Paper Info
Title
3D Graph-S(2)Net: Shape-Aware Self-ensembling Network for Semi-supervised Segmentation with Bilateral Graph Convolution
Abstract
Semi-supervised learning (SSL) algorithms have attracted much attentions in medical image segmentation due to challenge in acquiring pixel-wise annotations by using unlabeled data. However, most of existing SSLs neglected the geometric shape constraint in object, leading to unsatisfactory boundary and non-smooth of object. In this paper, we propose a shape-aware semi-supervised 3D medical image segmentation network, named 3D Graph-S(2)Net, which incorporates the flexible shape information and learns duality constraints between semantics and geometrics in the graph domain. Specifically, our method consists of two parts: a multi-task learning network (3D S(2)Net) and a graph-based cross-task module (3D BGCM). The 3D S(2)Net improves the existing self-ensembling model (i.e., Mean-Teacher model) by adding a signed distance map (SDM) prediction task, which encodes richer features of object shape and surface. Moreover, the 3D BGCM explores the co-occurrence relations between the semantics segmentation and SDM prediction task, so that the network learns stronger semantic and geometric correspondences from both labeled and unlabeled data. Experimental results on the Atrial Segmentation Challenge confirm that our 3D Graph-S(2)Net outperforms the state-of-the-arts in semi-supervised segmentation.
Year
DOI
Venue
2021
10.1007/978-3-030-87196-3_39
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2021, PT II
Keywords
DocType
Volume
Semi-supervised segmentation task, Signed distance map, 3D bilateral graph convolution
Conference
12902
ISSN
Citations 
PageRank 
0302-9743
1
0.38
References 
Authors
0
8
Name
Order
Citations
PageRank
Huimin Huang111.73
Nan Zhou210.72
Lanfen Lin348.67
Hongjie Hu4119.50
Yutaro Iwamoto51317.95
Xian-Hua Han61410.19
Yen-Wei Chen7720155.73
Ruofeng Tong846649.69