OCT-OCTA segmentation: combining structural and blood flow information to segment Bruch's membrane.

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Biology

Background:

• Bruch's membrane (BM) segmentation is crucial for diagnosing retinal diseases.
• Traditional methods struggle when pathology distorts the relationship between BM and the retinal pigment epithelium (RPE).
• Optical coherence tomography angiography (OCTA) offers new insights but requires robust segmentation techniques.

Purpose of the Study:

• To present a fully automated graph-based segmentation algorithm for Bruch's membrane (BM).
• To leverage both optical coherence tomography (OCT) and OCT angiography (OCTA) data for improved BM segmentation.
• To address challenges in BM segmentation caused by pathological distortions.

Main Methods:

• A novel graph-based segmentation algorithm was developed.
• The algorithm integrates data from OCT and OCTA.
• Validation was performed against manual segmentation in 18 eyes across various retinal conditions.

Main Results:

• The algorithm achieved a mean absolute error of approximately 0.91 pixels (4.1 μm) compared to manual segmentation.
• Successful segmentation was demonstrated in healthy controls and patients with diabetic retinopathy (DR), age-related macular degeneration (AMD), and choroidal neovascularization (CNV).
• The method proved effective even when the RPE was not a reliable surrogate for BM due to pathology.

Conclusions:

• The proposed OCT-OCTA segmentation algorithm provides accurate and automated BM segmentation.
• This technique is valuable for analyzing retinal pathologies where standard methods fail.
• The framework supports the increasing use of OCTA in ophthalmology research and clinical practice.