Automated evaluation of physical therapy exercises using multi-template dynamic time warping on wearable sensor signals

We develop an autonomous system to detect and evaluate physical therapy exercises using wearable motion sensors. We propose the multi-template multi-match dynamic time warping (MTMM-DTW) algorithm as a natural extension of DTW to detect multiple occurrences of more than one exercise type in the recording of a physical therapy session. While allowing some distortion (warping) in time, the algorithm provides a quantitative measure of similarity between an exercise execution and previously recorded templates, based on DTW distance. It can detect and classify the exercise types, and count and evaluate the exercises as correctly/incorrectly performed, identifying the error type, if any. To evaluate the algorithm's performance, we record a data set consisting of one reference template and 10 test executions of three execution types of eight exercises performed by five subjects. We thus record a total of 120 and 1200 exercise executions in the reference and test sets, respectively. The test sequences also contain idle time intervals. The accuracy of the proposed algorithm is 93.46% for exercise classification only and 88.65% for simultaneous exercise and execution type classification. The algorithm misses 8.58% of the exercise executions and demonstrates a false alarm rate of 4.91%, caused by some idle time intervals being incorrectly recognized as exercise executions. To test the robustness of the system to unknown exercises, we employ leave-one-exercise-out cross validation. This results in a false alarm rate lower than 1%, demonstrating the robustness of the system to unknown movements. The proposed system can be used for assessing the effectiveness of a physical therapy session and for providing feedback to the patient.     

Swimming performance and technique evaluation with wearable acceleration sensors

We are working towards a wearable computing system called SwimMaster, that will support swimmers in achieving their desired exercise goals by monitoring their swimming performance and technique and providing the necessary feedback. In this article, we describe our methods to extract the most relevant swimming performance and technique parameters from acceleration sensors worn at the wrist and at the back. We analyze the data and our methods with a SwimModel. Finally, we present the results of our evaluation studies with 18 swimmers—seven elite, eight recreational and three occasional swimmers.     

Ergonomic evaluation of a wearable assistive device for overhead work

Overhead work is an important risk factor for upper extremity (UE) musculoskeletal disorders. We examined the potential of a mechanical arm and an exoskeletal vest as a wearable assistive device (WADE) for overhead work. Twelve participants completed 10 minutes of simulated, intermittent overhead work, using each of three payloads (1.1, 3.4 and 8.1 kg) and with/without the WADE. Ratings of perceived discomfort (RPDs) and electromyography (EMG) were obtained for the upper arms, shoulders and low back. Using the WADE, UE RPDs decreased by ～50% with the heavier payloads, whereas smaller (～25%) and non-significant increases in low-back RPDs were found and were relatively independent of payload. Changes in RPDs with WADE use were consistent with physical demands indicated by EMG, though EMG-based differences in fatigue were less apparent. Participants generally preferred using the WADE, particularly with heavier payloads. These results supported the potential utility of a WADE as an intervention for overhead work.     

Implementation of various control algorithms for hand rehabilitation exercise using wearable robotic hand

In this paper, the control algorithms for strength exercise using wearable robotic hand are reviewed and the experimental results are analyzed and discussed. The SNU Exo-Glove is a soft exoskeleton that actuates motor function in disabled hands. This new type of device comprises a jointless simple mechanical structure and is actuated with wires. The strength exercise algorithms include isotonic, isokinetic, and impedance control exercises. An electromyography (EMG) regulation algorithm is proposed to limit the maximum level of activation of the muscles to prevent injury of the muscles and joints. The tension of the wire and the sEMG signal are analyzed to validate the effectiveness of rehabilitation with SNU Exo-Glove.     

Systematic usability evaluation on two harnesses for a wearable chairless exoskeleton

Harnesses are used in wearable robots in safety, sports, medical, and industrial environments. Wearable robot harnesses for lower extremities should be designed to prevent musculoskeletal disorders caused by excessive strain on the spine and lower extremities when users wear these robots for prolonged periods. Therefore, this study compares and analyzes the usability of the two harnesses—the former (old design) and the latter (new design)—for the Hyundai chairless exoskeleton (CEX). In total, 27 participants took part in this study. The wearability, stability, and overall wearing satisfaction of the former and the latter harnesses were compared. We conducted an experiment where the participants wore the former and the latter harnesses while completing a series of tasks that resembled the ones performed by the workers on a front-end module assembling line of an automobile factory. After the experiment, the participants took part in a survey and semi-structured interviews. The data was analyses by t-test, correlation, and regression analyses. The survey results demonstrated that for the latter harness the cushioning sensation (thigh r = 0.539; calf r = 0.516) had the strongest effect on the overall wearing satisfaction. In addition, the latter harness scored higher in overall wear satisfaction than the previous harness. This study was able to verify the difference between the new and old designs according to the usability evaluation of the harness. The results of this study are considered to be helpful in design to improve the usability of the harness in the future.     

