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

TL;DR

This paper presents a vessel-aware deep learning framework for OCTA images that incorporates vascular biomarkers like tortuosity and dropout maps to improve AMD detection and provide physiologically relevant insights.

Contribution

The study introduces an external multiplicative attention model that integrates vessel-specific biomarkers into deep learning for AMD detection from OCTA images, enhancing interpretability.

Findings

Arterial tortuosity is a strong discriminative biomarker.

Capillary dropout maps improve detection at larger scales.

The method offers physiologically meaningful insights.

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.