Longitudinal detection of new MS lesions using Deep Learning

no code implementations • 16 Jun 2022 • Reda Abdellah Kamraoui, Boris Mansencal, José V Manjon, Pierrick Coupé

First, we propose to use transfer-learning from a model trained on a segmentation task using single time-points.

Data Augmentation Segmentation +1

POPCORN: Progressive Pseudo-labeling with Consistency Regularization and Neighboring

no code implementations • 13 Sep 2021 • Reda Abdellah Kamraoui, Vinh-Thong Ta, Nicolas Papadakis, Fanny Compaire, José V Manjon, Pierrick Coupé

Semi-supervised learning (SSL) uses unlabeled data to compensate for the scarcity of annotated images and the lack of method generalization to unseen domains, two usual problems in medical segmentation tasks.

Image Segmentation Lesion Segmentation +2

DeepLesionBrain: Towards a broader deep-learning generalization for multiple sclerosis lesion segmentation

no code implementations • 14 Dec 2020 • Reda Abdellah Kamraoui, Vinh-Thong Ta, Thomas Tourdias, Boris Mansencal, José V Manjon, Pierrick Coupé

Instead of proposing another improvement of the segmentation accuracy, we propose a novel method robust to domain shift and performing well on unseen datasets, called DeepLesionBrain (DLB).

Data Augmentation Lesion Segmentation +2