What it does
Identifies whether a cell if infected by malaria
How I built it
- Using convolutional neural network, we trained over 5000 images and constructed a model with 96.5% of accuracy.
- The model is then connected to a web application, such that user can use it online.
- The web application is built with Softheon API and Google Cloud Platform.
Challenges I ran into
- Building a model, and improve its accuracy
- Connect python with front end
Accomplishments that I'm proud of
- Great accuracy improvement: our original model only has 50% of accuracy.
- Building a website with nice APIs
What's next for Hack Your Cell
- Higher accuracy
Log in or sign up for Devpost to join the conversation.