Inspiration
It is a tough task for players and coaches to determine what is their current health status and how prone they are to an injury while practicing or playing at a professional match.
What it does
Our App provides the users to determine the probability of them getting injured based upon their daily activities data tracked using Fitbit as well as their historical performance. It also provides the user with the probable factors that are impacting the current performance of the individual.
How we built it
We trained a machine learning model using Scikit-learn's multi layer perceptron classifier and bagging classifier wrapped in a bagging ensemble model. The model was trained and deployed on a Flask web service. The android app gets permission from the Fitbit device of the user and sends the relevant data to the web service everyday, which in turn helps to predict the current probability of injury of the user.
Challenges we ran into
Integration of the entire project.
Accomplishments that we're proud of
This is an innovative project which could really help the users to analyze their performance and modify their playing styles to avoid injury.
What we learned
We learned how to implement our skills and integrate modules together.
What's next for Sports Health Tracker
We look forward to publishing our app after getting more data from various studies.
Log in or sign up for Devpost to join the conversation.