Using Wearable Sensors to Measure Motor Abilities Following Stroke

Motor abilities of stroke survivors are often severely affected. Post-stroke rehabilitation is guided by the use of clinical assessments of motor abilities. Clinical assessment scores can be predicted by models based on features extracted from the wearable

Using Wearable Sensors to Measure Motor Abilities following Stroke Todd Hester1, Richard Hughes1, Delsey M. Sherrill1, Bethany Knorr2,

Metin Akay3, Joel Stein1, and Paolo Bonato1,4

1Dept of PM&R, Harvard Medical School, Spaulding Rehabilitation Hospital, Boston MA 2Thayer School of Engineering, Dartmouth College, Hanover NH

3Dept of Bioengineering, Arizona State University, Tempe AZ 4The Harvard-MIT Division of Health Science and Technology, Cambridge MA

tahester@http://doc.xuehai.net, rhughes1@http://doc.xuehai.net, delsey.sherrill@http://doc.xuehai.net,

bethany.r.knorr@dartmouth.edu, metin.akay@asu.edu, jstein@http://doc.xuehai.net,

pbonato@http://doc.xuehai.net

Abstract

Motor abilities of stroke survivors are often severely affected. Post-stroke rehabilitation is guided by the use of clinical assessments of motor abilities. Clinical assessment scores can be predicted by models based on features extracted from the wearable sensor data. Wearable sensors would allow monitoring of subjects in the home and provide accurate assessments to guide the rehabilitation process. We propose the use of a wearable sensor system to assess the motor abilities of stroke victims. Preliminary results from twelve subjects show the ability of this system to predict clinical scores of motor abilities.

Keywords: Wearable Sensors, Stroke, Clinical Assessment

1. Introduction

Approximately 700,000 people are affected by stroke each year in the United States and about 275,000 die from stroke each year [1]. Strokes affect a person’s cognitive, language, perceptual, sensory, and motor abilities [2]. More than 1,100,000 Americans have reported difficulties with functional limitations following stroke [3]. Recovery from stroke is a long process that continues beyond the hospital stay and into the home setting. The rehabilitation process is guided by clinical assessments of motor abilities.

Accurate assessment of motor abilities is important in selecting the best therapies for stroke survivors. These assessments are based on observations of subjects’ motor behavior using standardized clinical rating scales. The accuracy and consistency of observational assessments may vary greatly across clinicians [4]. Wearable sensors could be used to provide more accurate measures or could be used in addition to observational clinical tools. Wearable systems have the ability to measure motor behavior at home and for longer periods than could be observed in a clinical setting. Accelerometers can capture specific patterns of movement relating to motor disabilities. We propose that wearable systems can be used to predict clinical scores of motor abilities and we present an initial analysis of data demonstrating an association between accelerometer data and clinical scores.

2. Methods

Twelve subjects who had a stroke within the past 2 to 24 months were recruited for the study. Each subject was evaluated by a clinician using standardized clinical motor performance scales, including the Fugl-Meyer Assessment of Sensorimotor Recovery after Stroke, Chedoke-McMaster Stroke Assessment, Wolf Motor Performance Test, and the Reaching Performance Scale. These scales measure dimensions of upper limb motor behavior including movement quality, stage of motor recovery, use of compensatory movement strategies, and the ability to perform functional tasks. All testing was performed at Spaulding Rehabilitation Hospital. Subjects provided informed consent approved by the hospital’s research review board. Accelerometers were attached to the affected arm and the trunk (Figure 1).

第1页下一页