Knowledge distillation from multi-modal to mono-modal segmentation networks - Laboratoire Traitement et Communication de l'Information Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Knowledge distillation from multi-modal to mono-modal segmentation networks

Résumé

The joint use of multiple imaging modalities for medical image segmentation has been widely studied in recent years. The fusion of information from different modalities has demonstrated to improve the segmentation accuracy, with respect to mono-modal segmentations, in several applications. However, acquiring multiple modalities is usually not possible in a clinical setting due to a limited number of physicians and scanners, and to limit costs and scan time. Most of the time, only one modality is acquired. In this paper, we propose KD-Net, a framework to transfer knowledge from a trained multi-modal network (teacher) to a mono-modal one (student). The proposed method is an adaptation of the generalized distillation framework where the student network is trained on a subset (1 modality) of the teacher's inputs (n modalities). We illustrate the effectiveness of the proposed framework in brain tumor segmentation with the BraTS 2018 dataset. Using different architectures, we show that the student network effectively learns from the teacher and always outperforms the baseline mono-modal network in terms of seg-mentation accuracy.
Fichier principal
Vignette du fichier
paper2614.pdf (433.61 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02899529 , version 1 (15-07-2020)

Identifiants

  • HAL Id : hal-02899529 , version 1

Citer

Minhao Hu, Matthis Maillard, Ya Zhang, Tommaso Ciceri, Giammarco La Barbera, et al.. Knowledge distillation from multi-modal to mono-modal segmentation networks. MICCAI, Oct 2020, Lima, Peru. pp.772-781. ⟨hal-02899529⟩
395 Consultations
959 Téléchargements

Partager

Gmail Facebook X LinkedIn More