期刊
MEDICINE AND SCIENCE IN SPORTS AND EXERCISE
卷 47, 期 9, 页码 1988-1996出版社
LIPPINCOTT WILLIAMS & WILKINS
DOI: 10.1249/MSS.0000000000000615
关键词
RESULTANT FORCE; SHOES; PRINCIPAL COMPONENT ANALYSIS; SUPPORT VECTOR MACHINE; PATTERN RECOGNITION
资金
- Mizuno Corporation (Osaka, Japan)
- Biomechanigg Sport and Health Research (Calgary, Canada)
Purpose The effects of footwear on the musculoskeletal system are commonly assessed by interpreting the resultant force at the foot during the stance phase of running. However, this approach overlooks loading patterns across the entire foot. An alternative technique for assessing foot loading across different footwear conditions is possible using comprehensive analysis tools that extract different foot loading features, thus enhancing the functional interpretation of the differences across different interventions. The purpose of this article was to use pattern recognition techniques to develop and use a novel comprehensive method for assessing the effects of different footwear interventions on plantar loading. Methods A principal component analysis was used to extract different loading features from the stance phase of running, and a support vector machine (SVM) was used to determine whether and how these loading features were different across three shoe conditions. Results The results revealed distinct loading features at the foot during the stance phase of running. The loading features determined from the principal component analysis allowed successful classification of all three shoe conditions using the SVM. Several differences were found in the location and timing of the loading across each pairwise shoe comparison using the output from the SVM. Conclusions The analysis approach proposed can successfully be used to compare different loading patterns with a much greater resolution than has been reported previously. This study has several important applications. One such application is that it would not be relevant for a user to select a shoe or for a manufacturer to alter a shoe's construction if the classification across shoe conditions would not have been significant.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据