Precision Segmentation and Feature Extraction of Blood Vessels of Retina using Automated Algorithm

To mitigate the shortcomings of manual vessel segmentation of bloods in retinal images, an automatic algorithm is proposed to enhance diagnosis accuracy and reduce ophthalmologists' workload. The developed procedure for segmenting blood vessels of retinal fundus images has two sections: Image Partitioning and Attribute Extraction. Comprehensive experiments using different (DRIVE, HRF) datasets assessed the algorithm's performance in detecting retinal vessels of blood. Five quantitative metric performances parameters—accuracy, sensitivity, specificity, positive predictive value and negative predictive value—were validated through calculations of its efficiency against current state-of-the-art methods on the dataset used (DRIVE).The introduced technique for extracting blood vessels of retinal fundus images has some mentionable improvement by scoring an Accuracy, Sensitivity, Specificity, PPV, and NPV.