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

Abstract

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.

Revisiting Human Action Recognition: Personalization vs. Generalization

“…Differently, in [17], the performance of checking the correct execution of gymnastics sharply falls when the subject under testing is excluded from the training phase. A similar trend was registered by [18] and [19] for computer assisted rehabilitation tasks, as well as by [20] which performed a theoretical dissertation about within-subject and across-subjects noise using wearable motion sensors. Globally, [15], [16], [17], [18], [19], [20] did not mutually agree in their conclusions and, also, their investigation is actually limited by the use of private datasets explicitly designed for the considered application.…”

Localization and Tracking of Implantable Biomedical Sensors

“…In particular, wearable biomedical sensor technologies provide lower cost of care, minimally-invasive and effective procedures and shorter recovery times that improve the health outcomes. Detailed characteristics and various applications of wearable biomedical sensor systems can be found in [57,82,83,84,85,86]. Some of the more recent wearable sensor application areas are monitoring of vital signals, medical diagnosis and treatment, home-based rehabilitation and physical therapy, telesurgery, biomechanics, gait and posture recognition, detecting the emotional state and stress level of people and remote monitoring of the physically or mentally disabled, the elderly and children.…”

Investigating Inter-Subject and Inter-Activity Variations in Activity Recognition Using Wearable Motion Sensors

“…As expected, classification accuracy degrades significantly if one attempts to recognize the activities of a given subject with a system trained on other subjects' data. For this reason, most systems designed, for example, for physical therapy and rehabilitation, are individually trained for each subject so that the reference data are directly acquired from the subject who will use the system [16][17][18].…”