@techreport{oai:ipsj.ixsq.nii.ac.jp:00218327,
 author = {新崎, 義峰 and 伊藤, 弘大 and 小玉, 伽那 and 藤田, 和之 and 武田, 理宏 and 尾上, 孝雄 and 伊藤, 雄一 and Yoshio, Shinzaki and Kodai, Ito and Kana, Kodama and Kazuyuki, Fujita and Takeda, Toshihiro and Takao, Onoye and Yuichi, Itoh\n},
 issue = {2},
 month = {Jun},
 note = {For the elderly to live comfortably at home, it is necessary to prevent the deterioration of mobility called locomotive syndrome. However, many people visit a medical institution when weakness in their legs and backs becomes apparent, often before it is too late. In this study, we aim to develop a system to estimate leg strength from daily activities. As a preliminary step, we implemented a floor-type device that can acquire center-of-gravity and weight changes, and conducted an evaluation experiment of leg strength estimation using the device. In the experiment, center-of-gravity and weight changes were measured in three 22-year-old students during standing, sitting, and walking before and after squatting, and significant differences were investigated and classified by machine learning. The results showed that significant differences were found in many of the features, and the classification was successful with a 90.1% correct rate., For the elderly to live comfortably at home, it is necessary to prevent the deterioration of mobility called locomotive syndrome. However, many people visit a medical institution when weakness in their legs and backs becomes apparent, often before it is too late. In this study, we aim to develop a system to estimate leg strength from daily activities. As a preliminary step, we implemented a floor-type device that can acquire center-of-gravity and weight changes, and conducted an evaluation experiment of leg strength estimation using the device. In the experiment, center-of-gravity and weight changes were measured in three 22-year-old students during standing, sitting, and walking before and after squatting, and significant differences were investigated and classified by machine learning. The results showed that significant differences were found in many of the features, and the classification was successful with a 90.1% correct rate.},
 title = {スクワットによる脚力疲労にに関伴すうる日検常討生活動作時の重心揺動変化に関する検討},
 year = {2022}
}