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

Autor: Reda Abdellah Kamraoui, Thomas Tourdias, José V. Manjón, Vinh-Thong Ta, Pierrick Coupé, Boris Mansencal
Přispěvatelé: Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS), Neurocentre Magendie : Physiopathologie de la Plasticité Neuronale (U1215 Inserm - UB), Université de Bordeaux (UB)-Institut François Magendie-Institut National de la Santé et de la Recherche Médicale (INSERM), ITACA, Universitat Politècnica de València (UPV), Patch-based processing for medical and natural images (PICTURA), Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS)-Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS), KAMRAOUI, Reda Abdellah
Rok vydání: 2020
Předmět:
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI]
FOS: Computer and information sciences
Multiple Sclerosis
Image quality
Generalization
Property (programming)
Computer science
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
[INFO.INFO-IM] Computer Science [cs]/Medical Imaging
Initialization
Health Informatics
Convolutional neural network
030218 nuclear medicine & medical imaging
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
03 medical and health sciences
Consistency (database systems)
0302 clinical medicine
Deep Learning
[INFO.INFO-IM]Computer Science [cs]/Medical Imaging
FOS: Electrical engineering
electronic engineering
information engineering
Image Processing
Computer-Assisted
Humans
Radiology
Nuclear Medicine and imaging
Segmentation
[SDV.NEU] Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC]
Radiological and Ultrasound Technology
business.industry
Deep learning
Image and Video Processing (eess.IV)
Brain
Pattern recognition
Electrical Engineering and Systems Science - Image and Video Processing
Multiple Sclerosis Segmentation
Computer Graphics and Computer-Aided Design
Domain Generalization
[SDV.NEU]Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC]
Computer Vision and Pattern Recognition
Artificial intelligence
Neural Networks
Computer
business
030217 neurology & neurosurgery
Zdroj: Medical Image Analysis
Medical Image Analysis, 2022, 76, pp.102312. ⟨10.1016/j.media.2021.102312⟩
ISSN: 1361-8423
1361-8415
DOI: 10.1016/j.media.2021.102312⟩
Popis: Recently, segmentation methods based on Convolutional Neural Networks (CNNs) showed promising performance in automatic Multiple Sclerosis (MS) lesions segmentation. These techniques have even outperformed human experts in controlled evaluation conditions such as Longitudinal MS Lesion Segmentation Challenge (ISBI Challenge). However state-of-the-art approaches trained to perform well on highly-controlled datasets fail to generalize on clinical data from unseen datasets. 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). This generalization property results from three main contributions. First, DLB is based on a large group of compact 3D CNNs. This spatially distributed strategy ensures a robust prediction despite the risk of generalization failure of some individual networks. Second, DLB includes a new image quality data augmentation to reduce dependency to training data specificity (e.g., acquisition protocol). Finally, to learn a more generalizable representation of MS lesions, we propose a hierarchical specialization learning (HSL). HSL is performed by pre-training a generic network over the whole brain, before using its weights as initialization to locally specialized networks. By this end, DLB learns both generic features extracted at global image level and specific features extracted at local image level. DLB generalization was validated in cross-dataset experiments on MSSEG'16, ISBI challenge, and in-house datasets. During experiments, DLB showed higher segmentation accuracy, better segmentation consistency and greater generalization performance compared to state-of-the-art methods. Therefore, DLB offers a robust framework well-suited for clinical practice.
Databáze: OpenAIRE
Externí odkaz: