Prior-AttUNet: Retinal OCT Fluid Segmentation Based on Normal Anatomical Priors and Attention Gating

TL;DR

Prior-AttUNet is a novel segmentation model that combines anatomical priors and attention mechanisms to accurately delineate retinal fluid regions in OCT images across different devices.

Contribution

It introduces a hybrid dual-path architecture with generative anatomical priors and a triple-attention mechanism, improving segmentation accuracy and robustness in OCT imaging.

Findings

Achieves high Dice scores on multiple OCT devices

Maintains low computational cost of 0.37 TFLOPs

Demonstrates robustness across heterogeneous datasets

Abstract

Accurate segmentation of macular edema, a hallmark pathological feature in vision-threatening conditions such as age-related macular degeneration and diabetic macular edema, is essential for clinical diagnosis and management. To overcome the challenges of segmenting fluid regions in optical coherence tomography (OCT) images-notably ambiguous boundaries and cross-device heterogeneity-this study introduces Prior-AttUNet, a segmentation model augmented with generative anatomical priors. The framework adopts a hybrid dual-path architecture that integrates a generative prior pathway with a segmentation network. A variational autoencoder supplies multi-scale normative anatomical priors, while the segmentation backbone incorporates densely connected blocks and spatial pyramid pooling modules to capture richer contextual information. Additionally, a novel triple-attention mechanism, guided by anatomical priors, dynamically modulates feature importance across decoding stages, substantially enhancing boundary delineation. Evaluated on the public RETOUCH benchmark, Prior-AttUNet achieves excellent performance across three OCT imaging devices (Cirrus, Spectralis, and Topcon), with mean Dice similarity coefficients of 93.93%, 95.18%, and 93.47%, respectively. The model maintains a low computational cost of 0.37 TFLOPs, striking an effective balance between segmentation precision and inference efficiency. These results demonstrate its potential as a reliable tool for automated clinical analysis.