CareHive

Home Page
Model Detection

Inspiration

Diseases are prominent in today's world, especially respiratory illnesses and they cause many deaths. For example, SARS spread in 2003 with a death rate of 20%, and MERS was introduced in 2012, with a fatality rate of 38%. In 2020, there was the COVID global pandemic which caused major disturbances throughout the world and affected numerous people. This illness eventually leads to pneumonia, which is a dangerous infection that when misdiagnosed can be fatal to both young children as well as adults. This inspired us to create CareHive, a web platform that uses ML to predict pneumonic chest x-ray images, thereby helping to reduce medical errors and misdiagnoses.

What it does

CareHive is a web application with ML integration that helps determine normal chest x-rays vs pneumonic chest x-rays. We leveraged an open-source dataset from Kaggle to train our model. Our application can help ensure that there is a proper diagnosis and help reduce medical errors in imaging. The application gives users a portal to upload chest x-rays which are then run against a ML model to detect pneumonia in chest x-rays. In addition to this, they can create an account to store their data as well as contact doctors, after the data results have arrived.

How we built it

Figma: UI Design
React Js: For frontend
Python/Flask: For backend
ML: analyzing image data

Challenges we ran into

Some of our team members were still learning React, so they needed some help along the way
Located in different time zones within the US
Coming up with a feasible idea that adheres to the theme