Ambiguous Medical Image Segmentation Using Diffusion Schrödinger Bridge

2025-09-21

AI Summary

Medical image segmentation faces challenges from ambiguous lesion boundaries and inter-annotator variability, leading to inherent uncertainty. To address this, we introduce Schrödinger Bridge (SB) theory—the first such application in medical segmentation—formulating a joint image-mask dynamic evolution model. We propose the Diversity Divergence Index (D_DDI), an unsupervised metric that quantifies and regularizes annotation discrepancies without auxiliary supervision, thereby preserving predictive diversity while enhancing structural consistency. Furthermore, we design a diffusion-based SB framework coupled with a novel joint loss function. Evaluated on three benchmarks—LIDC-IDRI, COCA, and RACER—our method achieves state-of-the-art performance, particularly in boundary accuracy, morphological integrity, and clinical credibility. This work establishes a new paradigm for robust, uncertainty-aware medical segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Accurate segmentation of medical images is challenging due to unclear lesion boundaries and mask variability. We introduce emph{Segmentation Schödinger Bridge (SSB)}, the first application of Schödinger Bridge for ambiguous medical image segmentation, modelling joint image-mask dynamics to enhance performance. SSB preserves structural integrity, delineates unclear boundaries without additional guidance, and maintains diversity using a novel loss function. We further propose the emph{Diversity Divergence Index} ($D_{DDI}$) to quantify inter-rater variability, capturing both diversity and consensus. SSB achieves state-of-the-art performance on LIDC-IDRI, COCA, and RACER (in-house) datasets.

Problem

• Segments medical images with unclear lesion boundaries and mask variability
• Models joint image-mask dynamics to enhance segmentation performance
• Quantifies inter-rater variability while preserving structural integrity

Innovation

• Schrödinger Bridge models image-mask dynamics
• Novel loss function preserves structure and diversity
• Diversity Divergence Index quantifies inter-rater variability