Optimized PatchMatch for Near Real Time and Accurate Label Fusion

Vinh-Thong Ta (1) , Rémi Giraud (1) , D. Louis Collins (2) , Pierrick Coupé (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique
2 MNI - McConnell Brain Imaging Centre

Automatic segmentation methods are important tools for quantitative analysis of Magnetic Resonance Images. Recently, patch- based label fusion approaches demonstrated state-of-the-art segmenta- tion accuracy. In this paper, we introduce a new patch-based method using the PatchMatch algorithm to perform segmentation of anatomical structures. Based on an Optimized PAtchMatch Label fusion (OPAL) strategy, the proposed method provides competitive segmentation accu- racy in near real time. During our validation on hippocampus segmenta- tion of 80 healthy subjects, OPAL was compared to several state-of-the- art methods. Results show that OPAL obtained the highest median Dice coefficient (89.3%) in less than 1 sec per subject. These results highlight the excellent performance of OPAL in terms of computation time and segmentation accuracy compared to recently published methods.

Mots clés: PatchMatch; Patch-based Segmentation; Hippocampus

Domaines: Traitement des images [eess.IV]

Vinh-Thong Ta, Rémi Giraud, D. Louis Collins, Pierrick Coupé. Optimized PatchMatch for Near Real Time and Accurate Label Fusion. MICCAI 2014, Sep 2014, United States. 8 p. ⟨hal-01006329⟩