Automatically Evaluating Balance: A Machine Learning Approach

Compared to in-clinic balance training, in-home training is not as effective. This is, in part, due to the lack of feedback from physical therapists (PTs). In this paper, we analyze the feasibility of using trunk sway data and machine learning (ML) techniques to automatically evaluate balance, prov...

Mô tả chi tiết

Lưu vào:

Hiển thị chi tiết
Tác giả chính:	Bao, T.
Đồng tác giả:	Klatt, B. N.
Định dạng:	BB
Ngôn ngữ:	en_US
Thông tin xuất bản:	IEEE Xplore 2020
Chủ đề:	Balance rehabilitation balance performance machine learning telerehabilitation
Truy cập trực tuyến:	http://tailieuso.tlu.edu.vn/handle/DHTL/9876
Từ khóa:	Thêm từ khóa bạn đọc Không có từ khóa, Hãy là người đầu tiên gắn từ khóa cho biểu ghi này!

id	oai:localhost:DHTL-9876
record_format	dspace
spelling	oai:localhost:DHTL-98762020-12-09T08:37:08Z Automatically Evaluating Balance: A Machine Learning Approach Bao, T. Klatt, B. N. Whitney, S. L. Sienko, K. H. Wiens, J. Balance rehabilitation balance performance machine learning telerehabilitation Compared to in-clinic balance training, in-home training is not as effective. This is, in part, due to the lack of feedback from physical therapists (PTs). In this paper, we analyze the feasibility of using trunk sway data and machine learning (ML) techniques to automatically evaluate balance, providing accurate assessments outside of the clinic. We recruited sixteen participants to perform standing balance exercises. For each exercise, we recorded trunk sway data and had a PT rate balance performance on a scale of 1–5. The rating scale was adapted from the Functional Independence Measure. From the trunk sway data, we extracted a 61-dimensional feature vector representing the performance of each exercise. Given these labeled data, we trained a multi-class support vector machine (SVM) to map trunk sway features to PT ratings. Evaluated in a leave-one-participant-out scheme, the model achieved a classiﬁcation accuracy of 82%. Compared to participant self-assessment ratings, the SVM outputs were signiﬁcantly closer to PT ratings. The results of this pilot study suggest that in the absence of PTs, ML techniques can provide accurate assessments during standing balance exercises. Such automated assessments could reduce PT consultation time and increase user compliance outside of the clinic. 2020-12-09T08:35:59Z 2020-12-09T08:35:59Z 2019 BB http://tailieuso.tlu.edu.vn/handle/DHTL/9876 en_US IEEE Transactions on Neural Systems and Rehabilitation Engineering, (2019), VOL. 27, NO. 2, pp 179-186 application/pdf IEEE Xplore
institution	Trường Đại học Thủy Lợi
collection	DSpace
language	en_US
topic	Balance rehabilitation balance performance machine learning telerehabilitation
spellingShingle	Balance rehabilitation balance performance machine learning telerehabilitation Bao, T. Automatically Evaluating Balance: A Machine Learning Approach
description	Compared to in-clinic balance training, in-home training is not as effective. This is, in part, due to the lack of feedback from physical therapists (PTs). In this paper, we analyze the feasibility of using trunk sway data and machine learning (ML) techniques to automatically evaluate balance, providing accurate assessments outside of the clinic. We recruited sixteen participants to perform standing balance exercises. For each exercise, we recorded trunk sway data and had a PT rate balance performance on a scale of 1–5. The rating scale was adapted from the Functional Independence Measure. From the trunk sway data, we extracted a 61-dimensional feature vector representing the performance of each exercise. Given these labeled data, we trained a multi-class support vector machine (SVM) to map trunk sway features to PT ratings. Evaluated in a leave-one-participant-out scheme, the model achieved a classiﬁcation accuracy of 82%. Compared to participant self-assessment ratings, the SVM outputs were signiﬁcantly closer to PT ratings. The results of this pilot study suggest that in the absence of PTs, ML techniques can provide accurate assessments during standing balance exercises. Such automated assessments could reduce PT consultation time and increase user compliance outside of the clinic.
author2	Klatt, B. N.
author_facet	Klatt, B. N. Bao, T.
format	BB
author	Bao, T.
author_sort	Bao, T.
title	Automatically Evaluating Balance: A Machine Learning Approach
title_short	Automatically Evaluating Balance: A Machine Learning Approach
title_full	Automatically Evaluating Balance: A Machine Learning Approach
title_fullStr	Automatically Evaluating Balance: A Machine Learning Approach
title_full_unstemmed	Automatically Evaluating Balance: A Machine Learning Approach
title_sort	automatically evaluating balance: a machine learning approach
publisher	IEEE Xplore
publishDate	2020
url	http://tailieuso.tlu.edu.vn/handle/DHTL/9876
work_keys_str_mv	AT baot automaticallyevaluatingbalanceamachinelearningapproach
_version_	1787739876435689472

Automatically Evaluating Balance: A Machine Learning Approach

Tài liệu tương tự