Dual encoding feature filtering generalized attention UNET for retinal vessel segmentation

2025-06-02

AI Summary

Retinal vessel segmentation suffers from limited training samples, severe class imbalance, and inadequate feature representation, leading to poor generalizability and high false-positive rates. To address these challenges, we propose a novel U-Net variant featuring three key innovations: (1) a dual-encoder architecture that separately models domain-invariant preprocessed features and raw-image semantics; (2) a feature filtering fusion module and an attention-guided feature reconstruction fusion module—replacing conventional skip connections to enhance multi-scale feature selectivity and contextual awareness; and (3) a customized data augmentation and resampling strategy to mitigate distribution shift and class imbalance. Evaluated on five benchmark datasets—DRIVE, CHASE_DB1, STARE, HRF, and IOSTAR—our method consistently outperforms baseline models and state-of-the-art approaches, demonstrating substantial improvements in cross-dataset generalization performance.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Feature Construction/Reformulation
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Retinal blood vessel segmentation is crucial for diagnosing ocular and cardiovascular diseases. Although the introduction of U-Net in 2015 by Olaf Ronneberger significantly advanced this field, yet issues like limited training data, imbalance data distribution, and inadequate feature extraction persist, hindering both the segmentation performance and optimal model generalization. Addressing these critical issues, the DEFFA-Unet is proposed featuring an additional encoder to process domain-invariant pre-processed inputs, thereby improving both richer feature encoding and enhanced model generalization. A feature filtering fusion module is developed to ensure the precise feature filtering and robust hybrid feature fusion. In response to the task-specific need for higher precision where false positives are very costly, traditional skip connections are replaced with the attention-guided feature reconstructing fusion module. Additionally, innovative data augmentation and balancing methods are proposed to counter data scarcity and distribution imbalance, further boosting the robustness and generalization of the model. With a comprehensive suite of evaluation metrics, extensive validations on four benchmark datasets (DRIVE, CHASEDB1, STARE, and HRF) and an SLO dataset (IOSTAR), demonstrate the proposed method's superiority over both baseline and state-of-the-art models. Particularly the proposed method significantly outperforms the compared methods in cross-validation model generalization.

Problem

• Improves retinal vessel segmentation accuracy and generalization
• Addresses limited training data and imbalance distribution
• Enhances feature extraction with dual encoding and filtering

Innovation

• Dual encoder for richer feature encoding
• Feature filtering fusion module for precision
• Attention-guided feature reconstructing fusion