TY - JOUR
T1 - Dynamic graph convolutional autoencoder with node-attribute-wise attention for kidney and tumor segmentation from CT volumes
AU - Xuan, Ping
AU - Cui, Hui
AU - Zhang, Hongda
AU - Zhang, Tiangang
AU - Wang, Linlin
AU - Nakaguchi, Toshiya
AU - Duh, Henry B.L.
N1 - Funding Information:
This work is supported by the Natural Science Foundation of China ( 61972135 ), the Natural Science Foundation of Heilongjiang Province ( LH2019F049 ), the China Postdoctoral Science Foundation ( 2019M650069 ), and the Fundamental Research Foundation of Universities in Heilongjiang Province for Technology Innovation ( KJCX201805 ).
Publisher Copyright:
© 2021
PY - 2022/1/25
Y1 - 2022/1/25
N2 - Extraction and integration of semantic connections, spatial relations and dependencies are critical in volumetric image segmentation. This is a challenging issue, especially when there are long-distance objects with close semantic relations and neighboring objects with indistinct boundaries. We propose a novel dynamic graph convolution (DGC) autoencoder with node-attribute-wise attention (NodeAttri-Attention) for relation inference and reasoning, with applications on kidney and tumor segmentation from computerized tomography (CT) volumes. We first introduce a new graph construction strategy for 3D volumetric image data, where graph node attributes and connections represent topological relations and high-level correlations. Then NodeAttri-Attention mechanism is proposed to obtain attention-enhanced node attributes by discriminating adaptive contributions of various features. Finally, the DGC strategy is designed to learn and integrate the complex and underlying correlations across image regions. Our DGC dynamically updates graph topology and node attributes as the graph convolutional layer gradually deepens. Experimental results and ablation studies demonstrated the effectiveness of each of our major innovations in NodeAttri-Attention DGC, especially when objects are with weak boundaries, irregular shapes, and various sizes. The improved segmentation results of embedding NodeAttri-Attention DGC to different segmentation backbones show the generality of DGC autoencoder.
AB - Extraction and integration of semantic connections, spatial relations and dependencies are critical in volumetric image segmentation. This is a challenging issue, especially when there are long-distance objects with close semantic relations and neighboring objects with indistinct boundaries. We propose a novel dynamic graph convolution (DGC) autoencoder with node-attribute-wise attention (NodeAttri-Attention) for relation inference and reasoning, with applications on kidney and tumor segmentation from computerized tomography (CT) volumes. We first introduce a new graph construction strategy for 3D volumetric image data, where graph node attributes and connections represent topological relations and high-level correlations. Then NodeAttri-Attention mechanism is proposed to obtain attention-enhanced node attributes by discriminating adaptive contributions of various features. Finally, the DGC strategy is designed to learn and integrate the complex and underlying correlations across image regions. Our DGC dynamically updates graph topology and node attributes as the graph convolutional layer gradually deepens. Experimental results and ablation studies demonstrated the effectiveness of each of our major innovations in NodeAttri-Attention DGC, especially when objects are with weak boundaries, irregular shapes, and various sizes. The improved segmentation results of embedding NodeAttri-Attention DGC to different segmentation backbones show the generality of DGC autoencoder.
KW - Dynamic graph convolutional autoencoder
KW - Graph node attributes
KW - Kidney and tumor segmentation
KW - Long-distance relationship between nodes
KW - Node-attribute-wise attention
UR - http://www.scopus.com/inward/record.url?scp=85119173699&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2021.107360
DO - 10.1016/j.knosys.2021.107360
M3 - Journal article
AN - SCOPUS:85119173699
SN - 0950-7051
VL - 236
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
M1 - 107360
ER -