Benjamin T Schumacher, Michael J LaMonte, Andrea Z LaCroix, Eleanor M Simonsick, Steven P Hooker, Humberto Parada, John Bellettiere, Arun Kumar
{"title":"针对老年人开发、验证和运输几种基于机器学习的非运动型 VO2max 预测模型。","authors":"Benjamin T Schumacher, Michael J LaMonte, Andrea Z LaCroix, Eleanor M Simonsick, Steven P Hooker, Humberto Parada, John Bellettiere, Arun Kumar","doi":"10.1016/j.jshs.2024.02.004","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>There exist few maximal oxygen uptake (VO<sub>2max</sub>) non-exercise-based prediction equations, fewer using machine learning (ML), and none specifically for older adults. Since direct measurement of VO<sub>2max</sub> is infeasible in large epidemiologic cohort studies, we sought to develop, validate, compare, and assess the transportability of several ML VO<sub>2max</sub> prediction algorithms.</p><p><strong>Methods: </strong>The Baltimore Longitudinal Study of Aging (BLSA) participants with valid VO<sub>2max</sub> tests were included (n = 1080). Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine (SVM) algorithms were trained to predict VO<sub>2max</sub> values. We developed these algorithms for: (a) the overall BLSA, (b) by sex, (c) using all BLSA variables, and (d) variables common in aging cohorts. Finally, we quantified the associations between measured and predicted VO<sub>2max</sub> and mortality.</p><p><strong>Results: </strong>The age was 69.0 ± 10.4 years (mean ± SD) and the measured VO<sub>2max</sub> was 21.6 ± 5.9 mL/kg/min. Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine yielded root mean squared errors of 3.4 mL/kg/min, 3.6 mL/kg/min, 3.4 mL/kg/min, 3.6 mL/kg/min, and 3.5 mL/kg/min, respectively. Incremental quartiles of measured VO<sub>2max</sub> showed an inverse gradient in mortality risk. Predicted VO<sub>2max</sub> variables yielded similar effect estimates but were not robust to adjustment.</p><p><strong>Conclusion: </strong>Measured VO<sub>2max</sub> is a strong predictor of mortality. Using ML can improve the accuracy of prediction as compared to simpler approaches but estimates of association with mortality remain sensitive to adjustment. Future studies should seek to reproduce these results so that VO<sub>2max</sub>, an important vital sign, can be more broadly studied as a modifiable target for promoting functional resiliency and healthy aging.</p>","PeriodicalId":48897,"journal":{"name":"Journal of Sport and Health Science","volume":" ","pages":"611-620"},"PeriodicalIF":9.7000,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11282333/pdf/","citationCount":"0","resultStr":"{\"title\":\"Development, validation, and transportability of several machine-learned, non-exercise-based VO<sub>2max</sub> prediction models for older adults.\",\"authors\":\"Benjamin T Schumacher, Michael J LaMonte, Andrea Z LaCroix, Eleanor M Simonsick, Steven P Hooker, Humberto Parada, John Bellettiere, Arun Kumar\",\"doi\":\"10.1016/j.jshs.2024.02.004\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>There exist few maximal oxygen uptake (VO<sub>2max</sub>) non-exercise-based prediction equations, fewer using machine learning (ML), and none specifically for older adults. Since direct measurement of VO<sub>2max</sub> is infeasible in large epidemiologic cohort studies, we sought to develop, validate, compare, and assess the transportability of several ML VO<sub>2max</sub> prediction algorithms.</p><p><strong>Methods: </strong>The Baltimore Longitudinal Study of Aging (BLSA) participants with valid VO<sub>2max</sub> tests were included (n = 1080). Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine (SVM) algorithms were trained to predict VO<sub>2max</sub> values. We developed these algorithms for: (a) the overall BLSA, (b) by sex, (c) using all BLSA variables, and (d) variables common in aging cohorts. Finally, we quantified the associations between measured and predicted VO<sub>2max</sub> and mortality.</p><p><strong>Results: </strong>The age was 69.0 ± 10.4 years (mean ± SD) and the measured VO<sub>2max</sub> was 21.6 ± 5.9 mL/kg/min. Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine yielded root mean squared errors of 3.4 mL/kg/min, 3.6 mL/kg/min, 3.4 mL/kg/min, 3.6 mL/kg/min, and 3.5 mL/kg/min, respectively. Incremental quartiles of measured VO<sub>2max</sub> showed an inverse gradient in mortality risk. Predicted VO<sub>2max</sub> variables yielded similar effect estimates but were not robust to adjustment.