Patch-based DTI grading: Application to Alzheimer's disease classication

Kilian Hett (1, 2) , Vinh-Thong Ta (3, 1, 2) , Rémi Giraud (1, 2) , Mary Mondino (1, 2) , José V Manjón (4, 1, 2) , Pierrick Coupé (2, 1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique
2 PICTURA - Patch-based processing for medical and natural images
3 Bordeaux INP - Institut Polytechnique de Bordeaux
4 ITACA

Early diagnosis is one of the most important challenges related to Alzheimer's disease (AD). To address this issue, numerous studies proposed biomarkers based on anatomical MRI. Among them, patch-based grading demonstrated state-of-the-art results when applied to T1-weighted MRI. In this work, we propose to use a similar framework on dierent diusion parameters extracted from DTI. We also propose to use a fast patch-based search strategy to provide novel biomarkers for the early detection of AD. We intensively compare our new grading-based DTI features with basic MRI/DTI biomarkers and evaluate our method within a cross validation classication framework. Finally, we demonstrate that the proposed biomarkers obtain competitive results for the identication of the dierent stages of AD.

Mots clés: DWI; Mild Cognitive Impairment; Patch-based grading; Alzheimer's disease classication; DTI

Domaines: Imagerie médicale

Kilian Hett, Vinh-Thong Ta, Rémi Giraud, Mary Mondino, José V Manjón, et al.. Patch-based DTI grading: Application to Alzheimer's disease classication. Second International Workshop, Patch-MI 2016, Held in Conjunction with MICCAI 2016, Oct 2016, Athens, Greece. ⟨hal-01395487⟩