Human Gait Monitoring Using Wearable Fabric Based Strain Sensors And Deep Supervised Learning

Continuous lower body monitoring is an important step for real-time feedback training of runners and in-home rehabilitation assessment. Optical motion capture systems are the gold standards for gait analysis, but they are spatially limited to laboratories. Recently, wearable sensors have gained attention as unobtrusive methods to analyze gait metrics and health conditions. In this study, a wearable system capable of estimating lower body joint angles in sagittal, frontal, and transverse planes during gait was developed. A prototype with fiber strain sensors was fabricated. The positions of the sensors on the pelvis were optimized using a genetic algorithm. A cohort of ten people completed 15 minutes of running at 5 different speeds for gait analysis by our prototype device. The joint angles were estimated by a deep convolutional neural network in inter- and intra-participant scenarios. In intra-participant tests, root mean squared error (RMSE) and normalized root mean squared error (NRMSE) of less than 2.2° and 5.3 %, respectively, were obtained for hip, knee, and ankle joints in sagittal, frontal, and transverse planes. The RMSE and NRMSE in inter-participant tests were less than 6.4° and 10%, respectively, in the sagittal plane. The accuracy of this device and methodology could yield potential applications as a soft wearable device for gait monitoring.

Wearable Systems Based Gait Monitoring and Analysis provides a thorough overview of wearable gait monitoring techniques and their use in health analysis. The text starts with an examination of the relationship between the human body’s physical condition and gait, and then introduces and explains nine mainstream sensing mechanisms, including piezoresistive, resistive, capacitive, piezoelectric, inductive, optical, air pressure, EMG and IMU-based architectures. Gait sensor design considerations in terms of geometry and deployment are also introduced. Diverse processing algorithms for manipulating sensors outputs to transform raw data to understandable gait features are discussed. Furthermore, gait analysis-based health monitoring demonstrations are given at the end of this book, including both medical and occupational applications. The book will enable students of biomedical engineering, electrical engineering, signal processing, and ergonomics and practitioners to understand the medical and occupational applications of engineering-based gait analysis and falling injury prevention methods.

This book is the second edition of the one originally published in 2017. The original publication features the discovery of numerous novel applications for the use of smartphones and portable media devices for the quantification of gait, reflex response, and an assortment of other concepts that constitute first-in-the-world applications for these devices. Since the first edition, numerous evolutions involving the domain of wearable and wireless systems for healthcare have transpired warranting the publication of the second edition. This volume covers wearable and wireless systems for healthcare that are far more oriented to the unique requirements of the biomedical domain. The paradigm-shifting new wearables have been successfully applied to gait analysis, homebound therapy, and quantifiable exercise. Additionally, the confluence of wearable and wireless systems for healthcare with deep learning and neuromorphic applications for classification is addressed. The authors expect that these significant developments make this book valuable for all readers.

This book presents state-of-the-art technologies, trends and applications with a focus on the healthcare domain for ultra-wideband (3.1–10.6 GHz) and 60 GHz (57–66 GHz) wireless communication systems. Due to various key features such as miniaturized antenna design, low power, high data rate, less effects on the human body, relatively less crowded spectrum, these technologies are becoming popular in various fields of biomedical applications and day-to-day life. The book highlights various aspects of these technologies related to body-centric communication, including antenna design requirements, channel modeling and characterization for WBANs, current fabrication and antenna design strategies for textile, flexible and implanted antennas. Apart from the general requirements and study related to these frequency bands, various application specific topics such as localization and tracking, physical activity recognition and assessment, vital sign monitoring and medical imaging are covered in detail. The book concludes with the glimpses of future aspects of the UWB and 60 GHz technology which includes IoT for healthcare and smart living, novel antenna materials and application of machine learning algorithms for overall performance enhancement.

The embedded sensors in widely used smartphones, wearable devices and smart environments make the sensor data stream of human activity more accessible. With the development of deep neural networks, extensive studies have been conducted using deep learning methods to extract useful information from the sensor data to recognize the human activity, identify the person, or monitor the health condition of the person. However, applying deep neural networks to the sensor based human activity analysis task remains a challenging research problem in ubiquitous computing. Some of the reasons are: (i) The majority of the acquired data has no labels; (ii) Most of the previous works in activity and sensor stream analysis have been focusing on one aspect of the data, e.g. only recognizing the type of the activity or only identifying the person who performed the activity; (iii) Segmenting a continuous sensor stream and preserving the completeness of each human activity is difficult. In this dissertation, various deep learning techniques have been studied to address these problems in a weakly supervised, unsupervised, or semi-supervised manner. All the developed techniques use deep learning networks to learn embedding spaces in which activities group and thus classifiers can be trained efficiently. For this, both siamese network architectures for weakly supervised data and autoencoder-type networks for unsupervised techniques are learned and combined.

