Cognitive Decline / Emotional Instability Alarming System for Elderly and Disabled People Living Alone

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K12005
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61050:Intelligent robotics-related
• Research Institution: Iwate University
• Principal Investigator: Kim Chyon Hae 岩手大学, 研究支援・産学連携センター, 客員准教授 (30424815)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 見守り / 人工知能 / 認知症 / センサネットワーク / 発報 / 独居

Outline of Final Research Achievements

We have developed a sensor network, which is composed of door opening/closing sensors, motion sensors, and current sensors. Our AI learned an ADL model at each home of three 20’s single men. The AI judged whether or not to issue a report using our method that separates rare behaviors that do not close to the ADL model. We were able to confirm that reports were issued during rare behaviors such as "after the subject came home tired from extra work" or "when he cooked as a quite rare behavior". This time, we were unable to conduct experiments on the elderly due to the coronavirus pandemic, but as a big achievement the AI captured behaviors similar to rare behaviors that tend to occur when the symptoms of elderly people with cognitive decline or emotional instability worsen. A psychiatrist evaluated it as being sufficiently practical for watching over the elderly.

Free Research Field

知能ロボティクス

Academic Significance and Societal Importance of the Research Achievements

ADLモデルをリアルタイムに学習してレア行動を発報するAIやセンサネットワークの構成論が分かったことで、実用化レベルの高齢者見守り装置ができた。これにより、独居老人がいつの間にかに認知機能低下や情緒不安定を起こし、周囲でトラブルを起こしたり自身の健康を害するといった事を予防することができる見込みが立った。先進国における独居老人の数は年々増え続けており、本装置の需要は今後も高まると予想される。