2015 Fiscal Year Final Research Report
Functional subdivision of brain structures by clustering of PET images
Project/Area Number |
26861024
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Multi-year Fund |
Research Field |
Radiation science
|
Research Institution | 秋田県立脳血管研究センター(研究部門) |
Principal Investigator |
Matsubara Keisuke 秋田県立脳血管研究センター(研究局), その他部局等, 研究員 (40588430)
|
Research Collaborator |
KUWABARA Hiroto Johns Hopkins大学, 医学部, 准教授
|
Project Period (FY) |
2014-04-01 – 2016-03-31
|
Keywords | PET / 線条体 / クラスター分析 / 自動VOI抽出 |
Outline of Final Research Achievements |
We aimed to develop a method to estimate functional subdivision of brain structure by cluster analysis with multiple PET images. For the purpose, we validated accuracy of automated VOI labeling methods, and tested to generate functional subdivision map of striatum. Inter-method differences of accuracy of the automated VOI methods were different among the methods. We observed that inaccuracy of the automated VOI induced bias in binding potential estimated by PET kinetic analysis by 23.3% at maximum. We revealed inter-method differences of estimated functional subdivision of striatum among the clustering methods. Especially, Gaussian mixture model (GMM) resulted in subdivision similar to well-known anatomical subdivision of striatum. Further studies are required to reveal correlation between the estimated functional subdivisions and some brain functions.
|
Free Research Field |
医用画像工学、核医学
|