SAIL: Structure-Aware Interpretable Learning for Anatomy-Aligned Post-hoc Explanations in OCT

2026-05-04

AI Summary

Existing post-hoc explainable AI methods struggle to accurately delineate lesion structures, respect anatomical boundaries, and suppress noise in OCT images, leading to insufficient clinical trustworthiness. This work proposes the SAIL framework, which uniquely integrates retinal layer-wise anatomical priors with deep semantic features at the representation level, enabling the generation of anatomically aligned and structurally coherent attribution maps without modifying existing explainers. Evaluated across multiple OCT datasets, SAIL significantly enhances explanation quality, yielding results consistent with clinical understanding. Ablation studies further confirm that the fusion of structural and semantic information is crucial for improving the plausibility and fidelity of explanations.

Abstract

Optical coherence tomography (OCT), a commonly used retinal imaging modality, plays a central role in retinal disease diagnosis by providing high-resolution visualization of retinal layers. While deep learning (DL) has achieved expert-level accuracy in OCT-based retinal disease detection, its "black box" nature poses challenges for clinical adoption, where explainability is essential for clinical trust and regulatory approval. Existing post-hoc explainable AI (XAI) methods often struggle to delineate fine-grained lesion structures, respect anatomical boundaries, or suppress noise, limiting the trustworthiness of their explanations. To bridge these gaps, we propose a Structure-Aware Interpretable Learning (SAIL) framework that integrates retinal anatomical priors at the representation level and couples them with semantic features via a fusion design. Without modifying standard post-hoc explainability methods, this representation yields sharper and more anatomically aligned attribution maps. Comprehensive experiments on diverse OCT datasets demonstrate that our structure-aware method consistently enhances interpretability, producing clinically meaningful and anatomy-aware explanations. Ablation studies further show that strong interpretability requires both structural priors and semantic features, and that properly fusing the two is critical to achieve the best explanation quality. Together, these results highlight structure-aware representations as a key step toward reliable explainability in OCT.

Problem

• explainable AI
• optical coherence tomography
• anatomical alignment
• post-hoc explanation
• interpretability

Innovation

• Structure-Aware Learning
• Anatomy-Aligned Explanation
• Post-hoc XAI
• Retinal OCT
• Interpretable Deep Learning