An Optimized Hybrid approach to Detect Cataract

Cataract, a disease in which the lens of the eye slowly turns opaque resulting in blurry vision, is one of the most occurring diseases among older people. According to the sources, more than 50% of people aged 80 or more have had eye cataract. While cataracts only used to affect elderly people, nowadays it is prevalent to see them in ages below 50 due to modern lifestyles, lack of exercise, continuous usage of mobile phones and computers. Over 12 million people are blind in India and 62.6% of blindness is a result of cataract. If the detection is done before the final stages of cataract, this disease can be cured in less time. Artificial Intelligence (AI) is a significant tool for the development of computer-aided technology. The retinal image is an important medical reference that helps in diagnosing cataract. This paper proposes a hybrid method using the Fuzzy Technique, Retinal Image Quality Assessment (RIQA), and Convolutional Neural Networks (CNN). The fuzzy technique is used for image quality improvement, RIQA is used for image quality assessment and CNN is used for processing of the retina to automatic cataract detection.