Retinal Fundus Image Registration via Vascular Structure Graph Matching

TLDR

Retinal fundus images contain unique geometric structures in their vascular trees that can serve as distinctive features for matching. This study proposes a graph-based registration framework, GM‑ICP, to align pairwise retinal images. The method automatically extracts vessel graphs, performs graph matching to obtain global correspondences, refines alignment with a revised ICP using a quadratic transformation, and applies a structure‑based sample consensus (STRUCT‑SAC) to discard incorrect matches. GM‑ICP achieves a global optimum solution, is invariant to linear geometric transformations, does not require heavy local feature descriptors, and its effectiveness is shown on 48 clinical retinal image pairs.

Abstract

Motivated by the observation that a retinal fundus image may contain some unique geometric structures within its vascular trees which can be utilized for feature matching, in this paper, we proposed a graph‐based registration framework called GM‐ICP to align pairwise retinal images. First, the retinal vessels are automatically detected and represented as vascular structure graphs. A graph matching is then performed to find global correspondences between vascular bifurcations. Finally, a revised ICP algorithm incorporating with quadratic transformation model is used at fine level to register vessel shape models. In order to eliminate the incorrect matches from global correspondence set obtained via graph matching, we proposed a structure‐based sample consensus (STRUCT‐SAC) algorithm. The advantages of our approach are threefold: (1) global optimum solution can be achieved with graph matching; (2) our method is invariant to linear geometric transformations; and (3) heavy local feature descriptors are not required. The effectiveness of our method is demonstrated by the experiments with 48 pairs retinal images collected from clinical patients.