Data Analytics and Applications of the Wearable Sensors in Healthcare Printed Edition of the Special Issue Published in Sensors www.mdpi.com/journal/sensors Shabbir Syed-Abdul, Luis Fernandez Luque, Pei-Yun Sabrina Hsueh, Juan M. García-Gomez and Begoña Garcia-Zapirain Edited by Data Analytics and Applications of the Wearable Sensors in Healthcare Data Analytics and Applications of the Wearable Sensors in Healthcare Special Issue Editors Shabbir Syed-Abdul Luis Fernandez Luque Pei-Yun Sabrina Hsueh Juan M. Garc ́ ıa-Gomez Bego ̃ na Garcia-Zapirain MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin Pei-Yun Sabrina Hsueh IBM Thomas J. Watson Research Center USA Luis Fernandez Luque Qatar Computing Research Institute Qatar Bego ̃ na Garcia-Zapirain University of Deusto Spain Special Issue Editors Shabbir Syed-Abdul Taipei Medical University Taiwan Juan M. Garc ́ ıa-Gomez Universitat Politecnica de Valencia Spain Editorial Office MDPI St. Alban-Anlage 66 4052 Basel, Switzerland This is a reprint of articles from the Special Issue published online in the open access journal Sensors (ISSN 1424-8220) (available at: https://www.mdpi.com/journal/sensors/special issues/Wearable Sensors in the Healthcare). For citation purposes, cite each article independently as indicated on the article page online and as indicated below: LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. Journal Name Year , Article Number , Page Range. ISBN 978-3-03936-350-6 (Pbk) ISBN 978-3-03936-351-3 (PDF) c © 2020 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications. The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND. Contents About the Special Issue Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Mohy Uddin and Shabbir Syed-Abdul Data Analytics and Applications of the Wearable Sensors in Healthcare: An Overview Reprinted from: Sensors 2020 , 20 , 1379, doi:10.3390/s20051379 . . . . . . . . . . . . . . . . . . . . 1 Jose-Luis Bayo-Monton, Antonio Martinez-Millana, Weisi Han, Carlos Fernandez-Llatas and Vicente Traver Wearable Sensors Integrated with Internet of Things for Advancing eHealth Care Reprinted from: Sensors 2018 , 18 , 1851, doi:10.3390/s18061851 . . . . . . . . . . . . . . . . . . . . 7 Saurabh Singh Thakur, Shabbir Syed-Abdul, Hsiao-Yean (Shannon) Chiu, Ram Babu Roy, Po-Yu Huang, Shwetambara Malwade, Aldilas Achmad Nursetyo and Yu-Chuan (Jack) Li Artificial-Intelligence-Based Prediction of Clinical Events among Hemodialysis Patients Using Non-Contact Sensor Data Reprinted from: Sensors 2018 , 18 , 2833, doi:10.3390/s18092833 . . . . . . . . . . . . . . . . . . . . 25 Rob Argent, Patrick Slevin, Antonio Bevilacqua, Maurice Neligan, Ailish Daly and Brian Caulfield Wearable Sensor-Based Exercise Biofeedback for Orthopaedic Rehabilitation: A Mixed Methods User Evaluation of a Prototype System Reprinted from: Sensors 2019 , 19 , 432, doi:10.3390/s19020432 . . . . . . . . . . . . . . . . . . . . . 41 Miguel Enrique Iglesias Mart ́ ınez, Juan M. Garc ́ ıa-Gomez, Carlos S ́ aez, Pedro Fern ́ andez de C ́ ordoba and J. Alberto Conejero Feature Extraction and Similarity of Movement Detection during Sleep, Based on Higher Order Spectra and Entropy of the Actigraphy Signal: Results of the Hispanic Community Health Study/Study of Latinos Reprinted from: Sensors 2018 , 18 , 4310, doi:10.3390/s18124310 . . . . . . . . . . . . . . . . . . . . 55 Shih-Sung Lin, Chien-Wu Lan, Hao-Yen Hsu and Sheng-Tao Chen Data Analytics of a Wearable Device for Heat Stroke Detection Reprinted from: Sensors 2018 , 18 , 4347, doi:10.3390/s18124347 . . . . . . . . . . . . . . . . . . . . 73 Francisco Luna-Perej ́ on, Manuel Jes ́ us Dom ́ ınguez-Morales and Anton Civit-Balcells Wearable Fall Detector Using Recurrent Neural Networks Reprinted from: Sensors 2019 , 19 , 4885, doi:10.3390/s19224885 . . . . . . . . . . . . . . . . . 97 Katharina Stollenwerk, Jonas M ̈ uller, Andr ́ e Hinkenjann and Bj ̈ orn Kr ̈ uger Analyzing Spinal Shape Changes During Posture Training Using a Wearable Device Reprinted from: Sensors 2019 , 19 , 3625, doi:10.3390/s19163625 . . . . . . . . . . . . . . . . . . . . 117 Mario Vega-Barbas, Jose A. Diaz-Olivares, Ke Lu, Mikael Forsman, Fernando Seoane and Farhad Abtahi P-Ergonomics Platform: Toward Precise, Pervasive, and Personalized Ergonomics using Wearable Sensors and Edge Computing Reprinted from: Sensors 2019 , 19 , 1225, doi:10.3390/s19051225 . . . . . . . . . . . . . . . . . . . . 139 v Wen-Yen Lin, Hong-Lin Ke, Wen-Cheng Chou, Po-Cheng Chang, Tsai-Hsuan Tsai and Ming-Yih Lee Realization and Technology Acceptance Test of a Wearable Cardiac Health Monitoring and Early Warning System with Multi-Channel MCGs and ECG Reprinted from: Sensors 2018 , 18 , 3538, doi:10.3390/s18103538 . . . . . . . . . . . . . . . . . . . . 157 Se-Min Lim, Hyeong-Cheol Oh, Jaein Kim, Juwon Lee and Jooyoung Park LSTM-Guided Coaching Assistant for Table Tennis Practice Reprinted from: Sensors 2018 , 18 , 4112, doi:10.3390/s18124112 . . . . . . . . . . . . . . . . . . . . 