| Project/Area Number |
26860960
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Psychiatric science
|
|---|
| Research Institution | National Center of Neurology and Psychiatry |
|---|
Principal Investigator |
Fujii Takeshi 国立研究開発法人国立精神・神経医療研究センター, 病院, 医員 (80570837)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000)
Fiscal Year 2016: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2014: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
|---|
| Keywords | パターン認識 / MRI / うつ病 / 双極性障害 / 統計的ぱたーン認識 / 脳画像 / 気分障害 / 統計的パターン認識 |
|---|
| Outline of Final Research Achievements |
To support a differential diagnosis between Depression and Bipolar disorder with pattern recognition for MRI images, I conducted this research.
T1 weighted image and diffusion tensor image of each participant across depression, bipolar disorder and control groups were acquired. They were pre-processed and grey matter density image (GMD) and fractional anisotropy image (FA) were made. Support vector machine (SVM) learned features of these images for classification. Accuracy of classification with the SVM was examined.
To separate the MDD from HC with SVM, the GMD was more effective than FA. In contrast, to differentiate the BP from HC, the FA was more effective than GMD. However, to use both the GMD and FA did not improve classification accuracy between the MDD and BP. It is important to increase the sample size and keep improving means of application of pattern recognition for MRI images.
|
