This project presents a React-based Web Application which is able to take in X-Ray images of patients lungs, predicting whether or not the lungs are infected with Tuberculosis, with a given confidence. It is hosted under the following domain: https://tbclassify.netlify.app.
DISCLAIMER: This project is not meant to be used for medical purposes. It is for academical purposes only.
The data for this project comes from the kaggle dataset Pulmonary Chest X-Ray Abnormalities. It is divided into testing and training data on download.
This project was developed for the purposes handing in an assessment for completing the Computational Vision course at PUC-Rio, under the supervision of professor Marcelo Gattass.
The model used for predicting outcomes was trained using the training tool Teachable Machine, developed by Google. The training data was fed into the algorithm, which produced a model that could be loaded by the interface, being able to generate predictions.
This project is completely open-source and may be used by anyone for academical or commercial purposes. To run the project, follow these steps:
- Install NodeJS and yarn in your local machine
- cd into the project source folder
- Run
yarnin your command prompt, to install the necessary dependencies - Run
yarn startto run the project in your browser - Enjoy :)