KANResDiff: Learning Local Residual Diffusion via Kolmogorov-Arnold Network for Ambiguous Medical Image Segmentation
Authors: Fanding Li, Chenglin Wang, Xiangyu Li, Xingyu Qiu, Xinghua Ma, Xiangming Yin, Haiyang Li, Suyu Dong, +4 more
Organizations: Faculty of Computing, Harbin Institute of Technology, Harbin, China · College of Computer and Control Engineering, Northeast Forestry University, Harbin, China · Faculty of Computing, Harbin Institute of Technology, Shenzhen, China · Department of Computer and Data Science and Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio 44106, United States
Ambiguous medical image segmentation aims to provide a series of diverse but plausible segmentation hypotheses. However, existing methods introduce stochasticity in a fixed and pre-defined manner, failing to form a progressive semantic modeling process. To address these challenges, we propose KANResDiff to learn local residual diffusion with Kolmogorov-Arnold Network, thereby assigning distinct roles across stages for ambiguity modeling. Specifically, we propose Independent Time Encoding that offers spline-based time embeddings instead of linear ones from MLPs, which enhances the independence across inference stages and assigns progressive semantic roles to different stages. We propose Residual Schrodinger Bridge that injects deterministic residual prior with learnable weights by constructing local Schrodinger Bridge instead of following manually settings, achieving a flexible deterministic-stochastic interaction and stage-aware ambiguity modeling thanks to local optimal diffusion path. Extensive experimental results on two public datasets demonstrate that KANResDiff achieves SOTA performance on GED and HM-IoU, with maximum improvements of 16.8% and 7.7%, respectively, while maintaining competitive performance on the MDM metric. Source code is available at https://github.com/PerceptionComputingLab/KANResDiff.