Elongated Physiological Structure Segmentation via Spatial and Scale Uncertainty-aware Network
This work addresses segmentation challenges in medical imaging for structures like corneal endothelium and retinal vessels, which is incremental as it builds on existing uncertainty-based methods.
The paper tackles robust segmentation of elongated physiological structures like corneal endothelium and retinal vessels by proposing SSU-Net, which uses spatial and scale uncertainty to highlight ambiguous regions and integrate hierarchical contexts, achieving state-of-the-art performance on these tasks.
Robust and accurate segmentation for elongated physiological structures is challenging, especially in the ambiguous region, such as the corneal endothelium microscope image with uneven illumination or the fundus image with disease interference. In this paper, we present a spatial and scale uncertainty-aware network (SSU-Net) that fully uses both spatial and scale uncertainty to highlight ambiguous regions and integrate hierarchical structure contexts. First, we estimate epistemic and aleatoric spatial uncertainty maps using Monte Carlo dropout to approximate Bayesian networks. Based on these spatial uncertainty maps, we propose the gated soft uncertainty-aware (GSUA) module to guide the model to focus on ambiguous regions. Second, we extract the uncertainty under different scales and propose the multi-scale uncertainty-aware (MSUA) fusion module to integrate structure contexts from hierarchical predictions, strengthening the final prediction. Finally, we visualize the uncertainty map of final prediction, providing interpretability for segmentation results. Experiment results show that the SSU-Net performs best on cornea endothelial cell and retinal vessel segmentation tasks. Moreover, compared with counterpart uncertainty-based methods, SSU-Net is more accurate and robust.