179 Ke Lu, Liyun Yang, Fernando Seoane, Farhad Abtahi, Mikael Forsman and Kaj Lindecrantz Fusion of Heart Rate, Respiration and Motion Measurements from a Wearable Sensor System to Enhance Energy Expenditure Estimation Reprinted from: Sensors 2018 , 18 , 3092, doi:10.3390/s18093092 . . . . . . . . . . . . . . . . . . . . 193 Andreas Ejupi and Carlo Menon Detection of Talking in Respiratory Signals: A Feasibility Study Using Machine Learning and Wearable Textile-Based Sensors Reprinted from: Sensors 2018 , 18 , 2474, doi:10.3390/s18082474 . . . . . . . . . . . . . . . . . . . . 205 Ambra Cesareo, Ylenia Previtali, Emilia Biffi and Andrea Aliverti Assessment of Breathing Parameters Using an Inertial Measurement Unit (IMU)-Based System Reprinted from: Sensors 2019 , 19 , 88, doi:10.3390/s19010088 . . . . . . . . . . . . . . . . . . . . . . 217 Jose Manjarres, Pedro Narvaez, Kelly Gasser, Winston Percybrooks and Mauricio Pardo Physical Workload Tracking Using Human Activity Recognition with Wearable Devices Reprinted from: Sensors 2020 , 20 , 39, doi:10.3390/s20010039 . . . . . . . . . . . . . . . . . . . . . 241 Hyung Seok Nam, Woo Hyung Lee, Han Gil Seo, Yoon Jae Kim, Moon Suk Bang and Sungwan Kim Inertial Measurement Unit Based Upper Extremity Motion Characterization for Action Research Arm Test and Activities of Daily Living Reprinted from: Sensors 2019 , 19 , 1782, doi:10.3390/s19081782 . . . . . . . . . . . . . . . . . . . . 259 Wei Zhang, Michael Schwenk, Sabato Mellone, Anisoara Paraschiv-Ionescu, Beatrix Vereijken, Mirjam Pijnappels, A. Stefanie Mikolaizak, Elisabeth Boulton, Nini H. Jonkman, Andrea B. Maier, Jochen Klenk, Jorunn Helbostad, Kristin Taraldsen and Kamiar Aminian Complexity of Daily Physical Activity Is More Sensitive Than Conventional Metrics to Assess Functional Change in Younger Older Adults Reprinted from: Sensors 2018 , 18 , 2032, doi:10.3390/s18072032 . . . . . . . . . . . . . . . . . . . . 269 Chien-Chin Hsu, Bor-Shing Lin, Ke-Yi He and Bor-Shyh Lin Design of a Wearable 12-Lead Noncontact Electrocardiogram Monitoring System Reprinted from: Sensors 2019 , 19 , 1509, doi:10.3390/s19071509 . . . . . . . . . . . . . . . . . . . . 281 Udeni Jayasinghe and William S. Harwin, Faustina Hwang Comparing Clothing-Mounted Sensors with Wearable Sensors for Movement Analysis and Activity Classification Reprinted from: Sensors 2020 , 20 , 82, doi:10.3390/s20010082 . . . . . . . . . . . . . . . . . . . . . . 295 Hoda Allahbakhshi, Lindsey Conrow, Babak Naimi and Robert Weibel Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection Reprinted from: Sensors 2020 , 20 , 588, doi:10.3390/s20030588 . . . . . . . . . . . . . . . . . . . . . 309 vi Ying Kuen Cheung, Pei-Yun Sabrina Hsueh, Ipek Ensari, Joshua Z. Willey and Keith M. Diaz Quantile Coarsening Analysis of High-Volume Wearable Activity Data in a Longitudinal Observational Study Reprinted from: Sensors 2018 , 18 , 3056, doi:10.3390/s18093056 . . . . . . . . . . . . . . . . . . . . 331 Yashodhan Athavale and Sridhar Krishnan A Device-Independent Efficient Actigraphy Signal-Encoding System for Applications in Monitoring Daily Human Activities and Health Reprinted from: Sensors 2018 , 18 , 2966, doi:10.3390/s18092966 . . . . . . . . . . . . . . . . . . . . 343 Luis A. Trejo and Ari Yair Barrera-Animas Towards an Efficient One-Class Classifier for Mobile Devices and Wearable Sensors on the Context of Personal Risk Detection Reprinted from: Sensors 2018 , 18 , 2857, doi:10.3390/s18092857 . . . . . . . . . . . . . . . . . . . . 359 Aras Yurtman, Billur Barshan,* and Barı ̧ s Fidan Activity Recognition Invariant to Wearable Sensor Unit OrientationUsing Differential Rotational Transformations Represented by Quaternions Reprinted from: Sensors 2018 , 18 , 2725, doi:10.3390/s18082725 . . . . . . . . . . . . . . . . . . . . 379 Arindam Dutta, Owen Ma, Meynard Toledo, Alberto Florez Pregonero, Barbara E. Ainsworth, Matthew P. Buman and Daniel W. Bliss Identifying Free-Living Physical Activities Using Lab-Based Models with Wearable Accelerometers Reprinted from: Sensors 2018 , 18 , 3893, doi:10.3390/s18113893 . . . . . . . . . . . . . . . . . . . . 407 Samanta Rosati, Gabriella Balestra and Marco Knaflitz Comparison of Different Sets of Features for Human Activity Recognition by Wearable Sensors Reprinted from: Sensors 2018 , 18 , 4189, doi:10.3390/s18124189 . . . . . . . . . . . . . . . . . . . . 421 Davide Morelli, Alessio Rossi, Massimo Cairo and David A. Clifton Analysis of the Impact of Interpolation Methods of Missing RR-Intervals Caused by Motion Artifacts on HRV Features Estimations Reprinted from: Sensors 2019 , 19 , 3163, doi:10.3390/s19143163 . . . . . . . . . . . . . . . . . . . . 