Chien-Hua Huang , Tien-lung Sun , Min-Chi Chiu , Bih-O Lee
{"title":"Feasibility of predicting functional decline in the elderly through key posture information during sit-to-stand movement","authors":"Chien-Hua Huang , Tien-lung Sun , Min-Chi Chiu , Bih-O Lee","doi":"10.1016/j.humov.2024.103212","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><p>Early detection of functional decline in the elderly in day care centres facilitates timely implementation of preventive and treatment measures.</p></div><div><h3>Research question</h3><p>Whether or not a predictive model can be developed by applying image recognition to analyze elderly individuals' posture during the sit-to-stand (STS) manoeuvre.</p></div><div><h3>Methods</h3><p>We enrolled sixty-six participants (24 males and 42 females) in an observational study design. To estimate posture key point information, we employed a region-based convolutional neural network model and utilized nine key points and their coordinates to calculate seven eigenvalues (X1-X7) that represented the motion curve features during the STS manoeuvre. One-way analysis of variance was performed to evaluate four STS strategies and four types of compensation strategies for three groups with different capacities (college students, community-dwelling elderly, and day care center elderly). Finally, a machine learning predictive model was established.</p></div><div><h3>Results</h3><p>Significant differences (<em>p</em> < 0.05) were observed in all eigenvalues except X2 (momentum transfer phase, <em>p</em> = 0.168) between participant groups; significant differences (<em>p</em> < 0.05) were observed in all eigenvalues except X2 (<em>p</em> = 0.219) and X3 (hip-rising phase, <em>p</em> = 0.286) between STS patterns; significant differences (<em>p</em> < 0.05) were observed in all eigenvalues except X2 (<em>p</em> = 0.842) and X3 (<em>p</em> = 0.074) between compensation strategies. The motion curve eigenvalues of the seven posture key points were used to build a machine learning model with 85% accuracy in capacity detection, 70% accuracy in pattern detection, and 85% accuracy in compensation strategy detection.</p></div><div><h3>Significance</h3><p>This study preliminarily demonstrates that eigenvalues can be used to detect STS patterns and compensation strategies adopted by individuals with different capacities. Our machine learning model has excellent predictive accuracy and may be used to develop inexpensive and effective systems to help caregivers to continuously monitor STS patterns and compensation strategies of elderly individuals as warning signs of functional decline.</p></div>","PeriodicalId":55046,"journal":{"name":"Human Movement Science","volume":null,"pages":null},"PeriodicalIF":1.6000,"publicationDate":"2024-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0167945724000356/pdfft?md5=35b480489a9a61d807e16f7fb9e33f6f&pid=1-s2.0-S0167945724000356-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Human Movement Science","FirstCategoryId":"102","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0167945724000356","RegionNum":3,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"NEUROSCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Background
Early detection of functional decline in the elderly in day care centres facilitates timely implementation of preventive and treatment measures.
Research question
Whether or not a predictive model can be developed by applying image recognition to analyze elderly individuals' posture during the sit-to-stand (STS) manoeuvre.
Methods
We enrolled sixty-six participants (24 males and 42 females) in an observational study design. To estimate posture key point information, we employed a region-based convolutional neural network model and utilized nine key points and their coordinates to calculate seven eigenvalues (X1-X7) that represented the motion curve features during the STS manoeuvre. One-way analysis of variance was performed to evaluate four STS strategies and four types of compensation strategies for three groups with different capacities (college students, community-dwelling elderly, and day care center elderly). Finally, a machine learning predictive model was established.
Results
Significant differences (p < 0.05) were observed in all eigenvalues except X2 (momentum transfer phase, p = 0.168) between participant groups; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.219) and X3 (hip-rising phase, p = 0.286) between STS patterns; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.842) and X3 (p = 0.074) between compensation strategies. The motion curve eigenvalues of the seven posture key points were used to build a machine learning model with 85% accuracy in capacity detection, 70% accuracy in pattern detection, and 85% accuracy in compensation strategy detection.
Significance
This study preliminarily demonstrates that eigenvalues can be used to detect STS patterns and compensation strategies adopted by individuals with different capacities. Our machine learning model has excellent predictive accuracy and may be used to develop inexpensive and effective systems to help caregivers to continuously monitor STS patterns and compensation strategies of elderly individuals as warning signs of functional decline.
期刊介绍:
Human Movement Science provides a medium for publishing disciplinary and multidisciplinary studies on human movement. It brings together psychological, biomechanical and neurophysiological research on the control, organization and learning of human movement, including the perceptual support of movement. The overarching goal of the journal is to publish articles that help advance theoretical understanding of the control and organization of human movement, as well as changes therein as a function of development, learning and rehabilitation. The nature of the research reported may vary from fundamental theoretical or empirical studies to more applied studies in the fields of, for example, sport, dance and rehabilitation with the proviso that all studies have a distinct theoretical bearing. Also, reviews and meta-studies advancing the understanding of human movement are welcome.
These aims and scope imply that purely descriptive studies are not acceptable, while methodological articles are only acceptable if the methodology in question opens up new vistas in understanding the control and organization of human movement. The same holds for articles on exercise physiology, which in general are not supported, unless they speak to the control and organization of human movement. In general, it is required that the theoretical message of articles published in Human Movement Science is, to a certain extent, innovative and not dismissible as just "more of the same."