OCTA-Based Biomarker Characterization in nAMD
MAria Simona Tivadar, Ioana Damian, Adrian Groza, Simona Delia Nicoara
TL;DR
This study addresses improving nAMD diagnosis via OCTA by introducing a biomarker-focused pipeline, 3D visualizations, and a white-box ensemble of interpretable classifiers. The workflow extracts $A_{mCNV}$, $A_{Total}$, and VesselDensity $= rac{A_{mCNV}}{A_{Total}}$, supported by a fixed three-phase processing chain and a 3D representation of neovascularization. The white-box models (DT, SVM, DL-Learner) provide transparent decision rules, achieving $100\%$ training accuracy and $68\%$ testing accuracy, with high clinician–algorithm agreement (Jaccard $=0.8912$, Dice $=0.92$) against expert assessments. The work emphasizes explainability and clinical relevance, envisioning reliable decision support with potential for broader applicability across OCTA datasets and devices.
Abstract
We aim to enhance ophthalmologists' decision-making when diagnosing the Neovascular Age-Related Macular Degeneration (nAMD). We developed three tools to analyze Optical Coherence Tomography Angiography images: (1) extracting biomarkers such as mCNV area and vessel density using image processing; (2) generating a 3D visualization of the neovascularization for a better view of the affected regions; and (3) applying an ensemble of three white box machine learning algorithms (decision tree, support vector machines and DL-Learner) for nAMD diagnosis. The learned expressions reached 100% accuracy for the training data and 68% accuracy in testing. The main advantage is that all the learned models white-box, which ensures explainability and transparency, allowing clinicians to better understand the decision-making process.