An industrial evaluation of DRAM tests

DRAM production tests are currently necessary to reach a defect-per-million level that approaches the single-digit numbers. This implies that a single memory test is insufficient; rather, a set of tests is necessary. This application of 40 well-known memory tests to 1,896 1-Mbyte /spl times/ 4 DRAM chips, used up to 48 different stress combinations with each test. The results show the importance of selecting the right stress combination, and that the theoretically better tests - those covering more different functional faults - also have higher fault coverage.     

Introducing time in reaction systems

Reaction systems are a formal model of interactions between biochemical reactions. The main observation underlying the formulation of this model is that such interactions are based on two basic mechanisms: facilitation and inhibition. This paper continues the investigation of reaction systems, and in particular, it proposes a formal framework for introducing time into reaction systems. Within this framework one can formally define and investigate notions such as reaction times, creation times of compounds, their life spans, etc.     

Application level performance evaluation of wearable devices for stress classification with explainable AI

Stress has become one of the most prominent problems of modern societies and a key contributor to major health issues. Dealing with stress effectively requires detecting it in real-time, informing the user, and giving instructions on how to manage it. Over the past few years, wearable devices equipped with biosensors that can be utilized for stress detection have become increasingly popular. Since they come with various designs and technologies and acquire biosignals from different body locations, choosing a suitable device for a particular application has become a challenge for researchers and end-users. This study compares seven common wearable biosensors for stress detection applications. This was accomplished by collecting physiological sensor data during Baseline, Stress, Recovery, and Cycling sessions from 32 participants. Machine learning algorithms were used to classify four stress classes, and the results obtained from all wearables were compared. Following this, a state-of-the-art explainable artificial intelligence method was employed to clarify our models’ predictions and investigate the influence different features have on the models’ outputs. Despite the results showing that ECG wearables perform slightly better than the rest of the devices, adding a second biosignal (EDA) improved the results significantly, tipping the balance toward multisensor wearables. Finally, we concluded that although the output results of each model can be affected by various factors, in most cases, there is no significant difference in the accuracy of stress detection by different wearables. However, the decision to select a particular wearable for stress detection applications must be made carefully considering the trade-off between the users’ expectations and preferences and the pros and cons of each device.     

A low power and real-time hardware recurrent neural network for time series analysis on wearable devices

The research presented in this paper addresses the exploitation of Deep Learning methods on wearable devices. We propose a hardware architecture capable of analyzing time series signals through a Recurrent Neural Network implemented on FPGA technology. This architecture has been validated using a real dataset, which includes three-axial accelerometer data acquired by a wearable device used for fall detection. The experiments have been conducted considering different devices and demonstrates that the proposed hardware architecture outperforms the state of the art solutions both in terms of processing time and power consumption. Indeed, the proposed architecture is real-time compliant in the elaboration of the fall detection dataset adopted for the validation. The power consumption is in the order of dozens μW. Finally, futher functionalities could be added in the same chip since the resource usage is low.     

生理参数热舒适评价中的穿戴式数据采集模块设计

基于生理参数的热舒适评价是热环境质量评估的新方法。为了提高该方法的实用性,设计了一种穿戴式、多生理参数采集模块。以Zig Bee技术为核心,采用聚偏氟乙烯(PVDF)薄膜传感器、半导体温度传感器和Ω/V变换原理分别检测脉搏、皮肤温度和皮肤电阻。实验结果表明:模块可靠采集腕部的皮肤温度和皮肤电阻,并从脉搏信号获取心搏间期序列,数据发送正常。该模块适合在自然状态下采集生理信号,有利于研究和应用多种生理参数综合评价热舒适性。     

