Ambiguity-aware Truncated Flow Matching for Ambiguous Medical Image Segmentation
Authors
Affiliations (4)
Affiliations (4)
- 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, Case Western Reserve University, Cleveland, Ohio 44106, United States
Abstract
A simultaneous enhancement of accuracy and diversity of predictions remains a challenge in ambiguous medical image segmentation (AMIS) due to the inherent trade-offs. While truncated diffusion probabilistic models (TDPMs) hold strong potential with a paradigm optimization, existing TDPMs suffer from entangled accuracy and diversity of predictions with insufficient fidelity and plausibility. To address the aforementioned challenges, we propose Ambiguity-aware Truncated Flow Matching (ATFM), which introduces a novel inference paradigm and dedicated model components. Firstly, we propose Data-Hierarchical Inference, a redefinition of AMIS-specific inference paradigm, which enhances accuracy and diversity at data-distribution and data-sample level, respectively, for an effective disentanglement. Secondly, Gaussian Truncation Representation (GTR) is introduced to enhance both fidelity of predictions and reliability of truncation distribution, by explicitly modeling it as a Gaussian distribution at $T_{\text{trunc}}$ instead of using sampling-based approximations.Thirdly, Segmentation Flow Matching (SFM) is proposed to enhance the plausibility of diverse predictions by extending semantic-aware flow transformation in Flow Matching (FM). Comprehensive evaluations on LIDC and ISIC3 datasets demonstrate that ATFM outperforms SOTA methods and simultaneously achieves a more efficient inference. ATFM improves GED and HM-IoU by up to $12\%$ and $7.3\%$ compared to advanced methods.