omprehensive Study on Blood Vessel Segmentation and Extraction from Optical Coherence Tomography (OCT) Images: Modern Approaches and Applications

Optical Coherence Tomography (OCT) has become a powerful, non-invasive imaging modality that enables high-resolution visualization of microvascular structures, especially in the retina. Automated segmentation of retinal blood vessels in OCT images plays a critical role in the early detection, monitoring, and treatment of ocular diseases such as diabetic retinopathy, glaucoma, and macular degeneration.This study provides a comprehensive review of various vessel segmentation methods, ranging from traditional image-processing algorithms to advanced deep learning-based approaches. Classical techniques such as thresholding, morphological operations, and edge detection are compared with modern machine learning and convolutional neural network (CNN) architectures, including U-Net, SegNet, and Vision Transformers (ViT).The findings demonstrate that deep learning models—particularly transformer-based architectures—achieve superior accuracy in detecting small, low-contrast, and complex vascular structures. Furthermore, effective preprocessing techniques, data augmentation, and self-supervised learning approaches significantly improve model robustness and generalization. The integration of these advanced methods can accelerate diagnostic efficiency and enhance clinical decision-making in ophthalmology.

کلیدواژه ها:

Optical Coherence Tomography (OCT) ، Blood Vessel Segmentation ، Deep Learning ، Image Analysis ، Vision Transformer ، Medical Imaging