Particle Diffusion Matching: Random Walk Correspondence Search for the Alignment of Standard and Ultra-Widefield Fundus Images

2026-04-11

AI Summary

This study addresses the challenging problem of registering standard fundus images (SFIs) with ultra-widefield fundus images (UWFIs), which exhibit significant differences in scale, appearance, and sparse distinctive features. To overcome these challenges, the authors propose a diffusion-guided random walk correspondence search (RWCS) method that iteratively refines particle displacements by integrating local appearance cues, structural distributions, and global transformation estimates to achieve robust cross-modal correspondences. The introduced particle diffusion matching mechanism effectively establishes accurate point-to-point matches even in regions lacking salient features. Experimental results demonstrate that the proposed approach achieves state-of-the-art performance across multiple fundus image registration benchmarks, significantly outperforming existing methods on SFI–UWFI datasets. Furthermore, its clinical utility is validated, offering a novel paradigm for multimodal fundus image fusion.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

We propose a robust alignment technique for Standard Fundus Images (SFIs) and Ultra-Widefield Fundus Images (UWFIs), which are challenging to align due to differences in scale, appearance, and the scarcity of distinctive features. Our method, termed Particle Diffusion Matching (PDM), performs alignment through an iterative Random Walk Correspondence Search (RWCS) guided by a diffusion model. At each iteration, the model estimates displacement vectors for particle points by considering local appearance, the structural distribution of particles, and an estimated global transformation, enabling progressive refinement of correspondences even under difficult conditions. PDM achieves state-of-the-art performance across multiple retinal image alignment benchmarks, showing substantial improvement on a primary dataset of SFI-UWFI pairs and demonstrating its effectiveness in real-world clinical scenarios. By providing accurate and scalable correspondence estimation, PDM overcomes the limitations of existing methods and facilitates the integration of complementary retinal image modalities. This diffusion-guided search strategy offers a new direction for improving downstream supervised learning, disease diagnosis, and multi-modal image analysis in ophthalmology.

Problem

• Fundus Image Alignment
• Standard Fundus Images
• Ultra-Widefield Fundus Images
• Correspondence Search
• Image Registration

Innovation

• Particle Diffusion Matching
• Random Walk Correspondence Search
• Diffusion Model
• Retinal Image Alignment
• Multi-modal Registration