Over 93% Accuracy • Advanced Powerful AI • Easy & Simple to Use
This project is a functional full-stack web app that includes features such as login and signup functionality with a database through Flask-Login, beautiful design thanks to Materialize.css, an easy to understand backend because of Flask, best practices for production servers, and a state-of-the-art X-Ray Classifier powered by fast.ai. The AI was trained on over 100,000 anonymized chest x-ray images from more than 30,000 patients (NIH Dataset).
Website GitHub repo: https://github.com/HHousen/ai-respiratory-doctor
I also created an electron app for this project. It uses a self-contained python flask API, and pytorch/fastai to compute the probabilities of certain diseases in chest X-Rays using machine learning.
Asset bundling is handled by webpack through electron-webpack and building/compiling is done with electron-builder. The user interface was built using the Materialize-css framework. The original template this app was based on is electron-webpack-quick-start.
Additionally, Travis CI handles automatic building of the application on every commit.
Electron GitHub repo: https://github.com/HHousen/ai-respiratory-doctor-electron