MRI white matter lesion segmentation using an ensemble of neural networks and overcomplete patch-based voting

José V. Manjón (1) , Pierrick Coupé (2, 3) , Parnesh Raniga (4) , Ying Xia (4) , Patricia Desmond (5) , Jurgen Fripp (6) , Olivier Salvado (6)

1 ITACA
2 LaBRI - Laboratoire Bordelais de Recherche en Informatique
3 PICTURA - Patch-based processing for medical and natural images
4 CSIRO ICT Centre - CSIRO Information and Commuciation Technologies
5 University of Melbourne
6 CSIRO - Commonwealth Scientific and Industrial Research Organisation [Canberra]

Accurate quantification of white matter hyperintensities (WMH) from Magnetic Resonance Imaging (MRI) is a valuable tool for the analysis of normal brain ageing or neurodegeneration. Reliable automatic extraction of WMH lesions is challenging due to their heterogeneous spatial occurrence, their small size and their diffuse nature. In this paper, we present an automatic method to segment these lesions based on an ensemble of overcomplete patch-based neural networks. The proposed method successfully provides accurate and regular segmentations due to its overcomplete nature while minimizing the segmentation error by using a boosted ensemble of neural networks. The proposed method compared favourably to state of the art techniques using two different neurodegenerative datasets.

Mots clés: lesion segmentation; MRI; brain; patch-based; neural network; ensemble

Domaines: Imagerie médicale; Imagerie

José V. Manjón, Pierrick Coupé, Parnesh Raniga, Ying Xia, Patricia Desmond, et al.. MRI white matter lesion segmentation using an ensemble of neural networks and overcomplete patch-based voting. Computerized Medical Imaging and Graphics, In press, ⟨10.1016/j.compmedimag.2018.05.001⟩. ⟨hal-01808412⟩