Adaptive fusion of texture-based grading: Application to Alzheimer's disease detection

Kilian Hett (1) , Vinh-Thong Ta (2, 1) , José V Manjón (3) , Pierrick Coupé (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique
2 Bordeaux INP - Institut Polytechnique de Bordeaux
3 UPV - Universitat Politècnica de València = Universitad Politecnica de Valencia = Polytechnic University of Valencia

Alzheimer's disease is a neurodegenerative process leading to irreversible mental dysfunctions. The development of new biomark-ers is crucial to perform an early detection of this disease. Among new biomarkers proposed during the last decades, patch-based grading framework demonstrated state-of-the-art results. In this paper, we study the potential using texture information based on Gabor lters to improve patch-based grading method performance, with a focus on the hippocam-pal structure. We also propose a novel fusion framework to eciently combine multiple grading maps derived from a Gabor lters bank. Finally , we compare our new texture-based grading biomarker with the state-of-the-art approaches to demonstrate the high potential of the proposed method.

Mots clés: Alzheimer's disease classsification; Mild Cognitive Impairment; Patch-based grading fusion; Multi-features

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Machine Learning [stat.ML]; Neurosciences [q-bio.NC]; Imagerie

Kilian Hett, Vinh-Thong Ta, José V Manjón, Pierrick Coupé. Adaptive fusion of texture-based grading: Application to Alzheimer's disease detection. PatchMI: 3rd International Workshop on Patch-based Techniques in Medical Imaging, Sep 2017, Quebec, Canada. ⟨hal-01582003⟩