SPIRONet: Spatial-Frequency Learning and Topological Channel Interaction Network for Vessel Segmentation
This addresses the problem of accurate and fast vessel segmentation for medical interventional navigation systems, though it appears incremental as it builds on existing deep learning approaches with novel modules.
The paper tackles vessel segmentation in intraoperative images, which suffer from low signal-to-noise ratio and interference, by proposing SPIRONet, a network that achieves state-of-the-art performance on multiple datasets and runs at 21 FPS, exceeding clinical real-time requirements.
Automatic vessel segmentation is paramount for developing next-generation interventional navigation systems. However, current approaches suffer from suboptimal segmentation performances due to significant challenges in intraoperative images (i.e., low signal-to-noise ratio, small or slender vessels, and strong interference). In this paper, a novel spatial-frequency learning and topological channel interaction network (SPIRONet) is proposed to address the above issues. Specifically, dual encoders are utilized to comprehensively capture local spatial and global frequency vessel features. Then, a cross-attention fusion module is introduced to effectively fuse spatial and frequency features, thereby enhancing feature discriminability. Furthermore, a topological channel interaction module is designed to filter out task-irrelevant responses based on graph neural networks. Extensive experimental results on several challenging datasets (CADSA, CAXF, DCA1, and XCAD) demonstrate state-of-the-art performances of our method. Moreover, the inference speed of SPIRONet is 21 FPS with a 512x512 input size, surpassing clinical real-time requirements (6~12FPS). These promising outcomes indicate SPIRONet's potential for integration into vascular interventional navigation systems. Code is available at https://github.com/Dxhuang-CASIA/SPIRONet.