HeartInsight

• 

  Home page

• 

  Prompts user to enter information

• 

  User enters information

• 

  Returns a predicted risk and suggested articles

Inspiration

After learning that heart disease is one of the leading causes of death, our team wanted to create an accessible way for people to become aware of their risk of developing heart disease. We hope that presenting someone with their personal risk will inspire them to make possibly life-saving changes to their lifestyles, or at least provide some insight to their heart health.

What it does

Our app prompts the user to input the following information:

• sex
• age
• whether they currently smoke
• how many cigarettes they smoke in a day
• do they take blood pressure medication
• do they suffer from hypertension
• do they have diabetes
• total cholesterol
• systolic blood pressure
• diabolic blood pressure
• BMI
• resting heart rate
• glucose level

It then feeds the user input to our machine learning algorithm and returns a predicted risk of them developing coronary heart disease. The results are accompanied by suggested articles that vary based on the user's input.

How we built it

First, we developed our machine learning model by using Python to perform a logistic regression on a heart disease dataset from Kaggle. This involved filtering, transforming, and exploring the data, and then acquiring coefficients for each predicting variable so that we could make find the probability of the target variable (coronary heart disease).

Then we created the app using Flutter, which involved coding in Dart. Once we finished creating the GUI, we simply implemented a function that standardizes the user input, and uses the coefficients from our model to return a prediction.

Challenges we ran into

Since this was our first time in app development, figuring out how to start building our app was a bit of a struggle. We initially tried using React, but it didn't seem to be compatible with our computers. Once we finally decided to use Flutter, building an app with its software also proved to be a learning curve, but once we figured it out the rest of the development process ran smoothly.

Our team only consisted of two people, which also made this project challenging since we had settled on our idea with the assumption that there would be three of us. This unexpected increase in workload got a little overwhelming at times.

Accomplishments that we're proud of

We proud of the fact that we were able to develop a functioning machine learning model, especially since neither of us had any experience with data analysis or machine learning going into this project. We're also proud that we build our first mobile app!

What we learned

Working on this project gave us a lot of exposure to working with data, and how to use Python to implement a machine learning model. We also learned a lot about the app development process.

What's next for HeartInsight

We hope to continue building and improving our ML model in order to obtain more accurate results. We also hope that we can include more variables that may be relevant and provide deeper insight, such as exercise or diet.

Built With

• colab
• dart
• flutter
• python