HBRP Publication Page 13-24 2023. All Rights Reserved Page 13 Advancement in Image Processing and Pattern Recognition Volume 6 Issue 2 Automated Eye Diseases Recognition Web-Application Using Convolutional Neural Networks Aahn Deshpande 1 , Shubham Kumar 2 , Kalash Butola 3 , Harshit Pandey 4 , Jyoti Gupta 5 * 1,2 3 4 5 Department of Electronics and Communication Engineering, Bharati Vidyapeeth’s College of Engineering, New Delhi - 110063, India *Corresponding Author E-Mail Id: - jyoti.gupta@bharatividyapeeth.edu ABSTRACT The World Health Organization (WHO) estimates that more than 2 billion people worldwide experience close-up or distance vision issues. This study presents a method for creating an API for a proposed Deep learning CNN’s sequential model that can instantly and efficiently determine if a user has eye diseases such as glaucoma, crossed eyes, uveitis, cataracts, and bulging eyes. The API allows for easy integration into various applications, providing a valuable tool for developers and researchers. The system allows users to upload their eye images for diagnosis, with an accuracy of 97%. It is intended to assist ophthalmologists, not replace them. The proposed model aims to address the global issue of vision problems as reported by the World Health Organization, and provide a solution to ease the workload of ophthalmologists while increasing the detection accuracy of eye-related diseases. Keywords:-Convolution neural network, Sequential model, webApi, VGG-19, ReLU, strides, maxpooling, pickle file INTRODUCTION The artificial intelligence revolution in the human-computer interactions field has the potential to bring massive changes in multitudes of fields such as healthcare, travel, and finance. Finding ways to use artificial intelligence in healthcare has been in progress in recent decades, particularly using automated learning techniques to develop automated models to predict diseases is an important research direction. Automated learning techniques have been applied for the detection of diseases such as Hypertension [1], Asthma [2], Stroke [8], kidney diseases [9], Lung Cancer [10], Eye diseases [10]. In this paper, the focus is on the detection of eye diseases using a machine learning model CNN.The WHO (World Health Organization) estimates that 2.2 billion individuals worldwide have a close-up or distant vision issue. Estimates suggest that half of these situations might have been avoided or resolved. In addition to those who have near-sightedness impairment due to uncorrected presbyopia (826 million), Untreated cataract (94 million cases), glaucoma (more than 7 million cases), corneal opacities (more than 4 million cases), diabetic retinopathy (3.9 million cases), and trachoma (3.9 million cases) all cause moderate-to-severe distant vision impairment or blindness in 1 billion people (2 million)[11]. The primary causes of visual impairment are Age-related macular degeneration, uncorrected refractive errors, cataracts, glaucoma, corneal opacity, diabetic retinopathy, trachoma, hypertension, and other conditions. According to the National Library of Medicine and the National Center for Biotechnology, India's rate of uveitis per lakh is three times higher than that of the USA, which is approximately 730 per 100,000[25]. There