437 Alexis Fortin-Cˆ ot ́ e, Jean-S ́ ebastien Roy, Laurent Bouyer, Philip Jackson and Alexandre Campeau-Lecours Allumo: Preprocessing and Calibration Software for Wearable Accelerometers Used in Posture Tracking Reprinted from: 2020 , 20 , 229, doi:10.3390/s20010229 . . . . . . . . . . . . . . . . . . . . . . . . . 451 Robert W. Broadley, Jochen Klenk, Sibylle B. Thies, Laurence P. J. Kenney and Malcolm H. Granat Methods for the Real-World Evaluation of Fall Detection Technology: AScoping Review Reprinted from: Sensors 2018 , 18 , 2060, doi:10.3390/s18072060 . . . . . . . . . . . . . . . . . . . . 459 vii About the Special Issue Editors Shabbir Syed-Abdul M.D., M.S., Ph.D., Associate Professor at Taipei Medical University. Dr. Shabbir Syed-Abdul is an associate professor at the Institute of Biomedical Informatics, a leading researcher and a principal investigator at the International Center for Health Information Technology, Taipei, Taiwan. He is an educator for MOOCs on FutureLearn, one of his renowned courses is on the Internet of Things for active aging. Shabbir’s major research interests are long-term care with wearable technologies, mHealth, big data analysis and visualization, artificial intelligence, personal health records, social network in healthcare, and hospital information systems. He wants to empower care providers and improve patient engagement. He feels one of the methods to achieve this goal is to focus on the management and flow of the health and medical information among health care providers and seekers. Dr. Shabbir’s previous experiences working as a physician, researcher, and principal investigator both in the developed and developing world makes him an ideal candidate for this post leading advance global knowledge on the factors that impact on the development of sustainable healthcare, in particular focusing on innovation and ageing populations. Luis Fernandez Luque , Ph.D. eHealth | Researcher at QCRI | Co-founder at Salumedia (Seville, Spain). Qatar Computing Research Institute, HBKU, Qatar Foundation, University of Tromsø (UiT). Luis is committed to identifying how technology can support people to live healthier. His passions are health, computing, and improving reality, which led him to a career in eHealth, both as a researcher and as anentrepreneur. Luis Fernandex is currently an eHeath researcher at the Qatar Computing Research Institute—Qatar Foundation. He is also the co-founder and advisor in the digital health start-up Salumedia. Pei-Yun Sabrina Hsueh received her B.S. from the National Taiwan University, M.S. from University of California at Berkeley, and Ph.D. from the University of Edinburgh. She is currently an IBM Academy of Technology Member and Research Staff Member in the Center for Computational Health at IBM Watson Research Center, Yorktown Heights, NY. Currently, she is leading the efforts toward computational health behavioral understanding with applications in care management and health decision support (using patient-generated health data from wearable sensors, mobile devices, and patient-reported outcomes). She is a pioneer in the area of consumer and pervasive health informatics, and is a serial winner of IBM Inventor Plateau awards, Manager Choice Awards, Eminence and Excellence, and Research Achievement Award. She has authored 20+ patents, 50+ technical papers, and 3 textbook chapters. She co-chairs the IBM Health Informatics professional interest community and is actively serving on various scientific program committees in ACM, IEEE, AMIA, and IMIA. She was elected as the Chair of Consumer and Pervasive Health Informatics Work Group of American Medical Informatics Association in 2017 for the term of 2018–2022. Her expertise on emerging health informatics topics makes her a sought-after speaker and consultant in science-driven industry solutions. Prior to IBM, she worked in the EU FP6 an FP7 Augmented Multiparty Interaction project with 22 partner sites across 7 countries and has been selected as European Google Anita Borg Scholar. ix Juan M. Garc ́ ıa-Gomez , Ph.D. in Computer Engineering (2009) is Senior Lecturer at UPV, a leader of the research group in data-driven approaches to improving healthcare at the ITACA Institute, and has been the advisor of 6 Ph.D. theses about clinical decision support systems applied to brain tumors, breast cancer and biomedical data quality assessment. In 2007, he was visiting researcher at ESAT-KU Leuven. In 2016, he was involved at IHI-ULC as a visiting researcher. His research interests mainly focus on machine learning techniques for developing clinical decision support systems (CDSS), the definition of data quality metrics for assessing biomedical repositories, and the design of medical applications based on big data technologies. He is author of more than 40 articles published in specialized journals (including JAMIA , Nucleic Acids Research , PLoS ONE , and Bioinformatics among others). His work on an HL7-CDA wrapper for facilitating