Ultra-High-Definition Mental Health Care by Digital-Human Integration: New Strategy for utilizing Multi-Modal, Big, and Precision Data

2023 Fiscal Year Final Research Report

• Project Area: Ultra-High-Definition Mental Health Care by Digital-Human Integration: New Strategy for utilizing Multi-Modal, Big, and Precision Data
• Project/Area Number: 21H05065
• Research Category: Grant-in-Aid for Transformative Research Areas (B)
• Allocation Type: Single-year Grants
• Review Section: Transformative Research Areas, Section (I)
• Research Institution: Japan Advanced Institute of Science and Technology
• Principal Investigator: Nishimura Takuichi 北陸先端科学技術大学院大学, 先端科学技術研究科, 教授 (80357722)
• Project Period (FY): 2021-08-23 – 2024-03-31
• Keywords: 知識工学 / 暗黙知 / オントロジー / 知識構造化 / 知識連携

Outline of Final Research Achievements

A01 “Ontology Group” aimed to construct a computer-readable ontology of psychotherapy processes and psychiatric symptoms using expert artificial intelligence techniques, which can be used for the unified management of large amounts of diverse and precise data. The ontologies are processed to integrate data, and features are extracted from natural language and speech data using data-driven artificial intelligence techniques to identify psychiatric symptoms and predict outcomes. The goal of this project was to establish a foundation for systematically integrating the data of Groups A02 and A03, which identify psychiatric symptoms and predict outcomes, and Group A04, which clarify the network structure between various interventions and psychiatric symptoms, into a “variable growth database.

Free Research Field

人工知能、知識科学、サービス工学、パターン認識

Academic Significance and Societal Importance of the Research Achievements

本領域では，人間同士のコミュニケーションの多種・大量・精密データを用いて精神状態識別やアウトカム予測を可能とし，それを活用してデジタル-人間双方の手段を融合させた高精細のケアを実現しようとする領域を『精神の超高精細ケア』と定義し，高品質の心理療法臨床試験データを用いて，この領域の発展基盤の構築を目指した．本研究が始まる以前は，心理療法における知識は，教科書に記述されているもの以外の暗黙的なものは，ほとんどが属人化しており，セラピストや他の支援者と共有することは難しかった．教科書の文章も，様々な情報が記述されており，新人が勉強するには，要点把握までに時間を要しているという現状があった．