Heart Stroke Manager

• 

  Homepage

• 

  Sign-up Page

• 

  Login Page

• 

  Stroke Information Page

• 

  Stroke Questionnaire

• 

  Stroke Questionnaire Result

• 

  Daily Logs and Progress page

• 

  Statistics Page

• 

  Model Input

• 

  Model Output

Inspiration

According to the World Health Organization cardiovascular disease is the world's leading cause of death. More specifically, strokes are the 2nd leading cause of death globally. We wanted to create a project that helps tackle this issue.

What it does

Our project is essentially a web app that allows you to predict if you will have a stroke in your lifetime depending on your lifestyle. If you are vulnerable, this web app would help you improve your lifestyle to decrease your chances of getting a stroke. In addition, the website has a questionnaire to allow a user to figure out if they or someone they know are having a stroke and a daily log feature to track lifestyle improvements in various categories (water intake, BMI, weight, amount of exercise, glucose level). All of the data entered will be saved to your account.

How we built it

HTML, CSS and JS

• Used to design the front end of the website.

SQL

• To store user data we created a SQL database which allows us to access user data from any file in a organized manner

Machine Learning

• For the stroke prediction, we trained our very own machine learning model using stroke data on Kaggle. We then used a Jupyter Notebook and the Scikit Learn library to create the machine learning model. If interested, view the model training process here.

Flask

• To integrate the machine learning model and the databases onto the website so it can interact with the Front End of the website, we used Flask.

Challenges we ran into

Time

A major issue that we ran into is time, we had only 3 days to brainstorm an idea and come up with a solution. Polished projects usually take weeks, if not months to finish, but we managed to create a prototype of our solution in the time we had.

Integration

We used this hackathon to learn more about front-end and back-end. We had to learn how to integrate the login and sign up page and the machine learning model with the front-end of the website as none of us had this experience before.

Accomplishments that we're proud of

In the span of 3 days, we had successfully came up with a problem and brainstormed a solution for the problem. We were also able to integrate the login page and the machine learning model to the website and make it functional.

What we learned

We learned how to use Databases to store the user data. We learned new machine learning techniques such as removing missing values, encoding non-numerical data to numerical data and hyperparameter tuning to improve the machine learning model. We also learned about integrating both the login page and the machine learning model to the front-end of the website using Flask.

What's next for Heart Stroke Manager

• Hosting the website
• Adding graphs to show the trends of the user's health
• Improving the accuracy of the machine learning model
• Improving the front-end of the website by making the website more responsive

Built With

• css
• flask
• html
• machine-learning
• python
• scikit-learn
• sql