semantic interoperability to rule-based Clinical Decision Support Systems (doi:10.1016/j.cmpb.2012.10.003) was selected as the best of the medical informatics papers published in 2013, sub-field of Health Information Systems by the International Medical Informatics Association (IMIA) for the IMIA Yearbook 2014. He has coordinated the pattern recognition WP of the FP6 EU eTUMOUR project (LSHC-CT-2004- 503094), the predictive analysis of the HealthAgents Project (IST-2004-27214), and WP leader in the FP7 European project Help4Mood (FP7-ICT-2009-4; 248765). Bego ̃ na Garcia-Zapirain , Ph.D. graduated in communications engineering, specializing in telematics at the Basque Country University, Bilbao (Spain) in 1994. In 2003, she defended her doctoral thesis in the pathological speech digital processing field. After many years of working for ZIV Company, in 1997, Dr. Garc ́ ıa joined the University of Deusto faculty as a lecturer in signal theory and electronics. She has been the head of the Telecommunication Department at the University of Deusto since 2002. In 2001, she created the e-Life research group at the same university, playing the role of lead researcher. She has participated in more than 80 research projects on international, national, and regional levels; published more than 35 papers in international scientific journals; and presented more than 160 papers in international and national scientific conferences. x sensors Editorial Data Analytics and Applications of the Wearable Sensors in Healthcare: An Overview Mohy Uddin 1 and Shabbir Syed-Abdul 2, * 1 Executive O ffi ce, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences, King Abdulaziz Medical City, Ministry of National Guard—Health A ff airs, Riyadh 11426, Saudi Arabia; drmohyuddin@yahoo.com 2 Graduate Institute of Biomedical Informatics, Taipei Medical University, Taipei 10675, Taiwan * Correspondence: drshabbir@tmu.edu.tw; Tel.: + 886-2-6638-2736 (ext. 1514); Fax: + 886-2-6638-0233 Received: 26 February 2020; Accepted: 29 February 2020; Published: 3 March 2020 1. Introduction Improving health and lives of people is undoubtedly one of the prime goals of healthcare organizations, policy-makers, and leaders around the world. The need of ageing, disability, long-term care, and palliative care in our current society pose formidable challenges for disease burden and healthcare systems that must be addressed [ 1 ]. In order to tackle the leading causes of morbidity and mortality that may result from infections to chronic conditions especially in older adults and ageing population, the accessibility and provision of long-term care and palliative care, when and where needed by them, is crucial. With the continuous challenges and rising demands of the elderly, remote and home-based care, the technological innovations in the fields of digital health and health information and communication technologies, such as mobile health, wearable technologies, telemedicine and personalized medicine have transformed the ways of practice and delivery of healthcare in the recent decades [ 2 ]. Wearable technologies have been extensively used in the healthcare sector with multi-purpose applications ranging from patient care to personal health. In clinical and remote care, the applications of wearable devices / sensors, mobile applications, and tracking technologies are of immense importance for the diagnosis, prevention, monitoring, and management of chronic diseases and conditions [ 3 ]. The data generated from the wearable devices / sensors are a cornerstone for healthcare data analytics, especially when it is utilized by latest technologies, such as Artificial Intelligence (AI), Machine Learning (ML), Big Data Intelligence, and Internet of Things (IoT) Systems. The literature has many successful examples of utilization of these data in various branches of medicine, such as oncology, radiology, surgery, geriatrics, rheumatology, neurology, hematology, and cardiology. With the regular ongoing updates, the outcomes of data analytics and their applications are already making a huge impact in transforming and revolutionizing the healthcare industry. In this special issue, we aim to provide new insights on research data analytics and applications of wearable devices / sensors in healthcare by covering wide range of related topics. This issue represents the latest research that spans across 19 countries, 37 institutions and is covered by a total of 28 articles. To make better understanding of the research articles, we have arranged them in an order to show various covered aspects in this field, such as technology integration research, prediction systems, rehabilitation studies, prototype systems, community health studies, detection systems, ergonomics studies, technology acceptance studies, monitoring systems, warning systems, sports studies, clinical systems, feasibility studies, parameters measurement