accelerate the development of new therapies [ 15 ]. Wearable sensors benefit from the current technological advances to provide lightweight, portable, easy-to-use, inexpensive devices which can provide accurate measurements of physical variables [ 22 ]. Wearable motion sensors and ML methods have been widely used for objectively and rigorously assessing motor symptoms, motor fluctuations, and other complications that are relevant to adjust treatment and remote assistance [15,23–26]. In this context, this paper evaluates the potential of consumer smartwatches for estimating bradykinesia severity in PD. To this end, upper limb motion data were recorded from a triaxial accelerometer and triaxial gyroscope placed on the patient’s wrist. Then, signal processing, data augmentation, data transformation, and different ML and DL classification models were implemented to predict the bradykinesia severity following the standards of the MDS-UPDRS scale. The main contributions of this work are summarized as follows: • This study evaluates the potential of accelerometer and gyroscope sensors embedded in commodity smartwatches to detect bradykinesia severity using a set of standardized exercises. This approach can present an unobtrusive solution for bradykinesia monitoring in ambulatory and non-supervised environments using low-cost devices instead of using proprietary monitoring devices or sensors. • Different feature extraction methodologies proposed in the related literature are reproduced and evaluated with the data collected using commodity smartwatches. This task is performed to compare the predictive power of approaches based on ML and DL. In addition, the potential of different data representations and data augmentation techniques is evaluated with the aim of improving the performance of the systems for automatic bradykinesia severity scoring. • This work also introduces the use of convolutional neural networks (CNN) with patch input (implemented with 1D-convolutional layers) for automatic temporal Electronics 2022,11, 3879 3 of 19 window contextualization. The patch input strategy is proposed as a mechanism to automatically split and project the data of a single (multi-channel) sliding window into another dimension that can be exploited by classification algorithms. Additionally, the proposed approach is evaluated using an end-to-end neural network, and in combination with a Random Forest (RF) classifier located at the top of the neural network. • Finally, a methodology for the aggregation of a set of predictions (severity ratings) obtained from the classifiers during a single clinical visit is proposed and evaluated. This methodology is carried out with the aim of improving the outcomes of the bradykinesia assessment by providing a single severity indicator of the motor function of the upper limbs. The rest of this paper is organized as follows: An overview of the research studies focusing on bradykinesia detection using wearable sensors is provided in Section 2. Section 3 describes the data set used in this study, the implemented signal processing and ML methods, and the performance evaluation procedure. Results are reported in Section 4 and discussed in Section 5, together with conclusions. 2. Related Work The quantification of bradykinesia using wearable technologies has been widely explored in the last several decades. Besides commercial solutions, such as Kinesia ® (Great Lakes NeuroTechnologies Inc., Cleveland, OH, USA) [ 27 ] and PKG ® (Global Kinetics Pty Ltd., Melbourne, Australia) [ 28 ], several research studies focused on the detection of bradykinesia by characterizing the movement of patients. In [ 29 ], 50 PD patients were monitored to quantify bradykinesia and hypokinesia. Two accelerometers on the wrist were used for data collection, obtaining sensitivities of 60–71% and specificities of 66–76%. In [ 30 ], seven gyroscopes and two accelerometers were placed on the forearms, shins, and trunk for diagnosing the presence or absence of bradykinesia, tremor, body posture, and gait parameters, obtaining a Pearson correlation coefficient r of 0.71 with the UPDRS scale. In [ 31 ], a combination of a flexible sensor placed on the hand (triaxial accelerometer and gyroscope) and a consumer smartwatch (triaxial accelerometer) was employed to monitor 13 PD subjects. By using an RF algorithm, the authors achieved an AUC of 0.65 in a multiclass classification (MDS-UPDRS). In [ 32 ], an inertial measurement unit (IMU) wristband with an accelerometer was used to monitor 31 PD patients and 50 healthy controls. The authors proposed a methodology to extract bradykinesia digital biomarkers, providing a strong correlation (Pearson r = 0.67) between hand motion measurements and the MDS-UPDRS scoring. The leg agility task (MDS-UPDRS item 3.8) was addressed in different studies for the quantification of bradykinesia. In [ 33 , 34 ], 34 and 24 subjects were enrolled, respectively. Three IMUs were mounted on the patient’s chest and each thigh. Timeand frequencydomain features were extracted and selected to feed classification algorithms, i.e., Support Vector Machine (SVM) and k-Nearest Neighbors (kNN). Bradykinesia severity (UPDRS score) was estimated with an accuracy of 43% in both studies. In [ 35 ], 19 subjects with PD were monitored with ankle-mounted IMUs for leg agility evaluation and treatment response. Timeand frequency-domain features were computed to feed different classifiers, i.e., SVM, Decision Tree, and Logistic Regression. Pearson correlation with the UPDRS bradykinesia score was found to be r = 0.83. Finally, in [ 17 ], smartphones’ sensors and ML were used to detect bradykinesia using leg agility exercises, achieving an accuracy of 77.7% in a multi-class classification using the UPDRS scale. In recent years, the research community has explored the use of DL techniques for the automatic analysis of motor symptoms. DL methods make it possible to process the recorded inertial signals without the need for additional processing techniques, reducing the effort in the design and selection of discriminative feature sets [ 36 ]. However, despite the advantages of technology in PD, the application of these techniques requires a high amount of quality data and high computational processing power [37]. Electronics 2022,11, 3879 4 of 19 Relevant works using DL methods and wearable technology to assess bradykinesia have been proposed in [ 38 , 39 ], where CNN and sensors placed on the upper limbs have been employed. The results of these works indicate that they can outperform (shallow) ML approaches achieving an accuracy of 90.9% [ 38 ], and an AUC of 0.926 [ 39 ]. In [ 40 ], 30 PD patients were monitored during different activities using a single accelerometer on the wrist. CNN was used to process raw data and predict bradykinesia severity, achieving an accuracy of 0.67, sensitivity of 0.65, and specificity of 0.89. In [41], six flexible wearable sensors were used for recording data from 20 individuals with PD throughout multiple clinical assessments. Raw inertial data were input to a CNN algorithm, providing an AUC of 0.77. 3. Materials and Methods In this section, the methodology developed to obtain different bradykinesia detection methods is described. Section 3.1 describes the data used in this study, including information regarding subjects’ characteristics, experimental procedures, and clinical assessment of bradykinesia. The preprocessing procedures, including filtering, feature extraction, data transformation, and data augmentation, are reported in Section 3.2. Section 3.3 describes the ML and DL classification algorithms employed in the present work, together with their implementation details. Finally, details regarding the performance evaluation methods are provided in Section 3.5. 3.1. Bradykinesia Dataset The dataset employed in this study was collected using the Monipar application [ 42 ]. Monipar proposes a system based on wearable technology and artificial intelligence (AI) for monitoring motor activity in PD. The system consists of a mobile app that guides the user in performing 8 exercises of the MDS-UPDRS scale and a wearable module that records the subject’s movement using the triaxial accelerometer and gyroscope embedded in a consumer smartwatch. Specifically, tasks consisted of a series of 8 exercises belonging to the MDS-UPDRS scale part III, concerning the examination of the motor aspects [ 16 ]. The selected exercises include rest tremor amplitude, postural tremor of the hands, movement of the hands to the chest, finger tapping, hand movements, pronation–supination movements of the hands, arising from a chair, and gait. The duration of the entire procedure is approximately 7 min. 3.1.1. Data Acquisition Data were recorded from 6 subjects (3 females and 3 males, 64.2 ± 8.2 years) diagnosed with PD in the early stages of the disease, according to the Hoehn and Yarn scale [ 43 ] ( H&Y = 1 in all subjects) and from 7 healthy control subjects (4 females and 3 males, 64.0 ±5.4 years) . The data collection process was carried out for 8 and 9 weeks, respectively, using the Monipar application. Each week, subjects performed the pre-defined motor tasks in a controlled environment. A total of 105 weekly sessions were collected during the experimentation (46 sessions for PD; 59 sessions for healthy controls). These data correspond to more than 13 h of movement data collected with a triaxial accelerometer and a triaxial gyroscope. However, only relevant data related to the movement of the upper limbs, i.e., that recorded during finger tapping, hand movement, and pronation–supination movement of the hands, were analyzed in this study. The data from the three hand exercises correspond to 80 min (10% of the entire data set) of