This project is inspired by State Farm's Drive Safe and Save program, where users are rewarded for safe driving behaviors. In the world of digital health, our team came up with the idea of using an individual's health data to determine whether they are making progressing towards adopting a healthier lifestyle and should be rewarded in the form of reduced life insurance premium rates. Cardiovascular diseases are one of the biggest killers in America, and people with theses diseases are often charged a higher premium for life insurance. Therefore, our application focuses on evaluating an individual's lifestyle with regards to his or her cardiovascular health.
What it does
Fitbit owns >45% of the health-monitoring wearable's market. This app uses the Fitbit API to gather activity data from a user to determine their activity level, nutrition info, and sleep habits to determine their overall risk of exacerbating their cardiovascular conditions. Since this requires taking in multiple parameters and identifying how each of these parameters affect an individual's cardiovascular health, the app uses Machine Learning (logistic progression) to quantify how much progress an individual's making towards improving their cardiovascular health.
How we built it
We trained a machine learning model using a cardiovascular health dataset from Kaggle and developed a server using Python and Flask to integrate the machine learning model into the application. On the front end, we used Vue.js and Vuetify to develop a user interface.
Challenges we ran into:
The lack of training data for the machine learning model, which drove us to experiment with different machine learning models to see which one performs the best on the limited dataset.
Accomplishments that we're proud of
We're proud of how we adapted and improvised to multiple obstacles both in the technical aspects and algorithm.
What we learned
We learned a lot about full-stack development and the importance of our daily choices in determining our cardiovascular health as we were researching the data along the way.
What's next for LifeBit
Potentially using data from more sources to improve the accuracy of the machine learning model and incorporate more factors including an individual's social activity and driving habits to determine his or her qualification in life insurance discounts.