systems, design studies, location based systems, tracking systems, observational studies, risk assessment studies, activity recognition systems, impact measurement systems and systematic review. Sensors 2020 , 20 , 1379; doi:10.3390 / s20051379 www.mdpi.com / journal / sensors 1 Sensors 2020 , 20 , 1379 2. Summary of Special Issue Papers In order to provide a basic overview, we will go through and provide brief summary of all the articles of wearable devices / sensors covered in this issue one by one. Bayo-Monton et al. [ 4 ] provided an implementation of new portable system for remote management of chronic diseases by presenting and evaluating an embedded and scalable distributed system using wearable sensors for the connection of cheap health devices based on prototyping eHealth platforms. The results of their analysis showed that portable devices (p << 0.01) are suitable for supporting the transmission and analysis of biometric signals into scalable telemedicine systems. In an observational study, Thakur et al. [ 5 ] presented a supervised ML-based model for predicting the clinical events during dialysis sessions using data from a non-contact sensor device. The authors found the findings and performance of the ML model quite encouraging and suggested the use of non-contact sensors in clinical settings for monitoring patients’ vital parameters and in early warning solutions for predicting the clinical events. In a study involving patients that recently had knee replacement surgery, Argent et al. [ 6 ] explored and evaluated the feasibility, usability, impact and user experience of an exemplar exercise biofeedback system for orthopedic rehabilitation at home. In order to maximize the engagement and impact, the study incorporated user-centered design approaches by incorporating participants’ evaluation during the design of the system. The findings of the study support the ongoing development and evaluation of sensor-based biofeedback systems, and authors found the system highly usable and e ff ective for patient support and engagement. In a community health study, Martinez et al. [ 7 ] developed a new unsupervised exploratory method for characterizing feature extraction and detecting movement similarity in sleep by using actigraphy signals. The results of statistical analysis showed the potentials of this method for sleep disorders and their link with other conditions. The authors suggested the possible application of proposed approach for the extraction and comparison of sleep movements’ patterns in the field of medicine. Based on a previous work of a Wearable Heat stroke Detection Device (WHDD) [ 8 ] that was used for heat stroke prediction capability for any activity or exercise. Lin et al. [ 9 ] investigated the detailed information analysis and performed static and dynamic experiments for verifying the availability and e ff ectiveness of WHDD experimental subjects. The results of their work demonstrated the superior applicability of the WHDD for predicting the occurrence of heat stroke e ff ectively and ensuring the safety of runners. Using recurrent neural networks (RNNs)-based deep learning models, Luna-Perejon [ 10 ] presented a feasibility study of implementing a wearable system for the detection of falls and its associated risks / hazards in real time through accelerometer signals. Based on the results of the study, the authors recommended RNNs models as an e ff ective method for creation of autonomous wearable fall detection systems in real time. Using a large real-world database of posture data, Stollenwerk et al. [ 11 ] analyzed the postural changes that are induced under postural training in three di ff erent positions, sitting, standing, and hip hinging, and compared the snapshots of unguided-guided posture pair based on features resulted from 2D spine curve geometry. The results showed the novelty of the work in the field of wearable-sensor-based evaluation of spine curves. Vega-Barbas et al. [ 12 ] proposed a precise and pervasive ergonomic platform for accurate assessment of continuous risk and personalized automated coaching by utilizing in-house developed garments and a mobile application. The results of the study demonstrating a good usability score proved the acceptable usability of the platform. The authors expected that wearable technology in the field of ergonomics can have cost e ff ective risk assessment and economical solutions in the future. The study from Lin et al. [ 13 ] presented the design of a wearable cardiac health monitoring platform, implemented it as wearable smart clothing system with multi-channel mechanocardiograms and electrocardiograms measurements, and evaluated the usability of the system using technology acceptance model. The analysis and the results of the study showed the positive attitude of subjects for using this wearable system in providing early risk warnings. Based on deep learning, Lim