Elucidation of the neural basis of disorders of thinking in psychiatric disorders

2021 Fiscal Year Final Research Report

• Project Area: Correspondence and Fusion of Artificial Intelligence and Brain Science
• Project/Area Number: 16H06572
• Research Category: Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
• Allocation Type: Single-year Grants
• Review Section: Complex systems
• Research Institution: Tokyo Medical and Dental University (2017-2021); Kyoto University (2016)
• Principal Investigator: Takahashi Hidehiko 東京医科歯科大学, 大学院医歯学総合研究科, 教授 (60415429)
• Project Period (FY): 2016-06-30 – 2021-03-31
• Keywords: 精神疾患

Outline of Final Research Achievements

Thought disorder in psychiatric disorders include disorders of concept formation and abstraction. In schizophrenia, in particular, abnormalities in the semantic network of words have been assumed to underlie the thought disorder, which is a core symptom, but the neural basis of the disorder has remained unclear. We combined natural language processing, one of the artificial intelligence technologies, and brain image analysis technology to investigate abnormalities in semantic representations in the brain of schizophrenia patients. Brain activity while watching a video was measured by functional MRI. In healthy subjects, cluster formation was observed for each concept, such as words related to humans, organisms, and artifacts, but the degree of cluster formation was not clear in the schizophrenia group, suggesting an abnormal network of semantic representations in the brain.

Free Research Field

精神医学

Academic Significance and Societal Importance of the Research Achievements

思考の障害を有する精神疾患を対象にすることが特色である。人工知能は網羅的に概念を集めたり、はるか先まで先読みするのは得意である一方、人間ほどの柔軟性には至っていない。柔軟な学習アルゴリズムや脳機構の理解のためには、健常者の脳内機構を調べることと同時に、その破綻と考えられる精神疾患を扱うことは有意義で、柔軟な思考・判断が可能な人工知能開発に示唆を与える。統合失調症の中核障害である思考障害の客観的評価や脳基盤の解明は精神医学上大きなインパクトを与える。人工知能技術を用いて開発した複数の精神疾患のバイオマーカーを組み合わせ、精神疾患間の鑑別を可能にしたことで、個別化医療に貢献する可能性が高い。