movement data collected by each of the inertial sensors. The smartwatch employed for data collection was available on the market in 2019. This device employs Android Wear operating system and an internal memory of 4 GB (2 GB of free space). The device has a calibrated triaxial accelerometer with a maximum amplitude set to ±2 g, and triaxial gyroscope with a measurement range set to ±2000 dps. The smartwatch was placed on the wrist of the most affected side, according to the clinical indication of the physician attending to the patient and the dominant hand of Electronics 2022,11, 3879 5 of 19 healthy controls. Data were recorded using the accelerometer and gyroscope embedded in the smartwatch, with a sampling frequency of 50 Hz. Such a frequency is appropriate for human motion analysis, as the frequency content generated by common human movements lies in the 0–20 Hz band [ 44 ]. Figure 1summarizes the data collection process carried out using the Monipar app. Figure 1. Data collection methodology to detect bradykinesia using smartwatches and MDSUPDRS exercises. 3.1.2. Data Labeling Training supervised ML and DL methods to automatically detect motor symptoms requires data to be labeled by expert clinicians, who recognize symptoms and evaluate their severity. The labeling of the Monipar data was performed by a trained expert neurologist, who reviewed the videos of the weekly trials performed by the subjects. For each motor task, the clinician identified the presence of bradykinesia and evaluated its severity. According to the MDS-UPDRS guidelines, a score between 0 (no bradykinesia) and 4 (severe bradykinesia) was assigned to each task. To assign a single severity value to the data of each weekly assessment, the sub-scores of the three upper limbs exercises were averaged and rounded, and finally used as the reference metric. The distribution of the severity of bradykinesia in the group of PD patients and control subjects is reported in Figure 2. It can be observed that the recorded bradykinesia severity corresponds to four UPDRS ratings, including normal (0), slight (1), mild (2), and moderate (3). As evident from Figure 2, the data distribution is unbalanced, with more than 57% of the data corresponding to the UPDRS 0 severity (no bradykinesia). Moreover, movements belonging to the class UPDRS 4 (severe bradykinesia) are not represented. This is likely due to the intrinsic composition of the PD sample, which encompasses patients in the early stages of the disease. Figure 2. Distribution of the severity of bradykinesia in the dataset. Electronics 2022,11, 3879 6 of 19 3.2. Signal Preprocessing In order to prepare data for the subsequent classification step, some preprocessing procedures were performed. First, data were filtered and segmented (Section 3.2.1); then, different data transformation methods were applied (Section 3.2.2) to provide the input for ML and DL algorithms; finally, data augmentation was exploited to increase the data set size and provide a more balanced distribution of data (Section 3.2.3). 3.2.1. Filtering and Segmentation The low-frequency components of the sensor readings are related to postural changes (gross movements), while the high-frequency components reflect the actual accelerations of the body segments, associated with rapid movements [ 45 ]. To remove the gravity effect and the noise produced by trembling or shaking, inertial data were filtered using a fourth-order zero-lag Butterworth band-pass infinite impulse response (IIR) digital filter, with cut-off frequencies of 0.25 Hz and 3.5 Hz. The advantage of the Butterworth-type filter is that it allows a nearly constant gain in the passband. Then, inertial signals were segmented using non-overlapping sliding windows of 5.12 s (i.e., 256 samples). Figure 3shows a segment of the raw gyroscope signal (Figure 3a) and the corresponding filtered signal (Figure 3b). (a) Original signal (b) Filtered signal Figure 3. Filtering applied to the gyroscope signals. ( a ) sample of the original signal corresponding to exercise 4 (finger tapping); ( b ) gyroscope signal after applying a 0.25–3.5 Hz fourth-order Butterworth band pass filter. 