et al. [ 14 ] presented a coaching assistant method to provide useful information for table tennis practice, and used long short-term memory (LSTM) recurrent neural networks (RNNs) with deep state space model and probabilistic inference to support practice. The promising results provided by this method showed its 2 Sensors 2020 , 20 , 1379 capability in characterizing high-dimensional time series patterns and providing useful information with wearable sensors in table tennis coaching. Lu et al. [ 15 ] developed and tested a new method that combined information from heart rate, respiration, and accelerations measurements to estimate energy expenditure. These data measurements were taken from wearable sensor system and were integrated by neural network based model. The results of the proposed method showed improved accuracy over two existing established methods. The authors suggested that this model along with wearable system could be utilized in both occupational as well as general health applications. Ejupi [ 16 ] investigated the feasibility of wearable textile-based sensors to accurately monitor breathing patterns, develop algorithm to detect talking using ML algorithm, and evaluate the model’s performance with the study participants. The evaluation showed random forest classifier as the best performer in the dataset. The authors suggested that this approach could be used to quantify talking through social interaction and prevent social isolation and loneliness. Using a previously developed inertial measurement unit device based on three sensor [ 17 ], Cesareo et al. [ 18 ] presented an automatic and position-independent algorithm to derive the respiration-induced movement and determine the respiratory rate accurately. The results showed that principal component analysis (PCA) fusion method obtained overall highest performance in terms of breathing frequency estimation, in both supine as well as seated position. The authors suggested that PAC fusion, as dimension-reduction method, can be used to analyze further data in the future. Using wearable technology and ML algorithms, Manjarres et al. [ 19 ] developed a smart physical workload tracking system in real time for simultaneous remote monitoring of people. The established framework was based on the concept of ergonomics to facilitate the work of health professionals and fitness experts. The results of two case studies in real time showed good accuracy and reliability of the system. The authors recommended the future developments by combining ergonomics and ML to predict the physical e ff ort of activities and for injury prevention environments. Nam et al. [ 20 ] used an inertial measurement unit-based motion capture and analysis system to access arm movements. The study provided an important database on the dimensions of workspace and range of motions for arm movements. The validation results showed high accuracy and reliability of the system and emphasized on the importance of designing new exoskeletons for neurorehabilitation purposes. Zhang et al. [ 21 ] examined the relevance of di ff erent conventional physical activity metrics and complexity in the assessment of functional change after exercise intervention in younger and older adults. The findings of the study demonstrated the potential and usefulness of physical activity complexity metrics as compared to conventional metrics in assessment of functional changes for younger and older adults, and recommended them for the feasibility and e ff ectiveness of risk identification and interventions. Hsu et al. [ 22 ] proposed a wearable 12-lead electrocardiogram monitoring system to measure the electrocardiogram (ECG) signals of patients with myocardial ischemia and arrhythmia. The experimental results of the study provided a good ECG signal quality even while walking and detected ECG features of the mentioned patients. The authors suggested the possible usefulness of the proposed system in future mobile ECG monitoring applications. Jayasinghe et al. [ 23 ] investigated and quantified the data received from sensors in di ff erent types of clothing in order to characterize the activities as compared to the body worn sensors’ data. The case study analysis indicated that clothing sensors data correlated well with the body worn sensors data, and classification results from clothing sensors were also promising compared to body-worn sensors. The results of the study showed potentials of this approach in daily monitoring. Allahbakhshi et al. [ 24 ] examined the role of Global Positioning System (GPS) sensors data for detection of physical activity in semi-structured and real-life protocols using participants with wearable devices in a study. The results provided insights in assisting physical activity for future study designs and guidance related to detection of posture and transport related motion activities. Cheung et al. [ 25 ] proposed