Technological advancements in healthcare can contribute unquestionably in reducing healthcare strains by ensuring clinicians, doctors and other medical staff operate and conduct their daily activities more efficiently in the hospital vicinity. Since the turn of the 21st century, Human Activity Recognition (HAR) has undergone significant research in the healthcare domain. HAR utilised with powerful technologies can benefit remote patient monitoring, the elderly, patients suffering from chronic illness and ambient assisted living. Human activity recognition has shown to be effective in benefiting clinicians in the treatment and remote monitoring of patients. This field is not only vital for diagnosis and treatment, but also an assessment of how likely a medical patient will fall ill or die from certain diseases or health problems. To show the great importance of activity recognition in the health sector, analytically driving an improvement in accuracy in classifying patients' activities improves the relationship of patients and clinicians as well as reducing the possibility of a fatality. With Artificial Intelligence at the forefront of its revolutionary capabilities, a bright future is in store if we can implement it beneficially into our healthcare service. This book reveals how.

Smart wearable electronic devices capable of information exchanging (such as human-machine interfaces) have developed into key carriers for the interconnection, intercommunication, and interaction between humans and machines. Multimodal electronic textiles that incorporate multifunctional sensors into daily clothing are an emerging technology to realize smart wearable electronics. This has greatly advanced human-machine interface technology by bridging the gap between wearing comfort and traditional wearable electronic devices, which will facilitate the rapid development and wide application of natural human-machine interfaces. In this article, we provide a comprehensive summary of the latest research progress on multimodal electronic textiles for intelligent human-machine interfaces. Firstly, we introduce the most representative electronic textile manufacturing strategies in terms of functional fiber preparation and multimodal textile forming. Then, we explore the multifunctional sensing capability of multimodal electronic textiles and emphasize their advanced applications in intelligent human-machine interfaces. Finally, we present new insights on the future research directions and the challenges faced in practical applications of multimodal electronic textiles.

This book provides a collection of comprehensive research articles on data analytics and applications of wearable devices in healthcare. This Special Issue presents 28 research studies from 137 authors representing 37 institutions from 19 countries. To facilitate the understanding of the research articles, we have organized the book to show various aspects covered in this field, such as eHealth, technology-integrated research, prediction models, rehabilitation studies, prototype systems, community health studies, ergonomics design systems, technology acceptance model evaluation studies, telemonitoring systems, warning systems, application of sensors in sports studies, clinical systems, feasibility studies, geographical location based systems, tracking systems, observational studies, risk assessment studies, human activity recognition systems, impact measurement systems, and a systematic review. We would like to take this opportunity to invite high quality research articles for our next Special Issue entitled “Digital Health and Smart Sensors for Better Management of Cancer and Chronic Diseases” as a part of Sensors journal.

Remote gait monitoring system plays an important role in improving the process of gait rehabilitation while patients are not supervised by the physical therapists outside of the clinics. It can benefit patients, providers, and payers with low cost, high accuracy, real-time accessibility, detailed exercise reports, preventive treatment, and data privacy. The patients can use the system to record all the exercise during the recovery process while the providers can access the patients' recovery process with more convenience remotely. The remote gait monitoring system can also improve providers like insurance companies to create a more customized health plan and establish quantitative regulation. In this study, three gait parameters are discussed, including stride length, stride frequency, and stride velocity. A classification model was used to detect stationary epoch and non-stationary epoch to extract each stride sample. The accuracy of the classification model achieves 99.3%, which shows high reliability for detecting motion change points. Then, a mobile application on stride length estimation was developed with an OpenMP-based distributed deep learning optimized system (DDOS). The DDOS system used a convolutional neural network (CNN) to estimate the stride length of each stride sample. The system has the advantages of incremental learning, time flexibility, and customization, which can be used for multiple users at any place during the same time. The experiment results show a high accuracy with stride length estimation. OpenMP was used to accelerate operation time since the training process of CNN is time-consuming.

Gait recognition systems have gained tremendous attention due to its potential applications in healthcare, criminal investigation, sports biomechanics, and so forth. A new solution to gait recognition tasks can be provided by wearable sensors integrated in wearable objects or mobile devices. In this research a sock prototype designed with embedded soft robotic sensors (SRS) is implemented to measure foot ankle kinematic and kinetic data during three experiments designed to track participants’ feet ankle movement. Deep learning and statistical methods have been employed to model SRS data against Motion capture system (MoCap) to determine their ability to provide accurate kinematic and kinetic data using SRS measurements. In the first study, the capacitance of SRS related to foot-ankle basic movements was quantified during the gait movements of twenty participants on a flat surface and a cross-sloped surface. I have conducted another study regarding kinematic features in which deep learning models were trained to estimate the joint angles in sagittal and frontal planes measured by a MoCap system. Participant-specific models were established for ten healthy subjects walking on a treadmill. The prototype was tested at various walking speeds to assess its ability to track movements for multiple speeds and generalize models for estimating joint angles in sagittal and frontal planes. The focus of the last study is measuring the kinetic features and the goal is determining the validity of SRS measurements, to this end the pressure data measured with SRS embedded into the sock prototype would be compared with the force plate data.