3.2.2. Feature Extraction Classic ML models such as RF require features to be extracted from recorded signals. Two feature sets proposed in the reference literature were reproduced in this study, belonging to both the time and frequency domains. This was carried out to establish a reference model for comparison with the proposed methods. The two sets of features [ 31 , 46 ] include a total number of 74 and 43 features, respectively. As far as the input data for DL models are concerned, two different data representations were employed. The first consists of using the inertial readings, normalized in the range from − 1 to 1. The second was created as follows. Every single window obtained from the segmentation process was divided into two consecutive windows of 2.56 s (i.e., 128 samples). Then, the signals’ fast Fourier transform (FFT) was computed for both windows and used as an input feature set. This feature extraction method is based on contextual windows Electronics 2022,11, 3879 7 of 19 and will be referred to in the rest of the paper as Contextual FFT. The contextualization of adjacent FFT windows is based on methods proposed in the reference literature to improve the performance in FOG detection using accelerometers [47–49]. A summary of the feature set employed in this study is shown in Table 1. Table 1. Summary of the data representations. FFT: fast Fourier transform. Feature Set Number of Features Description of the Features Shawen et al. [ 31 ] 74 - Time domain features (24) - Frequency domain features (24) - Features extracted from the derivatives of the signals (16) - Entropy (4) - Peak correlation between signals (3) - Cross-correlation delay (3) Channa et al. [ 46 ] 43 - Time domain features (28) - Frequency domain features (12) - Peak correlation between signals (3) Filtered signal (256-sample window) 768 (256 × 3) Filtered signal obtained from the triaxial sensors (accelerometer or gyroscope). Contextual FFT 384 (128 × 3) Concatenated single-side FFT of two consecutive windows. A single window (256 samples) was divided into 2 windows of 128 samples before FFT computation. 3.2.3. Data Augmentation The synthetic minority over-sampling technique (SMOTE) [ 50 ] was used to balance the data input to classic ML classifiers. Specifically, the classes with a minority number of sliding windows (i.e., UPDRS 1 = 768; UPDRS 2 = 1027; UPDRS 3 = 2132) were resampled to provide the same number of sliding windows as the majority class (UPDRS 0 = 5253). The number of nearest neighbors used to construct the synthetic samples was set to 5. This procedure produced an increase in the dataset size of 53%. As far as the raw signals input to convolutional models are concerned, the application of signal permutation and magnitude warping [ 51 ] were employed to quadruple the amount of data. In the former case, the input data were sliced into four equal-length segments, and these segments were randomly permuted to create a new sliding window. As for the latter method, convolution between the input data and a smooth (randomly generated) curve was performed to change the magnitude of the samples of the sliding window. All the described data augmentation techniques were applied only to the training subsets, while the testing subset remained unchanged. Figure 4shows examples of the original data and the data augmentation techniques applied to the gyroscope signals. As shown in Figure 4b,e, portions of the signal were randomly permuted from the original signals (see Figure 4a,d), while, in Figure 4c,f, the amplitude of the original signals was modified by a randomly generated (smooth) curve. Electronics 2022,11, 3879 8 of 19 (a) Original signal (b) Permutation (c) Magnitude warping (d) Original signal (e) Permutation (f) Magnitude warping Figure 4. Data augmentation techniques applied in the gyroscope signals. ( a ) sample of the original signal corresponding to exercise 4 (finger tapping); ( b ) permutation of a sample signal of the exercise 4; ( c ) magnitude warping of a sample signal of exercise 4; ( d ) sample of the original signal corresponding to exercise 5 (hand movements); ( e ) permutation of a sample signal of the exercise 5; ( f ) magnitude warping of a sample signal of exercise 5. 3.3. Classification Algorithms Different algorithms were implemented to predict the bradykinesia severity in PD patients and control subjects, resulting in a multi-class classification task. The output of the implemented models is a value between 0 and 3, according to the clinical bradykinesia score provided by the MDS-UPDRS scale. For comparison proposes, two detection methods have been reproduced and evaluated to generate baseline metrics. The reproduced methods were the feature sets proposed in Shawen et al. [ 31 ] and Channa et al. [ 46 ]; these feature sets fed an RF classification model with 100 estimators, as proposed in [ 31 ]. Additional parameters of the RF classification algorithm were a minimum sample split equal to 2, a minimum sample leaf equal to 1, and the split criterion was Gini impurity [52]. The following DL algorithms were trained either using raw inertial signals or using the Contextual FFT data representation, as previously described in Section 3.2.2. CNN. It consists of an input layer (256 features and 3 channels), connected to three one-dimensional convolutional layers of (1D-CNN), all three with 64 filters of size equal to 8 and rectified linear unit (ReLU) activation functions. Then, a global average pooling (GAP) layer was connected. For classification tasks, a multi-layer-perceptron (MLP) block composed of a densely