Deep grading for MRI-based differential diagnosis of Alzheimer's disease and Frontotemporal dementia

Huy-Dung Nguyen, Michaël Clément, Vincent Planche, Boris Mansencal, Pierrick Coupé

Abstract: Highlights•Our framework enables to improve the interpretability of deep model outputs for a multi-class classification problem.•The grading map can be used to observe the impact on brain of different disease’s subtypes.•The combination of the structure grading and the structure atrophy enables to improve the generalization of our framework.