a novel quantile coarsening analysis (QCA) for reducing the dimension of data from wearable devices and demonstrated the feasibility of this approach in a small cohort of relatively healthy individuals. Because of the versatility of the QCA approach, the authors suggested that it can provide useful analytical tools for data in multi-modal monitoring. By explaining the role of 3 Sensors 2020 , 20 , 1379 actigraphs in personalized health, fitness monitoring and Internet of Medical Things (IoMT) paradigm, Athavale [ 26 ] presented a study utilizing wearable devices to capture and analyze physiological data at home-based health monitoring in an IoMT environment, and proposed a low level encoding scheme to improve actigraphy analysis. In order to ensure that there was no loss of information in encoding process, ML approach was used for the study validation. Based on the dataset Personal RIsk DEtection (PRIDE) [27], a study by Trejo [28] first explored the impact of using dimension reduction techniques and frequency domain features for personal risk detection through correlation matrix and principal component analysis, and then e ffi ciently accelerated the training and classification process of a given classifier for mobile devices. The results of the study were encouraging for timely detection of risk prone situations that can threaten a person’s physical integrity. Yurtman et al. [ 29 ] proposed a methodology to transform the recorded motion sensor sequences to sensor unit orientation unchangeably and incorporated it in pre-processing stage of the standard activity recognition scheme. The results from comparative evaluation of proposed method with the existing state-of-the-art classifiers showed its substantially better output in classifying stationary activities and hence its possible application in various wearable systems. Dutta et al. [ 30 ] used a novel framework to classify and model the physical activities performed by di ff erent participants in a supervised lab-based protocol and then utilized it to identify the physical activities in a free-living setting using the data from wrist worn accelerometers. The positive results of the study demonstrated its application for estimating physical activities in future cohort or intervention studies. In a study, Rosati et al. [ 31 ] compared two di ff erent feature sets for real-time human activity recognition (HAR) applications; one comprising time, frequency, and time-frequency related parameters used in the literature and the other containing only time-related variables linked with biomechanical meaning of acquired signals. The results showed that both set of features can reach high accuracy with support vector machine (SVM) classifier, but the new proposed variables can be easily interpreted and employed for better understanding of the alterations of biomechanical behavior in complex situations. In a study focusing on healthy subjects having normal heart activity, Morelli et al. [ 32 ] investigated the e ff ects of interpolation on time and duration with increasing missing values to assess the interpolation strategy for better results during the estimation of heart rate variability (HRV) features. The results concluded that interpolation in time is the most favorable method for producing better HRV features estimation as compared to interpolation on duration. Fortin-Cote et al. [ 33 ] presented a graphical software for the visualization and preprocessing of raw data received from accelerometer for human posture tracking and assessment. This tool was aimed to provide support for calibration of orientation estimate of inertial measurement units (IMUs) that are used for joint angle measurement. Two case studies were used to demonstrate the usefulness of this open source software. Broadley et al. [ 34 ] presented a systematic review to assess existing methods of evaluating fall detection systems, identify their limitations, and propose improved evaluation methods in the literature. The search results of articles that met the inclusion criteria identified few issues, such as use of small population datasets and inconsistency for performance quantification for these systems. Sensitivity, precision, and F-measures were derived as the most appropriate and robust measures for their realistic performance evaluation. Conflicts of Interest: The authors declare no conflict of interest. References 1. The, L. Global elderly care in crisis. Lancet 2014 , 383 , 927. [CrossRef] 2. Malwade, S.; Abdul, S.S.; Uddin, M.; Nursetyo, A.A.; Fernandez-Luque, L.; Zhu, X.; Cilliers, L.; Wong, C.-P.; Bamidis, P.; Li, Y.-C. Mobile and wearable technologies in healthcare for the ageing population. Comput. Methods Programs Biomed. 2018 , 161 , 233–237. [CrossRef] [PubMed] 3. Sim, I. Mobile Devices and Health. N. Engl. J. Med. 2019 , 381 , 956–968. [CrossRef] [PubMed] 4. Bayo-Monton, J.