An Activity Recognition Model Using Inertial Sensor Nodes in a Wireless Sensor Network for Frozen Shoulder Rehabilitation ExercisesSensors


Hsueh-Chun Lin, Shu-Yin Chiang, Kai Lee, Yao-Chiang Kan
Electrical and Electronic Engineering / Analytical Chemistry / Atomic and Molecular Physics, and Optics / Biochemistry


Service-Oriented Node Scheduling Scheme for Wireless Sensor Networks Using Markov Random Field Model

Hongju Cheng, Zhihuang Su, Jaime Lloret, Guolong Chen

A Localization Algorithm nin Wireless Sensor Networks Using a Mobile Beacon Node

Zhuhong you, Max Q.-H. Meng, Huawei Liang, Shuai Li, Yangming Li, Wanming Chen, Yajin Zhou, Shifu Miao, Kai Jiang, Qinglei Guo

Fault Node Recovery Algorithm for a Wireless Sensor Network

Hong-Chi Shih, Jiun-Huei Ho, Bin-Yih Liao, Jeng-Shyang Pan

Model Identification for Wireless Sensor Networks

Anand Oka, Lutz Lampe

Sensor nodes height selection method for wireless sensor networks in wide area

ByungHee Son, KwangJin Kim, JungTae Seo, ByungHo Joo, YoungWan Choi


Sensors 2015, 15, 2181-2204; doi:10.3390/s150102181 sensors

ISSN 1424-8220


An Activity Recognition Model Using Inertial Sensor Nodes in a Wireless Sensor Network for Frozen Shoulder

Rehabilitation Exercises

Hsueh-Chun Lin 1, Shu-Yin Chiang 2, Kai Lee 3 and Yao-Chiang Kan 3,* 1 Health Risk Management Department, China Medical University, 91 Hsueh-Shih Rd.,

Taichung 40402, Taiwan; E-Mail: 2 Department of Information and Telecommunications Engineering, Ming Chuan University, 5 De-Ming Rd., Gui Shan, Taoyuan 333, Taiwan; E-Mail: 3 Department of Communications Engineering, Yuan Ze University, 135 Yuan-Tung Rd., Chung-Li,

Taoyuan 32003, Taiwan; E-Mail: * Author to whom correspondence should be addressed; E-Mail:;

Tel.: +886-3-463-8800 (ext. 7011); Fax: +886-3-455-4264.

Academic Editor: Nauman Aslam

Received: 8 July 2014 / Accepted: 12 January 2015 / Published: 19 January 2015

Abstract: This paper proposes a model for recognizing motions performed during rehabilitation exercises for frozen shoulder conditions. The model consists of wearable wireless sensor network (WSN) inertial sensor nodes, which were developed for this study, and enables the ubiquitous measurement of bodily motions. The model employs the back propagation neural network (BPNN) algorithm to compute motion data that are formed in the WSN packets; herein, six types of rehabilitation exercises were recognized. The packets sent by each node are converted into six components of acceleration and angular velocity according to three axes. Motor features such as basic acceleration, angular velocity, and derivative tilt angle were input into the training procedure of the BPNN algorithm. In measurements of thirteen volunteers, the accelerations and included angles of nodes were adopted from possible features to demonstrate the procedure. Five exercises involving simple swinging and stretching movements were recognized with an accuracy of 85%–95%; however, the accuracy with which exercises entailing spiral rotations were recognized approximately 60%. Thus, a characteristic space and enveloped spectrum improving derivative features were suggested to enable identifying customized parameters. Finally, a real-time monitoring interface was developed for practical implementation. The


Sensors 2015, 15 2182 proposed model can be applied in ubiquitous healthcare self-management to recognize rehabilitation exercises.

Keywords: back propagation neural network (BPNN); frozen shoulder; inertial sensor node (ISN); rehabilitation activity; ubiquitous health care (UHC); wireless sensor network (WSN) 1. Introduction

The rapid innovations in information technology have promoted studies investigating human movements. Techniques for detecting bodily motions are widely applied in healthcare to ubiquitously monitor and rehabilitate disabled patients. Previous studies on motion analysis have involved tracking parts of a moving body by calculating data on image sequences of bodily movements [1]. Many vision-based approaches were implemented to classify large scale bodily motions, including movements of the head, arms, torso, or legs [2]. Computational algorithms have enabled image analyses of bodily gestures and were used in supporting the assistant interfaces, such as healthcare monitoring systems [3].

In addition, non-imaged tracking procedures were employed in systems for monitoring bodily motions [4].

Both types of system are used in ubiquitous healthcare (UHC) and facilitate maintaining traceable records that can be followed whenever and wherever patients require treatment [5]. According to privacy policies respecting patients opposed to public exhibition, non-imaged and noninvasive devices are appropriate for use in UHC programs. In the recent decade, the medical laboratory instruments mobilized with the ZigBee protocol have been subjected to experiments that involved remotely analyzing cardiac data on patients [6].

Rehabilitation through physical therapy is necessary for patients who exhibit limited ability in limb or bodily movement because of conditions such as hemiplegia and adhesive capsulitis [7]. A self-managed rehabilitation program for patients with frozen shoulder is a feasible UHC service. Frozen shoulder is a symptom of adhesive capsulitis and can cause stiffness and pain in the shoulder joint, reducing the ability to engage in a range of multidirectional motions [8,9]. The typical physical therapy needs to assign specific exercises for relaxing the restriction of capsulitis motion [10,11]. For example,

Codman’s pendulum exercise is used to train the patients who must abduct the arm through gravity and keep the supraspinatus relaxed without a fulcrum. This exercise extends the mobility of the shoulder joint by stretching and rotating the arms [12]. Many studies and clinical trials have suggested beneficial exercises for patients who require long-term rehabilitation to relieve pain; these exercises help by increasing the ranges of the joint motions such as forward flexion, elevation, abduction, and rotation [13].

Physiatrists interested in monitoring the daily rehabilitation progress of their patients have used customized programs [14] involving self-completed questionnaires, camcorders, and electromagnetic sensors in hospitals to monitor and manage the rehabilitation process [15,16]. An automatic process would be more efficient and cost effective for managing routine rehabilitation programs [17].

Wireless networks and telecommunication technology have enabled modern healthcare service providers to ubiquitously monitor patients who require self-management of regular rehabilitation at home. According to a UHC program, physiatrists can prescribe physical therapy supported by specific facilities and engage in daily supervision on the outpatients for a defined period of time [18–20].

Sensors 2015, 15 2183