Advanced mobile and wearable systems

The recent spectacular progress in the microelectronic, information, communication, material and sensor technologies created a big stimulus towards development of smart communicating cyber-physical systems (CPS) and Internet of Things (IoT). CPS and IoT are undergoing an explosive growth to a large degree related to advanced mobile systems like smart automotive and avionic systems, mobile robots and wearable devices. The huge and rapidly developing markets of sophisticated mobile cyber-physical systems represent great opportunities, but these opportunities come with a price of unusual system complexity, as well as, stringent and difficult to satisfy requirements of many modern applications. Specifically, smart cars and various wearable systems to a growing degree involve big instant data from multiple complex sensors or other systems, and are required to provide continuous autonomous service in a long time. In consequence, they demand a guaranteed (ultra-)high performance and/or (ultra-)low energy consumption, while requiring a high reliability, safety and security. To adequately address these demands, sophisticated embedded computing and embedded design technologies are needed. After an introduction to modern mobile systems, this paper discusses the huge heterogeneous area of these systems, and considers serious issues and challenges in their design. Subsequently, it discusses the embedded computing and design technologies needed to adequately address the issues and overcome the challenges in order to satisfy the stringent requirements of the modern mobile systems.     

Implementation and evaluation of failsafe computer-controlled systems

Fault tolerance is considered as the ideal candidate not only for the failsafe system, but also for the reduction of the failure effect and the continuation of the remaining task. The proposed fault-tolerant architecture includes the software design of error detection and diagnosis, as well as the error recovery. Multi-tasks are managed by a computer with a redundant one or by multi-computers with redundancy from each other are employed and evaluated in terms of the reliability and effectiveness. The executing program is supervised by the watchdog, which warns a failure condition of the software program in case that the execution time of each subprogram runs over its default value. The computers are mutually sending the heartbeat signals periodically. The message of the receiving signal indicates whether the system is under failure. The entire detection of the heartbeat function is supervised by the time daemon to ensure that the fault recovery is feasible. Once a computer is failed, the other computer immediately takes over its position and accomplishes the remaining task.     

Implementation and evaluation of a microthread architecture

Future many-core processor systems require scalable solutions that conventional architectures currently do not provide. This paper presents a novel architecture that demonstrates the required scalability. It is based on a model of computation developed in the AETHER project to provide a safe and composable approach to concurrent programming. The model supports a dynamic approach to concurrency that enables self-adaptivity in any environment so the model is quite general. It is implemented here in the instruction set of a dynamically scheduled RISC processor and many such processors form a microgrid. Binary compatibility over arbitrary clusters of such processors and an inherent scalability in both area and performance with concurrency exploited make this a very promising development for the era of many-core chips. This paper introduces the model, the processor and chip architecture and its emulation on a range of computational kernels. It also estimates the area of the structures required to support this model in silicon.     

反应时和动态视野测量研究

以动态视野的测量为例,分析了反应时、简单反应时的关系,指出传统动态视野测量中由于简单反应时的影响而不能准确评价人的动态视野.通过减除反应时中包含的简单反应时,重新定义了动态视野.通过简单反应时实验和动态视野测量,表明了这种新定义的正确性.巧妙的实验程序设计思路和自行研发的测量装置,都是反应时和动态视野测量准确性的保证.     

两种线程池的实现和性能评价

线程池被广泛地应用在中间件如Web应用服务器、事务监控器等的实现中.实现线程池的方法主要有半同步/半异步(Half-Sync/Half-Asyn)模式和领导者/跟随者(Leader/Follower)模式.实现了这两种线程池,从理论上分析了它们的性能,并进行了性能实验.实验数据表明Leader/Follower模式具有较高的性能优势.     

Context-awareness in wearable and ubiquitous computing

A common focus shared by researchers in mobile, ubiquitous and wearable computing is the attempt to break away from the traditional desktop computing paradigm. Computational services need to become as mobile as their users. Whether that service mobility is achieved by equipping the user with computational power or by instrumenting the environment, all services need to be extended to take advantage of the constantly changing context in which they are accessed. This paper will report on work carried out by the Future Computing Environments Group at Georgia Tech to provide infrastructure for context-aware computing. We will describe some of the fundamental issues involved in contextaware computing, solutions we have generated to provide a flexible infrastructure and several applications that take advantage of context awareness to allow freedom from traditional desktop computing.