転倒歴と疫学データに基づく主成分分析を用いた転倒リスクの要因抽出

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2022/07/01

タイトル(英語): Extraction of Factors of Fall Risk Using Principal Component Analysis Based on Fall History and Epidemiological Data

著者名: 河越　祐太（鳥取大学大学院持続性社会創生科学研究科），櫛田　大輔（鳥取大学工学部（クロス情報科学研究センター）），松本　浩実（川崎医療福祉大学リハビリテーション学部）

著者名(英語): Yuta Kawagoe (Graduate School of Sustainability Science, Tottori Univerisy), Daisuke Kushida (Faculty of Engineering (Cross-Informatics Research Center), Tottori Univerisy), Hiromi Matsumoto (Faculty of Rehabilitation, Kawasaki University of Medical Welfare)

キーワード: 転倒リスク，疫学データ，主成分分析，リスクマップ，要因抽出　　fall risk，epidemiological data，principal component analysis，risk map，extraction of factors

要約(英語): Early detection of fracture risk is important to prevent fractures, which account for 12% of the causes of needing care. The current diagnosis of fracture risk is based on the FRAX index defined by the WHO. However, since this is a numerical assessment only for experts, it is difficult for patients to recognize their own fracture risk, and this does not lead to improved awareness of fracture prevention. Therefore, the authors obtained epidemiological data on 397 fracture-related items and conducted a follow-up survey on the fact of falling one year later in 136 elderly women in Hino-cho, Hino-gun, Tottori Prefecture. In this paper, we used the fact of falling, which is a risk factor for fracture, as a classification criterion, and extracted epidemiological data with high relevance to the fact of falling by principal component analysis of the epidemiological data, and expressed them in the form of a two-dimensional map. In addition, the reliability of the two-dimensional map was demonstrated by referring to the correlation between epidemiological data and the fact of falling.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.142 No.7 （2022） 特集：2021年電子・情報・システム部門大会

本誌掲載ページ: 706-712 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/142/7/142_706/_article/-char/ja/