A Deep Learning Approach for Retinal Image Feature Extraction

Retinal image analysis is crucially important to detect the different kinds of life-threatening cardiovascular and ophthalmic diseases as human retinal microvasculature exhibits remarkable abnormalities responding these disorders. The high dimensionality random accumulation images enlarge data size, that creating complexity in managing understating data. Deep Learning (DL) has been introduced deal with this big challenge by developing intelligent tools. Convolutional Neural Network (CNN), a DL approach, designed extract hierarchical features more abstraction. To assist ophthalmologist eye screening disease diagnosis, CNN being explored create automatic systems for microvascular pattern analysis, feature extraction, quantification images. Extraction true vessel significant further such diameter bifurcation angle quantification. This study proposes feature, segments extraction approach exploiting Faster RCNN. fundamental Image Processing principles have employed pre-processing A combined database assembling from publicly available databases used train, test, evaluate proposed method. method obtained 92.81% sensitivity 63.34 positive predictive value extracting top first tier colour It expected integrate into diagnostic tools evaluation validation analysing performance.