Development of computer-aided detection system for early Alzheimer's disease on CT iamges

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K08706
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Medical Physics and Radiological Technology
• Research Institution: 秋田県立脳血管研究センター(研究部門)
• Principal Investigator: Takahashi Noriyuki 秋田県立脳血管研究センター(研究部門), 放射線医学研究部, 特任研究員 (90595076)
• Co-Investigator(Kenkyū-buntansha): 木下 俊文 秋田県立脳血管研究センター(研究部門), その他部局等, 研究員 (70314599)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 早期アルツハイマー / Computed tomography / 画像統計解析

Outline of Final Research Achievements

Enlargement of the temporal horn of the lateral ventricle on CT images is an accurate diagnostic marker of Alzheimer's disease. We developed a z-score-based semi-quantitative analysis of the temporal horn volume of the lateral ventricle for detection of early Alzheimer's disease on CT images. The previous study showed that regional mean z score of the temporal horn of the lateral ventricle calculated using the method was highly correlated with the volume of the temporal horn of the lateral ventricle for 50 subjects (R2=0.94). Also, This scheme was applied to CT scans from 51 AD patients and 31 controls to evaluate the performance of the method for detecting Alzheimer's. The estimated temporal horn of the lateral ventricle volume was markedly increased in the Alzheimer's group compared with the controls (P < .0001). Therefore, this computerized method may have the potential to accurately detect early Alzheimer's on CT images.

Free Research Field

医用画像工学