Development of the Fall Risk Assessment Tool Based on a Perception-Motion Model

2024 Fiscal Year Final Research Report

• Project/Area Number: 20K19858
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 61030:Intelligent informatics-related
• Research Institution: The University of Tokyo (2022-2024); Tokyo Institute of Technology (2020-2021)
• Principal Investigator: Uchiyama Emiko 東京大学, 大学院工学系研究科(工学部), 助教 (30845269)
• Project Period (FY): 2020-04-01 – 2025-03-31
• Keywords: 転倒予防 / 姿勢制御 / リスクアセスメント / 知覚モデル / 運動モデル / データ科学

Outline of Final Research Achievements

For the stumble risk, the depth perception is known as the risk factor, and the effect of the posture (depression angle) to detect stimuli from the depth direction was experimentally shown. For the stagger risk factor, the head-shaking test, which tests the balance ability of the person was proposed, and the model which examines the balance ability using the control parameter identification method under the assumption of the control model in the static standing before and after this test was also proposed. Through the analysis of the model for the head-shaking test, we confirmed the muscle activity increased after the head shaking. Also, as the way of searching the fall risk factors, several method using data science were developed: The similarity verification method using Akaike information quantity, the text mining method for exploring fall factors was developed, and technique which extracts common features related to the fall risk from multi-modal multiple data sets.

Free Research Field

人間情報学

Academic Significance and Societal Importance of the Research Achievements

本研究の成果では，知覚と姿勢の関係，姿勢制御を，モデルによって定量的に議論する方法が得られた．これらはヒトの行動に数理的な表現を与え，客観的にリスクアセスメントを行うための基盤となる．また，本研究で開発されたバランス機能テストは，5秒で10回首を振り，その前後での立位姿勢の変化を確認するもので，医療・介護に関わる専門職のリスクアセスメントの負担低減という点に社会的意義がある．データ科学的な転倒因子探索手法は，近年ますます高まるオープンサイエンスにおいて，ますます活用が期待される．特に，本研究成果を活用し，オープンデータを組み合わせて転倒因子を探索できれば，データ取得コストの大幅減が期待される．