</p><p><strong>Conclusion: </strong>Measured VO<sub>2max</sub> is a strong predictor of mortality. Using ML can improve the accuracy of prediction as compared to simpler approaches but estimates of association with mortality remain sensitive to adjustment. Future studies should seek to reproduce these results so that VO<sub>2max</sub>, an important vital sign, can be more broadly studied as a modifiable target for promoting functional resiliency and healthy aging.</p>\",\"PeriodicalId\":48897,\"journal\":{\"name\":\"Journal of Sport and Health Science\",\"volume\":\" \",\"pages\":\"611-620\"},\"PeriodicalIF\":9.7000,\"publicationDate\":\"2024-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11282333/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Sport and Health Science\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1016/j.jshs.2024.02.004\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2024/2/29 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"HOSPITALITY, LEISURE, SPORT & TOURISM\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Sport and Health Science","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.jshs.2024.02.004","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/2/29 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"HOSPITALITY, LEISURE, SPORT & TOURISM","Score":null,"Total":0}
Development, validation, and transportability of several machine-learned, non-exercise-based VO2max prediction models for older adults.
Background: There exist few maximal oxygen uptake (VO2max) non-exercise-based prediction equations, fewer using machine learning (ML), and none specifically for older adults. Since direct measurement of VO2max is infeasible in large epidemiologic cohort studies, we sought to develop, validate, compare, and assess the transportability of several ML VO2max prediction algorithms.
Methods: The Baltimore Longitudinal Study of Aging (BLSA) participants with valid VO2max tests were included (n = 1080). Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine (SVM) algorithms were trained to predict VO2max values. We developed these algorithms for: (a) the overall BLSA, (b) by sex, (c) using all BLSA variables, and (d) variables common in aging cohorts. Finally, we quantified the associations between measured and predicted VO2max and mortality.
Results: The age was 69.0 ± 10.4 years (mean ± SD) and the measured VO2max was 21.6 ± 5.9 mL/kg/min. Least absolute shrinkage and selection operator, linear- and tree-boosted extreme gradient boosting, random forest, and support vector machine yielded root mean squared errors of 3.4 mL/kg/min, 3.6 mL/kg/min, 3.4 mL/kg/min, 3.6 mL/kg/min, and 3.5 mL/kg/min, respectively. Incremental quartiles of measured VO2max showed an inverse gradient in mortality risk. Predicted VO2max variables yielded similar effect estimates but were not robust to adjustment.
Conclusion: Measured VO2max is a strong predictor of mortality. Using ML can improve the accuracy of prediction as compared to simpler approaches but estimates of association with mortality remain sensitive to adjustment. Future studies should seek to reproduce these results so that VO2max, an important vital sign, can be more broadly studied as a modifiable target for promoting functional resiliency and healthy aging.
期刊介绍：
The Journal of Sport and Health Science (JSHS) is an international, multidisciplinary journal that aims to advance the fields of sport, exercise, physical activity, and health sciences. Published by Elsevier B.V. on behalf of Shanghai University of Sport, JSHS is dedicated to promoting original and impactful research, as well as topical reviews, editorials, opinions, and commentary papers.
With a focus on physical and mental health, injury and disease prevention, traditional Chinese exercise, and human performance, JSHS offers a platform for scholars and researchers to share their findings and contribute to the advancement of these fields. Our journal is peer-reviewed, ensuring that all published works meet the highest academic standards.
Supported by a carefully selected international editorial board, JSHS upholds impeccable integrity and provides an efficient publication platform. We invite submissions from scholars and researchers worldwide, and we are committed to disseminating insightful and influential research in the field of sport and health science.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
