OLVA: Optimal Latent Vector Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

OLVA: Optimal Latent Vector Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation

Résumé

This paper addresses the domain shift problem for segmentation. As a solution, we propose OLVA, a novel and lightweight unsupervised domain adaptation method based on a Variational Auto-Encoder (VAE) and optimal transport (OT) theory. Thanks to the VAE, our model learns a shared cross-domain latent space that follows a normal distribution, which reduces the domain shift. To guarantee valid segmentations, our shared latent space is designed to model the shape rather than the intensity variations. We further rely on an OT loss to match and align the remaining discrepancy between the two domains in the latent space. We demonstrate OLVA's effectiveness for the segmentation of multiple cardiac structures on the public Multi-Modality Whole Heart Segmentation (MM-WHS) dataset, where the source domain consists of annotated 3D MR images and the unlabelled target domain of 3D CTs. Our results show remarkable improvements with an additional margin of 12.5% dice score over concurrent generative training approaches.
Fichier principal
Vignette du fichier
miccai2021.pdf (897.93 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03261428 , version 1 (15-06-2021)

Identifiants

  • HAL Id : hal-03261428 , version 1

Citer

Dawood Al Chanti, Diana Mateus. OLVA: Optimal Latent Vector Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation. the 24th International Conference on Medical Image Computing and Computer Assisted Intervention, Sep 2021, Strasbourg (virtuel), France. ⟨hal-03261428⟩
157 Consultations
233 Téléchargements

Partager

Gmail Facebook X LinkedIn More