A study on loss functions and decision thresholds for the segmentation of multiple sclerosis lesions on spinal cord MRI - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

A study on loss functions and decision thresholds for the segmentation of multiple sclerosis lesions on spinal cord MRI

Résumé

Multiple sclerosis (MS) patients often present hyper-intense T2-w lesions in the spinal cord. The severe imbalance between background and lesion classes poses a major challenge to Deep Learning segmentation approaches, requiring for ad hoc strategies. Careful selection of the loss function and adjustment of the conventional 0.5-thresholding may help mitigating this issue. Our results show the performance advantages of loss functions based on the Tversky Index and the benefits of threshold tuning over more standard settings and the state-of-the-art model for MS lesion segmentation on spinal cord MRI.
Fichier principal
Vignette du fichier
ISBI_Primus_10052023.pdf (1.01 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03865212 , version 1 (22-11-2022)
hal-03865212 , version 2 (07-03-2023)
hal-03865212 , version 3 (10-05-2023)

Identifiants

  • HAL Id : hal-03865212 , version 3

Citer

Burhan Rashid Hussein, Cédric Meurée, Malo Gaubert, Arthur Masson, Anne Kerbrat, et al.. A study on loss functions and decision thresholds for the segmentation of multiple sclerosis lesions on spinal cord MRI. 20th IEEE International Symposium on Biomedical Imaging (ISBI 2023), Apr 2023, Cartagena (Colombia), Colombia. ⟨hal-03865212v3⟩
213 Consultations
203 Téléchargements

Partager

Gmail Facebook X LinkedIn More