M3

The M3 (multi-modal imaging and multi-level characteristic with multi-classifier) is a brain imaging classification tool, which can help researchers to discriminate patients from normal controls. The M3 includes three steps: feature selection, maximum uncertainty linear discriminant analysis (MLDA)-based classification and multi-classifier. A leave-one-out cross-validation (LOOCV) is further used to estimate the performance of the M3. Finally, the most discriminative features are identified.