Development of an objective MDD screening system adopting autonomic nervous responses to mental tasks

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K12603
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 90110:Biomedical engineering-related
• Research Institution: Tokyo Metropolitan University
• Principal Investigator: Matsui Takemi 東京都立大学, システムデザイン研究科, 教授 (50404934)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 生体医工学 / 生体情報 / ストレス

Outline of Final Research Achievements

In this research, we aimed to establish a method that screens patients with depression from healthy people without being influenced by the relationship between patients and doctors and a subjective diagnosis made by them based on their experiences.
We developed an automatic prototype system instructing subjects to speak out random numbers as a mental burden and obtained heart rate waves using a photoplethysmograph sensor. Also, we created an algorithm that discriminated patients with depression from healthy subjects using Linear Discriminant Analysis.We conducted the screening test with patients at a mental hospital and control subjects at our university. Due to unstable heart rate measurements, we had poor results, so we installed machine learning to our algorithm and successfully obtained high accuracy in discrimination.

Free Research Field

生体医工学

Academic Significance and Societal Importance of the Research Achievements

うつ病診断は、アンケートや医師の問診により行われるが、主観に基づくアンケートの信頼性は高くなく、また問診は医師自身の経験に大きく左右される。そこで、本研究では、うつ病患者と健常者で異なる精神負荷に対する自律神経応答から、判別を行った。個人差が大きい自律神経応答に対し、機械学習を用いたアルゴリズムを構築することで、高精度な判別を可能とした。本システムは精神科のみならず、うつ病罹患者の在宅で自身の状態確認や、専門ではない産業医の診療補助として使用が考えられる。客観的数値により、産業医の精神科専門医への受診促進、潜在的うつ病患者の受診促進への応用が期待される。