Deep Network Analysis and Prediction of Ophthalmic Disorders

ABSTRACT

Everybody, around the globe, is aware that their kids, relatives, and family are suffering from the pandemic COVID-19. S everal people are still facing post-COVID-19 issues. During COVID-19's second wave, mucormycosis, sometimes known as "black fungus, " plagued people, especially those who had previously been infected with the virus. The clinical manifestations of mucormycosis are quite varied, the disease affects the skin, subcutaneous fatty tissue, and visceral organs such as the eyes and brain. This paper surveys the Mucormycosis-affected eye diseases due to post-COVID-19 complications and leverages the Machine learning model to differentiate it from other eye diseases. COVID-19-associated Mucormycosis carries a very high mortality rate and timely detection that can assist people in starting therapy at an early stage of the disease, increasing their chances of recovery. Though it was evaluated for a specific disease (COVID-19-associated mucormycosis) we ended up developing a framework that can detect other eye diseases. Thus, the goal of this research is to distinguish Mucormycosis from other eye diseases such as Bulging Eyes, Cataracts, Crossed Eyes, Glaucoma, and Uveitis. This study implies Deep learning techniques with a Convolutional Neural Network based on the TensorFlow and Keras model to detect and make use of computer vision to accurately classify eye diseases. We achieved a precision of 70% in this study by developing a webpage using the trained model for an eye diseases evaluation. © 2022 IEEE