Two different approaches to small sample size - a common problem in MRI-based studies
| Authors | |
|---|---|
| Year of publication | 2010 |
| Type | Conference abstract |
| MU Faculty or unit | |
| Citation | |
| Description | Recently, the small sample size problem and huge image data have often been discussed in MRI-based studies. Two methods for data reduction are compared here and further modified to solve classification of 3-D MRI data sets in the schizophrenia research. The results show that PCA based on covariance matrix of patients (pPCA) is more suitable for large MRI data reduction than 2DPCA. The results also indicate that deformation images are more appropriate for classification than GM density images. |
| Related projects: |