connected layer with 260 units and ReLU activation was densely connected to a softmax layer with 4 units, corresponding to the number of output classes (i.e., bradykinesia severity score from 0 to 3). Contextual CNN. The features extracted by the contextual windows method were evaluated. In this case, the architecture of the CNN used is composed of an input layer (256 features and 3 channels), connected to three 1D-CNN, the first one with 64 filters of size 8 and the next two with 20 filters of size 8, all of them with ReLU activation function. A GAP layer was then connected. For the classification tasks, an MLP block composed of a densely connected layer with 180 units and a ReLU activation function was connected to the classification layer, made of 4 units with a softmax activation function. CNN (PI). As a novel approach, a CNN with patch input (PI) was proposed and evaluated. The patch extraction was implemented using a 1D-CNN. For this task, the kernel Electronics 2022,11, 3879 9 of 19 and stride parameters were set with the same value. In this way, the convolutional layers can act as an automatic patch extractor and bring equivalent results to patching extraction strategies such as those employed in Transformer-based models and isotropic computervision models [ 53 , 54 ], in which images are divided into non-overlapping square patches in raster-scan order. The proposed patching input strategy adapted to process multi-channel signals is shown in Figure 5. Figure 5. Patch input strategy with 1D-Convolution. The training and evaluation of this latter model were performed using the filtered inertial signals. The architecture consists of an input layer implementing the PI strategy using 64 filters with kernel size and stride equal to 8 (in both cases). The input layer was connected to one 1D convolutional layer with 64 filters, a kernel size of 3, and a ReLU activation function. Then, a max pooling layer with a pool size of 2 and a subsequent GAP layer were connected. For the classification tasks, the MLP block included two densely connected layers, with 100 and 50 units, respectively, both with ReLU activation functions. Finally, these layers were connected to the final classification with 4 units and a softmax activation function. The architecture for the DNN with convolutional layers and PI is shown in Figure 6. Figure 6. Proposed architecture for a CNN with patch input and MLP. CNN (PI) + RF. The combination of CNN with PI and RF classification algorithm was evaluated. In this approach, the convolutional (with path input) block acts as a feature extractor, while the RF model (with 100 estimators) performs the classification tasks. While the CNN block allows the automatic extraction of features from the raw signal, the RF classifier takes advantage of a large number of individual decision trees that operate as an ensemble, providing good performance and high generalization capabilities. Electronics 2022,11, 3879 16 of 19 procedure may be carried out, as achieved in [ 40 ], where the presence or absence of bradykinesia was estimated during unconstrained ADLs. The proposed solution, further improved and validated on a larger cohort of PD patients, may be used to complement traditional outpatient visits. Specifically, data collected during the sporadic clinical examination can be employed for further training the proposed automatic scoring system. Afterward, the wearable solution can be used in the home setting to passively collect information regarding bradykinesia presence and severity. Finally, the information will be able to be assessed by clinicians for evaluating the evolution of the symptom over time and its fluctuations throughout the day, eventually planning proper therapy adjustments. The present study is intended to show the potential of consumer wearable technology and DL approaches to detect the severity of bradykinesia by using data recorded during standardized MDS-UPDRS upper limbs’ motor tasks. Moreover, different ML and DL methodologies are proposed and compared, further discussing the effect of data augmentation, input type, and architectures. Future studies will be in the direction of increasing the data set size, by enrolling a larger patient cohort. Then, the development of an automatic scoring system working in non-supervised conditions [ 40 ] would pave the way to continuous, long-term, unobtrusive monitoring of PD in home environments. Author Contributions: L.S.: Conceptualization, Software, Methodology, Validation, Formal analysis, Writing—review and editing. B.D.: Data Curation, Software, Investigation, Methodology, Formal analysis, Writing—Original Draft. L.B.: Conceptualization, Formal analysis, Validation, Writing— review and editing. N.C.: Funding acquisition, Supervision, Writing—review and editing. S.C.: Validation, Formal analysis, Writing—review and editing. P.A.: Resources, Supervision, Writing— review and editing. J.M.L.: Resources, Supervision, Writing—review and editing. G.D.A.: Funding acquisition, Project administration, Writing—review and editing. I.P.: Formal analysis, Project administration, Writing—review and editing. All authors have read and agreed to the published version of the manuscript. Funding: Part of this research was funded by the project “Tecnologías Capacitadoras para la Asistencia, Seguimiento y Rehabilitación de Pacientes con Enfermedad de Parkinson”. Centro Internacional sobre el envejecimiento, CENIE (código 0348_CIE_6_E) Interreg V-A España-Portugal (POCTEP); and (2) FCT—Fundação para a Ciência e Tecnologia within the R&D Units Project Scope: UIDB/00319/2020. