Background: Quantification of neovascularization changes in fibrovascular membrane (FVM) acquired from optical coherence tomography angiography (OCTA) imaging is extremely important for diagnosis and treatment monitoring of proliferative diabetic retinopathy (PDR). However, few vessel extraction methods have been reported for quantifying the neovascular changes on fibrovascular membrane (FVM) in proliferative diabetic retinopathy PDR based on OCTA imaging.
Methods: We propose an optimized approach to segment blood vessels based on an improved vascular connectivity analysis (VCA) algorithm with a combined effort of morphological characterization as well as noise and artifacts elimination. The length and width of vessels are also obtained in quantitative assessment of the microvascular network. Furthermore, we also study the feasibility of the proposed method by monitoring changes of neovascularization on FVM, which is based on the OCTA images of PDR patients after treatment by intravitreal injection of conbercept (IVC).
Results: The proposed method has achieved better performance compared with existing algorithms on accuracy and can be used for PRD treatment monitoring. In the study of PDR treatment monitoring, the data show that from the beginning (0 day) to 5th day of treatment, the total length of neovascularization on FVM in this area has been significantly shortened by an 77.8% reduction, indicating significant effects from the treatment applied. Besides, the average width of the neovascularization on FVM at the 7th day after treatment has been increased by 158%, which indicates that most of the narrow neovascularization has been reduced.
Conclusion: The result and analysis have confirmed that the proposed optimization process with improved VCA method is both effective and feasible to segment and quantify the neovascularization on FVM with less noise and artifacts, thus can be readily applied to monitor the fibrovascular regression within the treatment period.
Clinical Trial Registration: This trial is registered with the Chinese Clinical Trial Registry (Registered 27 December 2017, http://www.chictr.org.cn, registration number 17014160).