Development of diagnostic assistance software for Parkinson's syndrome by brain network analysis using graph theory

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K17157
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 52040:Radiological sciences-related
• Research Institution: National Center of Neurology and Psychiatry
• Principal Investigator: Shigemoto Yoko 国立研究開発法人国立精神・神経医療研究センター, 病院 放射線診療部, 医師 (00815384)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: ネットワーク解析 / 健常者 / 年齢 / 性別 / パーキンソン病

Outline of Final Research Achievements

Structural network analysis was performed using brain MRI of 812 healthy subjects, and a database of local network index images was created. Next, patients with degenerative diseases with parkinsonism (Parkinson's disease, progressive supranuclear palsy, multiple system atrophy-parkinsonism variant) were similarly analyzed to detect disease-specific network abnormalities. As a result, the clustering coefficient, which is one of the local network indicators, coincided with the site of decline in brain function compared to the site of atrophy in gray matter, and was considered to be useful for differentiation.

Free Research Field

放射線医学

Academic Significance and Societal Importance of the Research Achievements

MRIのグラフ理論を用いたネットワーク解析は、パーキンソニズムをきたす疾患の鑑別に有用であった。この手法は他の疾患にも応用可能と思われるため、さまざまな神経変性疾患の診断補助として日常診療に役立てることができうると考える。