Retinal Vessel Classification and Segmentation Using Advanced Machine Learning Techniques

The automated segmentation of blood vessels in retinal fundus images plays a critical role in the field of medical imaging analysis. To create a decision support system to help in the initial diagnosis and treatment of retinal sickness, precise vascular segmentation is regarded to be crucial. Using fuzzy edge detection and learning approaches, this research creates a system that can segment and classify retinal blood vessels. To begin, the proposed RBVSC-FEDLM model employs a contrast enhancement method. In addition to employing GOA to fine-tune the membership functions, a Fuzzy Cluster Network Algorithm for Edge Detection (FCNAED) technique is employed to locate the edges of retinal fundus images. The ORB (Oriented FAST and Rotated BRIEF) feature extractor is used for the extraction of feature vectors. Finally, the use of Auto Encoder (AE) for categorizing retinal images is a novel contribution to the field. This study uses a benchmark dataset to verify the RBVSC-FEDLM model's accuracy, and found that it outperforms state-of-the-art alternatives.