Improving segmentation of retinal arteries and veins using cardiac signal in doppler holograms

2025-11-18

AI Summary

Traditional retinal artery-vein segmentation methods rely solely on spatial features, neglecting the high-temporal-resolution hemodynamic dynamics embedded in Doppler holography data—thereby limiting the accuracy of quantitative hemodynamic analysis. To address this, we propose a pulse-signal-guided spatiotemporal fusion segmentation framework: a lightweight temporal解析 preprocessing module explicitly encodes cardiac pulsatility priors and integrates seamlessly with standard segmentation networks (e.g., U-Net), enabling effective extraction of dynamic flow features without attention mechanisms or iterative refinement. The method preserves architectural simplicity while significantly improving artery-vein segmentation accuracy (≥5.2% absolute gain over baselines) and advancing quantitative retinal hemodynamic assessment. Furthermore, we publicly release the first Doppler holography dataset specifically designed for artery-vein temporal segmentation, enabling reproducible research in dynamic retinal vascular analysis.

Technology Category

• Computer Vision: Segmentation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Doppler holography is an emerging retinal imaging technique that captures the dynamic behavior of blood flow with high temporal resolution, enabling quantitative assessment of retinal hemodynamics. This requires accurate segmentation of retinal arteries and veins, but traditional segmentation methods focus solely on spatial information and overlook the temporal richness of holographic data. In this work, we propose a simple yet effective approach for artery-vein segmentation in temporal Doppler holograms using standard segmentation architectures. By incorporating features derived from a dedicated pulse analysis pipeline, our method allows conventional U-Nets to exploit temporal dynamics and achieve performance comparable to more complex attention- or iteration-based models. These findings demonstrate that time-resolved preprocessing can unlock the full potential of deep learning for Doppler holography, opening new perspectives for quantitative exploration of retinal hemodynamics. The dataset is publicly available at https://huggingface.co/datasets/DigitalHolography/

Problem

• Segmenting retinal arteries and veins using cardiac signals in Doppler holograms
• Exploiting temporal dynamics in holographic data for improved segmentation accuracy
• Enabling quantitative hemodynamic assessment through time-resolved preprocessing techniques

Innovation

• Uses cardiac signal features for artery-vein segmentation
• Incorporates pulse analysis pipeline into U-Net architecture
• Employs time-resolved preprocessing to exploit temporal dynamics