Related Experiment Video
Updated: Mar 11, 2026
Evaluation of Capillary and Other Vessel Contribution to Macular Perfusion Density Measured with Optical Coherence Tomography Angiography
Published on: February 18, 2022
Novel Quantitative OCTA Biomarkers of Choroidal Neovascularization and Associations With Disease Activity and
Yuhong Gan1, Jiaxin Pu1, Shishi Yang1
1State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangzhou, People's Republic of China.
Purpose:
The purpose of this study was to develop a semi-automated optical coherence tomography angiography (OCTA) framework for quantitative morphological assessment of choroidal neovascularization (CNV) to examine associations between quantitative metrics and CNV etiology, activity, and lesion patterns.
Methods:
This retrospective study analyzed treatment-naïve eyes with neovascular age-related macular degeneration (nAMD), myopic CNV (mCNV), or inflammatory CNV (iCNV). OCTA images were processed with a standardized pipeline combining AngioTool, ImageJ, and custom Python algorithms to extract vascular-network and contour-heterogeneity metrics.
Results:
Sixty‑six eyes were analyzed (21 with nAMD, 23 with mCNV, and 22 with iCNV). The nAMD lesions were largest and most branched, showing the greatest median vessel area (1.24 mm2), perimeter (8.369 mm), MaxFeret (2.603 mm), MinFeret (1.598 mm), and total vessel length (14.583 mm)-all significantly greater than mCNV (all P < 0.017). The mCNV lesions were compact and regular. The iCNV lesions exhibited pronounced contour irregularity with lowest circularity (0.621) and vertex index (0.843), and highest maximum (0.243 mm) and mean (0.137 mm) depression depths. For activity assessment, a five-parameter model (vessel area, vessel density, total junctions, average depression, and circularity) discriminated active versus inactive CNV with excellent accuracy (area under the curve [AUC] = 0.901).
Conclusions:
Quantitative OCTA biomarkers distinguish CNV etiologies and activity, providing a noninvasive, objective basis for clinical assessment and individualized treatment planning.
Translational Relevance:
This study bridges the gap between image processing and clinical practice by validating a quantitative OCTA pipeline sensitive to etiology-specific morphological variations. The resulting multivariate model provides an objective, noninvasive tool for assessing CNV activity. Ultimately, these OCTA biomarkers offer a reproducible method to optimize disease monitoring and guide personalized management of CNV.
