Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Vessel-Aware Deep Learning for OCTA-Based Detection of AMD

Margalit G. Mitzner, Moinak Bhattacharya, Zhilin Zou, Chao Chen, Prateek Prasanna

TL;DR

An external multiplicative attention framework that incorporates vessel-specific tortuosity maps and vasculature dropout maps derived from arteries, veins, and capillaries is introduced that offers interpretable insights aligned with known AMD pathophysiology.

Abstract

Age-related macular degeneration (AMD) is characterized by early micro-vascular alterations that can be captured non-invasively using optical coherence tomography angiography (OCTA), yet most deep learning (DL) models rely on global features and fail to exploit clinically meaningful vascular biomarkers. We introduce an external multiplicative attention framework that incorporates vessel-specific tortuosity maps and vasculature dropout maps derived from arteries, veins, and capillaries. These biomarker maps are generated from vessel segmentations and smoothed across multiple spatial scales to highlight coherent patterns of vascular remodeling and capillary rarefaction. Tortuosity reflects abnormalities in vessel geometry linked to impaired auto-regulation, while dropout maps capture localized perfusion deficits that precede structural retinal damage. The maps are fused with the OCTA projection to guide a deep classifier toward physiologically relevant regions. Arterial tortuosity provided the most consistent discriminative value, while capillary dropout maps performed best among density-based variants, especially at larger smoothing scales. Our proposed method offers interpretable insights aligned with known AMD pathophysiology.

Vessel-Aware Deep Learning for OCTA-Based Detection of AMD

TL;DR

An external multiplicative attention framework that incorporates vessel-specific tortuosity maps and vasculature dropout maps derived from arteries, veins, and capillaries is introduced that offers interpretable insights aligned with known AMD pathophysiology.

Abstract

Age-related macular degeneration (AMD) is characterized by early micro-vascular alterations that can be captured non-invasively using optical coherence tomography angiography (OCTA), yet most deep learning (DL) models rely on global features and fail to exploit clinically meaningful vascular biomarkers. We introduce an external multiplicative attention framework that incorporates vessel-specific tortuosity maps and vasculature dropout maps derived from arteries, veins, and capillaries. These biomarker maps are generated from vessel segmentations and smoothed across multiple spatial scales to highlight coherent patterns of vascular remodeling and capillary rarefaction. Tortuosity reflects abnormalities in vessel geometry linked to impaired auto-regulation, while dropout maps capture localized perfusion deficits that precede structural retinal damage. The maps are fused with the OCTA projection to guide a deep classifier toward physiologically relevant regions. Arterial tortuosity provided the most consistent discriminative value, while capillary dropout maps performed best among density-based variants, especially at larger smoothing scales. Our proposed method offers interpretable insights aligned with known AMD pathophysiology.
Paper Structure (12 sections, 3 equations, 3 figures, 2 tables)

This paper contains 12 sections, 3 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Tortuosity Heatmap Computation
  4. Vessel Density Computation
  5. Classification of AMD vs Normal Eyes
  6. External Multiplicative Attention from Tortuosity and Density Heatmaps
  7. Experiments and Results
  8. Experiments
  9. Results and Discussion
  10. Conclusion
  11. Compliance with Ethical Standards
  12. Acknowledgements

Figures (3)

  • Figure 1: Overview of the proposed pipeline vessel segmentations are used to generate tortuosity or density heatmaps (multi-$\sigma$), which are then separately fused with the full OCTA projection via multiplicative weighting and classified using a ResNet-18 backbone for AMD detection.
  • Figure 2: Attention heatmaps. Segmentation labels, tortuosity dot maps showing relative tortuosity of each segment, high tortuosity (red hotspots indicate high tortuosity) and low vessel density heatmaps (red hotspots indicate low density).
  • Figure 3: AUC-ROC vs. $\sigma$ for vessel tortuosity (A) and density (B) heatmap-attentions are shown. GradCAM results are shown in C.