LSR-Diff: A Diffusion Model Synthesizing Level Set Representations for Reliable Segmentation of Medical Images With Ambiguous Edges
IEEE Transactions on Image Processing · 2026 · 35 · 7277-7292
DOI: 10.1109/tip.2026.3707793 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2026. Verfasst von Wenbo Gao und 3 weiteren Autor:innen. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shenzhen Science and Technology Program.
Abstract (Originalsprache)
Accurate boundary segmentation is critical for high-stakes applications such as disease diagnosis, yet remains challenging due to complex topology, boundary ambiguity, and annotation uncertainty. Diffusion Probabilistic Models (DPMs) generate multiple masks with inherent uncertainty, enhancing boundary delineation compared to deterministic models. However, most existing DPM based segmentation approaches learn discrete binary masks, conflicting with the continuous diffusion process and leading to hard-to-learn degradation during noise addition. Moreover, the prevailing approach of averaging stochastic predictions and applying a fixed threshold disregards structural consistency, often leading to imprecise boundaries, isolated artifacts, and holes. To address these challenges, we propose the $L$ evel $S$ et $R$ epresentation $D$ iffusion model (LSR-Diff), which incorporates a diffusion model with a hybrid mask representation to better capture boundary information, and a novel strategy $E$ nsemble $A$ ggregation via Level $S$ et $E$ volution (EASE) to merge prediction candidates while respecting structural information. The hybrid representation takes advantage of both discrete binary masks and continuous implicit masks, with an intermediate representation to ensure a smooth transition. The EASE module guided by ambiguity estimation and anatomical structure then refines boundary topology, preventing arbitrary mask assembly during the aggregation of stochastic predictions. We conduct extensive experiments across various clinical applications including multiple modalities and tissues, showing that LSR-Diff achieves competitive overall performance and improved edge quality and topology accuracy on the tested tasks.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- National Natural Science Foundation of China
- Shenzhen Science and Technology Program
Zitieren
APA
Gao, W., Cao, H., Cheung, J. C., & Yang, Y. (2026). LSR-Diff: A Diffusion Model Synthesizing Level Set Representations for Reliable Segmentation of Medical Images With Ambiguous Edges. IEEE Transactions on Image Processing, 35, 7277-7292. https://doi.org/10.1109/tip.2026.3707793
Vancouver
Gao W, Cao H, Cheung JC, Yang Y. LSR-Diff: A Diffusion Model Synthesizing Level Set Representations for Reliable Segmentation of Medical Images With Ambiguous Edges. IEEE Transactions on Image Processing. 2026;35:7277-7292. doi:10.1109/tip.2026.3707793
BibTeX
@article{Gao2026,
author = {Gao, Wenbo and Cao, Haoyu and Cheung, James Chung-Wai and Yang, Yunyun},
title = {LSR-Diff: A Diffusion Model Synthesizing Level Set Representations for Reliable Segmentation of Medical Images With Ambiguous Edges},
journal = {IEEE Transactions on Image Processing},
year = {2026},
volume = {35},
pages = {7277--7292},
doi = {10.1109/tip.2026.3707793}
}
EUVIMED