Hippocampus Subfield Segmentation Using a Patch-Based Boosted Ensemble of Autocontext Neural Networks

José V Manjón (1) , Pierrick Coupé (2, 3)

1 UPV - Universitat Politècnica de València = Universitad Politecnica de Valencia = Polytechnic University of Valencia
2 PICTURA - Patch-based processing for medical and natural images
3 LaBRI - Laboratoire Bordelais de Recherche en Informatique

The hippocampus is a brain structure that is involved in several cogni-tive functions such as memory and learning. It is a structure of great interest in the study of the healthy and diseased brain due to its relationship to several neu-rodegenerative pathologies. In this work, we propose a novel patch-based method that uses an ensemble of boosted neural networks to perform the hippocampus subfield segmentation on multimodal MRI. This new method minimizes both random and systematic errors using an overcomplete autocontext patch-based labeling using a novel boosting strategy. The proposed method works well on high resolution MRI but also on standard resolution images after superresolution. Finally , the proposed method was compared with a similar state-of-the-art methods showing better results in terms of both accuracy and efficiency.

Domaines: Imagerie médicale

José V Manjón, Pierrick Coupé. Hippocampus Subfield Segmentation Using a Patch-Based Boosted Ensemble of Autocontext Neural Networks. International Workshop on Patch-based Techniques in Medical Imaging (MICCAI), Sep 2017, Québec, Canada. ⟨10.1007/978-3-319-67434-6_4⟩. ⟨hal-01626265⟩