-L.; Martinez-Millana, A.; Han, W.; Fernandez-Llatas, C.; Sun, Y.; Traver, V. Wearable Sensors Integrated with Internet of Things for Advancing eHealth Care. Sensors 2018 , 18 , 1851. [CrossRef] 4 Sensors 2020 , 20 , 1379 5. Thakur, S.S.; Abdul, S.S.; Chiu, H.-Y.; Roy, R.B.; Huang, P.-Y.; Malwade, S.; Nursetyo, A.A.; Li, Y.-C. Artificial-Intelligence-Based Prediction of Clinical Events among Hemodialysis Patients Using Non-Contact Sensor Data. Sensors 2018 , 18 , 2833. [CrossRef] 6. Argent, R.; Slevin, P.; Bevilacqua, A.; Neligan, M.; Daly, A.; Caulfield, B. Wearable Sensor-Based Exercise Biofeedback for Orthopaedic Rehabilitation: A Mixed Methods User Evaluation of a Prototype System. Sensors 2019 , 19 , 432. [CrossRef] 7. Iglesias Mart í nez, M.E.; Garc í a-Gomez, J.M.; S á ez, C.; Fern á ndez de C ó rdoba, P.; Alberto Conejero, J. Feature Extraction and Similarity of Movement Detection during Sleep, Based on Higher Order Spectra and Entropy of the Actigraphy Signal: Results of the Hispanic Community Health Study / Study of Latinos. Sensors 2018 , 18 , 4310. [CrossRef] 8. Chen, S.T.; Lin, S.S.; Lan, C.W.; Hsu, H.Y. Design and Development of a Wearable Device for Heat Stroke Detection. Sensors 2017 , 18 , 17. [CrossRef] 9. Lin, S.-S.; Lan, C.-W.; Hsu, H.-Y.; Chen, S.-T. Data Analytics of a Wearable Device for Heat Stroke Detection. Sensors 2018 , 18 , 4347. [CrossRef] 10. Luna-Perej ó n, F.; Dom í nguez-Morales, M.J.; Civit-Balcells, A. Wearable Fall Detector Using Recurrent Neural Networks. Sensors 2019 , 19 , 4885. [CrossRef] 11. Stollenwerk, K.; Müller, J.; Hinkenjann, A.; Krüger, B. Analyzing Spinal Shape Changes During Posture Training Using a Wearable Device. Sensors 2019 , 19 , 3625. [CrossRef] [PubMed] 12. Vega-Barbas, M.; Diaz-Olivares, J.A.; Lu, K.; Forsman, M.; Seoane, F.; Abtahi, F. P-Ergonomics Platform: Toward Precise, Pervasive, and Personalized Ergonomics using Wearable Sensors and Edge Computing. Sensors 2019 , 19 , 1225. [CrossRef] [PubMed] 13. Lin, W.-Y.; Ke, H.-L.; Chou, W.-C.; Chang, P.-C.; Tsai, T.-H.; Lee, M.-Y. Realization and Technology Acceptance Test of a Wearable Cardiac Health Monitoring and Early Warning System with Multi-Channel MCGs and ECG. Sensors 2018 , 18 , 3538. [CrossRef] [PubMed] 14. Lim, S.-M.; Oh, H.-C.; Kim, J.; Lee, J.; Park, J. LSTM-Guided Coaching Assistant for Table Tennis Practice. Sensors 2018 , 18 , 4112. [CrossRef] [PubMed] 15. Lu, K.; Yang, L.; Seoane, F.; Abtahi, F.; Forsman, M.; Lindecrantz, K. Fusion of Heart Rate, Respiration and Motion Measurements from a Wearable Sensor System to Enhance Energy Expenditure Estimation. Sensors 2018 , 18 , 3092. [CrossRef] 16. Ejupi, A.; Menon, C. Detection of Talking in Respiratory Signals: A Feasibility Study Using Machine Learning and Wearable Textile-Based Sensors. Sensors 2018 , 18 , 2474. [CrossRef] 17. Cesareo, A.; Gandolfi, S.; Pini, I.; Bi ffi , E.; Reni, G.; Aliverti, A. A novel, low cost, wearable contact-based device for breathing frequency monitoring. In Proceedings of the 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Seogwipo, Korea, 11–15 July 2017; pp. 2402–2405. 18. Cesareo, A.; Previtali, Y.; Bi ffi , E.; Aliverti, A. Assessment of Breathing Parameters Using an Inertial Measurement Unit (IMU)-Based System. Sensors 2018 , 19 , 88. [CrossRef] 19. Manjarres, J.; Narvaez, P.; Gasser, K.; Percybrooks, W.; Pardo, M. Physical Workload Tracking Using Human Activity Recognition with Wearable Devices. Sensors 2019 , 20 , 39. [CrossRef] 20. Nam, H.S.; Lee, W.H.; Seo, H.G.; Kim, Y.J.; Bang, M.S.; Kim, S. Inertial Measurement Unit Based Upper Extremity Motion Characterization for Action Research Arm Test and Activities of Daily Living. Sensors 2019 , 19 , 1782. [CrossRef] 21. Zhang, W.; Schwenk, M.; Mellone, S.; Paraschiv-Ionescu, A.; Vereijken, B.; Pijnappels, M.; Mikolaizak, A.S.; Boulton, E.; Jonkman, N.H.; Maier, A.B.; et al. Complexity of Daily Physical Activity Is More Sensitive Than Conventional Metrics to Assess Functional Change in Younger Older Adults. Sensors 2018 , 18 , 2032. [CrossRef] 22. Hsu, C.-C.; Lin, B.-S.; He, K.-Y.; Lin, B.-S. Design of a Wearable 12-Lead Noncontact Electrocardiogram Monitoring System. Sensors 2019 , 19 , 1509. [CrossRef] 23. Jayasinghe, U.; Harwin, W.S.; Hwang, F. Comparing Clothing-Mounted Sensors with Wearable Sensors for Movement Analysis and Activity Classification. Sensors 2019 , 20 , 82. [CrossRef] [PubMed] 24. Allahbakhshi, H.; Conrow, L.; Naimi, B.; Weibel, R. Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection. Sensors 2020 , 20 , 588. [CrossRef] [PubMed] 5 Sensors 2020 , 20 , 1379 25. Cheung, Y.K.; Hsueh, P.-Y.S.; Ensari, I.; Willey, J.Z.; Diaz, K.M. Quantile Coarsening Analysis of High-Volume Wearable Activity Data in a Longitudinal Observational Study. Sensors 2018 , 18 , 3056. [CrossRef] [PubMed] 26. Athavale, Y.; Krishnan, S. A Device-Independent E ffi cient Actigraphy Signal-Encoding System for Applications in Monit