2016 Fiscal Year Final Research Report
In silico disease classification based on brain perfusion SPECT
| Project/Area Number |
25870406
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Physical pharmacy
Intelligent informatics
|
|---|
| Research Institution | Hokuriku University (2015-2016)
Osaka University (2013-2014) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2013-04-01 – 2017-03-31
|
| Keywords | 画像の識別 / データマイニング / 画像診断 / 知識の発見 |
|---|
| Outline of Final Research Achievements |
Brain perfusion SPECT is an imaging technique to evaluate a cerebral blood flow (CBF) in diagnosis of several neurodegenerative diseases. In this study, our aims are to build the model which enables us to classify some neurodegenerative diseases based on SPECT images, and to supply the information about the brain regions that it is important for the prediction of the diseases.
We built the model to classify three kinds of diseases (Alzheimer disease, Parkinson disease or other neurodegenerative diseases) by support vector machine. In the prediction, brain regions whose CBF was different between the diseases and which were included into the SVM model were not conflict with clinical knowledge. Therefore, the model and information about the brain regions obtained in this study should be useful for computer aided diagnosis.
|
| Free Research Field |
情報系薬学
|