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki. In addition, this study was approved by the Institutional Review Board (Ethics Committee) of the Universidad Politécnica de Madrid (date of approval: 18 June 2018) and the Ethics Committee of the University of Minho with the document identification CE.CSH 031/2018 (date of approval: 11 December 2018). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available because they contain protected patient health information. Acknowledgments: This work has been supported by: (1) Grupo de Investigación en Instrumentación y Acústica Aplicada (I2A2). ETSI Industriales. Universidad Politécnica de Madrid; and (2) ALGORITMI Research Centre, University of Minho (Portugal). Conflicts of Interest: The authors declare no conflict of interest. Electronics 2022,11, 3879 17 of 19 Abbreviations The following abbreviations are used in this manuscript: AI Artificial intelligence ADAM Adaptive moment estimation ADLs Activities of daily living AUC Area under the curve CNN Convolutional neural network CV Cross-validation DA Data augmentation DL Deep learning DNN Deep neural network FFT Fast Fourier Transform GAP Global average pooling kNN k-nearest neighbors IMU Inertial measurement unit LR Linear regression ML Machine learning MLP Multi-layer perceptron PCA Principal component analysis PD Parkinson’s disease QoL Quality of life ReLU Rectified linear unit RF Random forest ROC Receiver operating characteristic RMSE Root mean square error SMOTE Synthetic minority over-sampling technique References 1. Pringsheim, T.; Jette, N.; Frolkis, A.; Steeves, T.D. The prevalence of Parkinson’s disease: A systematic review and meta-analysis. Mov. Disord. 2014,29, 1583–1590. [CrossRef] [PubMed] 2. Jankovic, J. Parkinson’s disease: Clinical features and diagnosis. J. Neurol. Neurosurg. Psychiatry 2008 ,79, 368–376. [CrossRef] [PubMed] 3. Kalia, L.; Lang, A. Parkinson’s disease. Lancet 2015,386, 896–912. [CrossRef] 4. Hou, J.; Lai, E. Non-motor Symptoms of Parkinson’s Disease. Int. J. Gerontol. 2007,1, 53–64. [CrossRef] 5. Williams, D.; Watt, H.; Lees, A. Predictors of falls and fractures in bradykinetic rigid syndromes: A retrospective study. J. Neurol. Neurosurg. Psychiatry 2006,77, 468–473. [CrossRef] [PubMed] 6. Bronte-Stewart, H. Postural instability in idiopathic Parkinson’s disease: The role of medication and unilateral pallidotomy. Brain 2002,125, 2100–2114. [CrossRef] [PubMed] 7. Patel, S.; Lorincz, K.; Hughes, R.; Huggins, N.; Growdon, J.; Standaert, D.; Akay, M.; Dy, J.; Welsh, M.; Bonato, P. Monitoring motor fluctuations in patients with Parkinson’s disease using wearable sensors. IEEE Trans. Inf. Technol. Biomed. 2009 ,13, 864–873. [CrossRef] [PubMed] 8. Davie, C. A review of Parkinson’s disease. Br. Med. Bull. 2008,86, 109–127. [CrossRef] 9. Fahn, S. Parkinson disease, the effect of levodopa, and the ELLDOPA trial. Arch. Neurol. 1999,56, 529–535. [CrossRef] 10. Olanow, C.W.; Stern, M.B.; Sethi, K. The scientific and clinical basis for the treatment of Parkinson’s disease. Neurology 2009 , 72, S1–S136. [CrossRef] [PubMed] 11. Parkinson Study Group. Evaluation of dyskinesias in a pilot, randomized, placebo-controlled trial of remacemide in advanced Parkinson disease. Arch. Neurol. 2001,58, 1660–1668. [CrossRef] [PubMed] 12. Hughes, A.J.; Daniel, S.E.; Blankson, S.; Lees, A.J. A clinicopathologic study of 100 cases of Parkinson’s disease. Arch. Neurol. 1993,50, 140–148. [CrossRef] 13. Berardelli, A. Pathophysiology of bradykinesia in Parkinson’s disease. Brain 2001,124, 2131–2146. [CrossRef] [PubMed] 14. Vingerhoets, F.; Schulzer, M.; Calne, D.; Snow, B. Which clinical sign of Parkinson’s disease best reflects the nigrostriatal lesion? Ann. Neurol. 1997,41, 58–64. [CrossRef] [PubMed] 15. Monje, M.H.G.; Foffani, G.; Obeso, J.; Sànchez-Ferro, Á. New Sensor and Wearable Technologies to Aid in the Diagnosis and Treatment Monitoring of Parkinson’s Disease. Annu. Rev. Biomed. Eng. 2019,21, 111–143. [CrossRef] [PubMed] 16. Goetz, C.G.; Tilley, B.C.; Shaftman, S.R.; Stebbins, G.T.; Fahn, S.; Martinez-Martin, P.; Poewe, W.; Sampaio, C.; Stern, M.B.; Dodel, R.; et al. Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Scale presentation and clinimetric testing results. Mov. Disord. Off. J. Mov. Disord. Soc. 2008,23, 2129–2170. [CrossRef] Electronics 2022,11, 3879 18 of 19 17. Borzì, L.; Varrecchia, M.; Sibille, S.; Olmo, G.; Artusi, C.A.; Fabbri, M.; Rizzone, M.G.; Romagnolo, A.; Zibetti, M.; Lopiano, L. Smartphone-Based Estimation of Item 3.8 of the MDS-UPDRS-III for Assessing Leg Agility in People with Parkinson’s Disease. IEEE Open J. Eng. Med. Biol. 2020,1, 140–147. [CrossRef] 18. Espay, A.J.; Hausdorff, J.M.; Sánchez-Ferro, Á.; Klucken, J.; Merola, A.; Bonato, P.; Paul, S.S.; Horak, F.B.; Vizcarra, J.A.; Mestre, T.A.; et al. A Roadmap for Implementation of Patient-Centered Digital Outcome Measures in Parkinson’s disease Obtained Using Mobile Health Technologies. Mov. Disord. 2019,34, 657–663. [CrossRef] 19. Luis-Martínez, R.; Monje, M.H.; Antonini, A.; Sánchez-Ferro, Á.; Mestre, T.A. Technology-enabled care: Integrating multidisciplinary care in Parkinson’s disease through digital technology. Front. Neurol. 2020,11, 575975. [CrossRef] 20. Virginia Anikwe, C.; Friday Nweke, H.; Chukwu Ikegwu, A.; Adolphus Egwuonwu, C.; Uchenna Onu, F.; Rita Alo, U.; Wah Teh, Y. Mobile and wearable sensors for data-driven health monitoring system: State-of-the-art and future prospect. Expert Syst. Appl. 2022,202, 117362. [CrossRef] 21. Tau¸tan, A.M.; Ionescu, B.; Santarnecchi, E. Artificial intelligence in neurodegenerative diseases: A review of available tools with a focus on machine learning techniques. Artif. Intell. Med. 2021,117, 102081. [CrossRef] [PubMed] 22. Rovini, E.; Maremmani, C.; Cavallo, F. How Wearable Sensors Can Support Parkinson’s Disease Diagnosis and Treatment: A Systematic Review. Front Neurosci. 2017,11, 555. [CrossRef] [PubMed] 23. Hubble, R.; Naughton, G.; Silburn, P.; Cole, M.H. Wearable Sensor Use for Assessing Standing Balance and Walking Stability in People with Parkinson’s Disease: A Systematic Review. PLoS ONE 2015,10, e0123705. [CrossRef] 24. Borzì, L.; Olmo, G.; Artusi, C.A.; Fabbri, M.; Rizzone, M.G.; Romagnolo, A.; Zibetti, M.; Lopiano, L. A new index to assess turning quality and postural stability in patients with Parkinson’s disease. Biomed. Signal Process. Control 2020,62, 102059. [CrossRef] 25. Mei, J.; Desrosiers, C.; Frasnelli, J. Machine Learning for the Diagnosis of Parkinson’s Disease: A Review of Literature. Front. Aging Neurosci. 2021,13, 633752. [CrossRef] 26. Cubo, E.; Mir Rivera, P.; Sánchez Ferro, Á. Manual SEN de Nuevas Tecnologías en Trastornos del Movimiento; Ediciones SEN: Madrid, Spain, 2021. 27. Heldman, D.A.; Urrea-Mendoza, E.; Lovera, L.C.; Schmerler, D.A.; Garcia, X.; Mohammad, M.E.; McFarlane, M.C.U.; Giuffrida, J.P.; Espay, A.J.; Fernandez, H.H. App-based bradykinesia tasks for clinic and home assessment in Parkinson’s disease: Reliability and responsiveness. J. Park. Dis. 2017,7, 741–747. [CrossRef] [PubMed] 28. Griffiths, R.I.; Kotschet, K.; Arfon, S.; Xu, Z.M.; Johnson, W.; Drago, J.; Evans, A.; Kempster, P.; Raghav, S.; Horne, M.K. Automated assessment of bradykinesia and dyskinesia in Parkinson’s disease. J. Park. Dis. 2012,2, 47–55. [CrossRef] 29. Dunnewold, R.J.; Hoff, J.I.; van Pelt, H.C.; Fredrikze, P.Q.; Wagemans, E.A.; van Hilten, B.J. Ambulatory quantitative assessment of body position, bradykinesia, and hypokinesia in Parkinson’s disease. J. Clin. Neurophysiol. 1998 ,15, 235–242. [CrossRef] [PubMed] 30. Salarian, A. Ambulatory Monitoring of Motor Functions in Patients with Parkinson’s Disease Using Kinematic Sensors; Technical Report; EPFL: Lausanne, Switzerland, 2006. 31. Shawen, N.; O’Brien, M.K.; Venkatesan, S.; Lonini, L.; Simuni, T.; Hamilton, J.L.; Ghaffari, R.; Rogers, J.A.; Jayaraman, A. Role of data measurement characteristics in the accurate detection of Parkinson’s disease symptoms using wearable sensors. J. Neuroeng. Rehabil. 2020,17, 52. [CrossRef] [PubMed] 32. Mahadevan, N.; Demanuele, C.; Zhang, H.; Volfson, D.; Ho, B.; Erb, M.K.; Patel, S. Development of digital biomarkers for resting tremor and bradykinesia using a wrist-worn wearable device. NPJ Digit. Med. 2020,3, 5. [CrossRef] [PubMed] 33. Parisi, F.; Ferrari, G.; Giuberti, M.; Contin, L.; Cimolin, V.; Azzaro, C.; Albani, G.; Mauro, A. Body-Sensor-Network-Based Kinematic Characterization and Comparative Outlook of UPDRS Scoring in Leg Agility, Sit-to-Stand, and Gait Tasks in Parkinson’s Disease. IEEE J. Biomed. Health Inform. 2015,19, 1777–1793. [CrossRef] 34. Giuberti, M.; Ferrari, G.; Contin, L.; Cimolin, V.; Azzaro, C.; Albani, G.; Mauro, A. Automatic UPDRS Evaluation in the Sit-to-Stand Task of Parkinsonians: Kinematic Analysis and Comparative Outlook on the Leg Agility Task. IEEE J. Biomed. Health Inform. 2015,19, 803–814. [CrossRef] [PubMed] 35. Aghanavesi, S.; Bergquist, F.; Nyholm, D.; Senek, M.; Memedi, M. Motion sensor-based assessment of Parkinson’s disease motor symptoms during leg agility tests: Results from levodopa challenge. IEEE J. Biomed. Health Inform. 