WDT-MD: Wavelet Diffusion Transformers for Microaneurysm Detection in Fundus Images

TL;DR

This paper introduces WDT-MD, a novel wavelet diffusion transformer model that significantly improves microaneurysm detection in fundus images by addressing key limitations of existing diffusion-based methods, enhancing accuracy and robustness.

Contribution

The paper presents a wavelet diffusion transformer framework with noise-encoded conditioning, pseudo-normal pattern synthesis, and multi-scale analysis, advancing automated microaneurysm detection in retinal images.

Findings

Outperforms state-of-the-art methods on IDRiD and e-ophtha MA datasets.

Effectively reduces false positives and improves normal feature reconstruction.

Demonstrates robustness across diverse lesion characteristics.

Abstract

Microaneurysms (MAs), the earliest pathognomonic signs of Diabetic Retinopathy (DR), present as sub-60 $\mu m$ lesions in fundus images with highly variable photometric and morphological characteristics, rendering manual screening not only labor-intensive but inherently error-prone. While diffusion-based anomaly detection has emerged as a promising approach for automated MA screening, its clinical application is hindered by three fundamental limitations. First, these models often fall prey to "identity mapping", where they inadvertently replicate the input image. Second, they struggle to distinguish MAs from other anomalies, leading to high false positives. Third, their suboptimal reconstruction of normal features hampers overall performance. To address these challenges, we propose a Wavelet Diffusion Transformer framework for MA Detection (WDT-MD), which features three key innovations: a noise-encoded image conditioning mechanism to avoid "identity mapping" by perturbing image conditions during training; pseudo-normal pattern synthesis via inpainting to introduce pixel-level supervision, enabling discrimination between MAs and other anomalies; and a wavelet diffusion Transformer architecture that combines the global modeling capability of diffusion Transformers with multi-scale wavelet analysis to enhance reconstruction of normal retinal features. Comprehensive experiments on the IDRiD and e-ophtha MA datasets demonstrate that WDT-MD outperforms state-of-the-art methods in both pixel-level and image-level MA detection. This advancement holds significant promise for improving early DR screening.