Removing Segmentation Inconsistencies with Semi-Supervised Non-Adjacency Constraint - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Medical Image Analysis Année : 2019

Removing Segmentation Inconsistencies with Semi-Supervised Non-Adjacency Constraint

Résumé

The advent of deep learning has pushed medical image analysis to new levels, rapidly replacing more traditional machine learning and computer vision pipelines. However segmenting and labelling anatomical regions remains challenging owing to appearance variations, imaging artifacts, the paucity and variability of annotated data, and the difficulty of fully exploiting domain constraints such as anatomical knowledge about inter-region relationships. We address the last point, improving the network's region-labeling consistency by introducing NonAdjLoss, an adjacency-graph based auxiliary training loss that penalizes outputs containing regions with anatomically-incorrect ad-jacency relationships. NonAdjLoss supports both fully-supervised training and a semi-supervised extension in which it is applied to unlabeled supplementary training data. The approach substantially reduces segmentation anomalies on the MICCAI-2012, IB-SRv2 brain MRI datasets and the Anatomy3 whole body CT dataset, especially when semi-supervised training is included.
Fichier principal
Vignette du fichier
ganaye2019-postprint_cmp.pdf (996.29 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02275956 , version 1 (05-09-2019)

Identifiants

Citer

Pierre-Antoine Ganaye, Michaël Sdika, Bill Triggs, Hugues Benoit-Cattin. Removing Segmentation Inconsistencies with Semi-Supervised Non-Adjacency Constraint. Medical Image Analysis, 2019, 58, pp.101551. ⟨10.1016/j.media.2019.101551⟩. ⟨hal-02275956⟩
462 Consultations
683 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More