2019 ,24, 111–119. [CrossRef] [PubMed] 36. Bengio, Y. Deep learning of representations: Looking forward. In Statistical Language and Speech Processing, Proceedings of the First International Conference, SLSP 2013, Tarragona, Spain, 29–31 July 2013; Springer: Berlin/Heidelberg, Germany, 2013; pp. 1–37. 37. Glorot, X.; Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Sardinia, Italy, 13–15 May 2010; Teh, Y.W., Titterington, M., Eds.; PMLR: Sardinia, Italy, 2010; Volume 9, pp. 249–256. 38. Eskofier, B.M.; Lee, S.I.; Daneault, J.F.; Golabchi, F.N.; Ferreira-Carvalho, G.; Vergara-Diaz, G.; Sapienza, S.; Costante, G.; Klucken, J.; Kautz, T.; et al. Recent machine learning advancements in sensor-based mobility analysis: Deep learning for Parkinson’s disease assessment. In Proceedings of the 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 16–20 August 2016; pp. 655–658. 39. Park, D.J.; Lee, J.W.; Lee, M.J.; Ahn, S.J.; Kim, J.; Kim, G.L.; Ra, Y.J.; Cho, Y.N.; Jeong, W.B. Evaluation for Parkinsonian Bradykinesia by deep learning modeling of kinematic parameters. J. Neural Transm. 2021,128, 181–189. [CrossRef] Electronics 2022,11, 3879 19 of 19 40. Pfister, F.M.; Um, T.T.; Pichler, D.C.; Goschenhofer, J.; Abedinpour, K.; Lang, M.; Endo, S.; Ceballos-Baumann, A.O.; Hirche, S.; Bischl, B.; et al. High-Resolution Motor State Detection in Parkinson’s Disease Using Convolutional Neural Networks. Sci. Rep. 2020,10, 5860. [CrossRef] 41. Lonini, L.; Dai, A.; Shawen, N.; Simuni, T.; Poon, C.; Shimanovich, L.; Daeschler, M.; Ghaffari, R.; Rogers, J.A.; Jayaraman, A. Wearable sensors for Parkinson’s disease: Which data are worth collecting for training symptom detection models. Npj Digit. Med. 2018,1, 64. [CrossRef] 42. Sigcha, L.; Pavón, I.; Costa, N.; Costa, S.; Gago, M.; Arezes, P.; López, J.M.; De Arcas, G. Automatic Resting Tremor Assessment in Parkinson’s Disease Using Smartwatches and Multitask Convolutional Neural Networks. Sensors 2021,21, 291. [CrossRef] 43. Hoehn, M.M.; Yahr, M.D. Parkinsonism: Onset, progression, and mortality. Neurology 1998,50, 318. [CrossRef] 44. Bao, L.; Intille, S.S. Activity recognition from user-annotated acceleration data. In Pervasive Computing, Proceedings of theSecond International Conference, PERVASIVE 2004, Vienna, Austria, 21–23 April 2004; Springer: Berlin/Heidelberg, Germany, 2004; pp. 1–17. 45. Bonato, P.; Sherrill, D.M.; Standaert, D.G.; Salles, S.S.; Akay, M. Data mining techniques to detect motor fluctuations in Parkinson’s disease. In Proceedings of the 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 26–30 August 2004; Volume 2, pp. 4766–4769. 46. Channa, A.; Ifrim, R.C.; Popescu, D.; Popescu, N. A-WEAR bracelet for detection of hand tremor and bradykinesia in Parkinson’s patients. Sensors 2021,21, 981. [CrossRef] 47. San-Segundo, R.; Navarro-Hellín, H.; Torres-Sánchez, R.; Hodgins, J.; De la Torre, F. Increasing robustness in the detection of freezing of gait in Parkinson’s disease. Electronics 2019,8, 119. [CrossRef] 48. Sigcha, L.; Costa, N.; Pavón, I.; Costa, S.; Arezes, P.; López, J.; De Arcas, G. Deep Learning Approaches for Detecting Freezing of Gait in Parkinson’s Disease Patients through On-Body Acceleration Sensors. Sensors 2020,20, 1895. [CrossRef] 49. Sigcha, L.; Borzì, L.; Pavón, I.; Costa, N.; Costa, S.; Arezes, P.; López, J.M.; De Arcas, G. Improvement of Performance in Freezing of Gait detection in Parkinson’s Disease using Transformer networks and a single waist-worn triaxial accelerometer. Eng. Appl. Artif. Intell. 2022,116, 105482. [CrossRef] 50. Chawla, N.V.; Bowyer, K.W.; Hall, L.O.; Kegelmeyer, W.P. SMOTE: Synthetic minority over-sampling technique. J. Artif. Intell. Res. 2002,16, 321–357. [CrossRef] 51. Um, T.T.; Pfister, F.M.J.; Pichler, D.; Endo, S.; Lang, M.; Hirche, S.; Fietzek, U.; Kuli´c, D. Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks. In Proceedings of the 19th ACM International Conference on Multimodal Interaction, Glasgow, UK, 13–17 November 2017. [CrossRef] 52. Jost, L. Entropy and diversity. Oikos 2006,113, 363–375. [CrossRef] 53. Trockman, A.; Kolter, J.Z. Patches are all you need? arXiv 2022, arXiv:2201.09792. 54. Hassani, A.; Walton, S.; Shah, N.; Abuduweili, A.; Li, J.; Shi, H. Escaping the big data paradigm with compact transformers. arXiv 2021, arXiv:2104.05704. 55. Li, L.; Jamieson, K.; DeSalvo, G.; Rostamizadeh, A.; Talwalkar, A. Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization. J. Mach. Learn. Res. 2017,18, 6765–6816. 56. Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv 2014, arXiv:1412.6980. 57. Japkowicz, N.; Shah, M. Evaluating Learning Algorithms: A Classification Perspective; Cambridge University Press: Cambridge, UK, 2011. 58. Breiman, L. Random Forests. Mach. Learn. 2001,45, 5–32. [CrossRef]