A Novel OCTA-Based Image Processing Method for Enhanced Microaneurysm Detection and 3D Visualization in Diabetic Retinopathy.

Abstract

The early detection of microaneurysms (MA) in diabetic retinopathy (DR) is crucial for disease management. Although optical coherence tomography angiography (OCTA) offers non-invasive, high-resolution imags, it still suffers from insufficient sensitivity and limitations of two-dimensional imaging in MA's detection. This study proposes an innovative processing method based on retinal OCTA images that integrates blood flow signals with the hyperreflective features of MA in structural OCT, combined with three-dimensional reconstruction technology, to construct a retinal vascular image (3D-MA map) that enhances MA's detection rate and provides 3D visualization. Using the open-source software ImageJ, this method enables visual analysis of the spatial morphology and arteriovenous origin of MAs. The results show that the 3D-MA image preserves three-dimensional stereoscopic effects while enhancing the display of vascular abnormalities, particularly improving the recognition of MAs that are challenging to detect with OCTA alone. This study provides a powerful imaging tool for early diagnosis and mechanistic research of diabetic retinopathy.