RetinoPathyScanner

CMPE272-Team1-Project:

Introduction to the problem statement : Millions of people suffer from diabetic retinopathy, the leading cause of blindness among working aged adults. If diabetic retinopathy can be detected in early stages, then this would be very helpful in preventing it in the first place.

Currently only trained doctors can provide reliable diagnosis. In this we hope to provide a web based service where anybody can upload their retina scan and get a first diagnosis for this.

Abstract : We want to build a web/app service to help people get early diagnosis for diabetic retinopathy. Catching the growth of disease early on is crucial to cure it. We hope to leverage the current research in computer vision and deep learning to provide a first level diagnosis for diabetic retinopathy..

Anyone can signup and utilize the service and upload their retina scans to get a first diagnosis for their chance to develop diabetic retinopathy. This will help a far greater number of people be able to get the diagnosis. Incase of a positive diagnosis, further tests by a professional doctor can be done.

Approach : We will be using deep learning technology to create a model. This model will help in detecting the diabetic retinopathy in the early stages in millions of people. The diabetic retinopathy data will be displayed using ReactJS frontend. For the backend we will be using flask technology to implement our project.

Functionalities : Users will sign up in the portal. Users will log into the portal. User will upload his/her retina scan. User will get the probability of him getting the diabetic retinopathy information and useful links according to result.

Persona :

User: User who wants to get to know the probability of him getting the diabetic retinopathy.

Doctor: Who wants to upload image to check the probability of his patients getting the diabetic retinopathy.

Dataset links : dataset sources: Aptos 2019 Blindness Detection ： https://www.kaggle.com/c/aptos2019-blindness-detection/data

How to run :

1) Git clone the repo 2) In front end folder, run

npm install 3) > npm start 4) Scan QR code to run in app or select run in web browser.