RV-ESA

RV-ESA analyzes retinal vessel shapes using elastic shape analysis to quantify geometric changes in retinal vasculature associated with diabetic retinopathy.

Key Features:

• Shape Modeling Method: Produces precise elastic geometric representations of retinal vessels for quantitative analysis.
• Statistical Analysis Tools: Computes Graph Mean and Graph Principal Component Analysis (PCA) for quantitative assessment of vascular shape variation.
• Geodesic Visualization Tool: Computes and visualizes geodesic deformation paths to represent vessel shape changes over time.

Scientific Applications:

• Progression Quantification: Quantifies geometric changes in retinal vasculature associated with diabetic retinopathy severity, with Graph PCA scores reported to negatively correlate with DR grades.
• Validation and Cohort Studies: Applied to the EyePACS dataset and a longitudinal proprietary dataset of 98 patients to support population-level and longitudinal analyses.
• Morphometric Research for Early Detection: Enables detailed morphometric studies of microvascular alterations to support research into early detection and monitoring of diabetic retinopathy.

Methodology:

Generates elastic geometric representations of retinal vessels, computes Graph Mean and Graph PCA for statistical analysis, and uses geodesic computations to visualize vessel deformations.