Safety-Net: Automatic identification of segmentation errors for Multiple Sclerosis lesions. Safety-Net : Identification automatique des erreurs de segmentation des lésions de la Sclérose-en-Plaques

Archive ouverte

Lambert, Benjamin | Forbes, Florence | Doyle, Senan | Tucholka, Alan | Dojat, Michel

Edité par CCSD -

International audience. In the context of Multiple Sclerosis (MS), neural networks have the potential to greatly assist neurologists in personalizing therapy through automatic quantification of the evolution of brain lesions. However, these models produce predictions (lesions) without an estimation of their certainty. To address this, we propose Safety-Net, a neural network capable of segmenting MS lesions and associating a map of uncertainty, allowing the neurologist to quickly identify problematic areas and, if necessary, correct the prediction, thereby increasing the acceptability of the system. . Dans le contexte de la Sclérose-En-Plaques (SEP), les réseaux de neurones ont le potentiel d'assister grandement le neurologue dans la personnalisation de la thérapie par la quantification automatique de l'évolution des lésions cérébrales. Néanmoins, ces modèles produisent des prédictions (lésions) sans estimation de leur certitude. Pour remédier à cela, nous proposons Safety-Net, un réseau de neurones capable de segmenter les lésions SEP, et d'associer une carte de doute, permettant au neurologue d'identifier rapidement les zones problématiques, le cas échéant de corriger la prédiction, augmentant ainsi l'acceptabilité du système.

Suggestions

Du même auteur

Uncertainty-based Quality Control for Subcortical Structures Segmentation in T1-weighted Brain MRI

Archive ouverte | Lambert, Benjamin | CCSD

International audience. Deep Learning (DL) models are presently the gold standard for medical image segmentation. However, their performance may drastically drop in the presence of characteristics in test images not...

Leveraging 3D information in unsupervised brain MRI segmentation

Archive ouverte | Lambert, Benjamin | CCSD

International audience. Automatic segmentation of brain abnormalities is challenging, as they vary considerably from one pathology to another. Current methods are supervised and require numerous annotated images for...

Trustworthy clinical AI solutions: a unified review of uncertainty quantification in deep learning models for medical image analysis

Archive ouverte | Lambert, Benjamin | CCSD

The full acceptance of Deep Learning (DL) models in the clinical field is rather low with respect to the quantity of high-performing solutions reported in the literature. Particularly, end users are reluctant to rely on the rough ...

Chargement des enrichissements...