3D Retinal Vessel Density Mapping With OCT-Angiography.

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Optical Coherence Tomography Angiography (OCTA) offers high-resolution, non-invasive imaging of retinal capillaries.
• Current OCTA analysis often relies on limited 2D summary statistics, hindering detailed longitudinal and cross-population comparisons.
• There is a need for advanced methods to analyze 3D retinal vasculature for localized disease assessment.

Purpose of the Study:

• To develop a novel 3D approach for mapping retinal vessel density from OCTA images.
• To enable localized and longitudinal comparisons of retinal vasculature.
• To assess microvascular changes in conditions like diabetic retinopathy and aging.

Main Methods:

• A 3D vessel network representation was generated using curvelet-based denoising and optimally oriented flux (OOF).
• A 3D vessel density image (VDI) was created by diffusing vessel masks.
• Non-linear 3D OCT image registration was employed for localized cross-subject comparisons.

Main Results:

• The method successfully detected localized vascular changes in synthetic data with simulated capillary dropout.
• Quantitative validation on normal control and diabetic retinopathy datasets showed detection of age-related and disease-specific changes.
• The framework demonstrated efficacy in detecting microvascular alterations even in the presence of edema.

Conclusions:

• The proposed 3D retinal vessel density mapping framework provides a powerful tool for localized microvascular analysis.
• This method facilitates the monitoring of retinal disease progression and the detection of subtle vascular changes.
• The approach holds significant potential for clinical applications in ophthalmology and retinal research.