Heart disease is the number one leading cause of death in the US and around the world. This is largely due to the fact that symptom progression is very difficult to track and monitor.

What it does offers the perfect solution for patients living with heart disease. The app, which connects to the Apple Watch, seamlessly monitors patients’ symptoms/ vitals in real time and visualizes data in a user friendly and easy to use format. Every day, the app creates personalized health reports, which users can send to their healthcare providers at any time. Lastly, the app utilizes a push notification system that alerts users if any abnormalities are detected based on a machine learning algorithm. If the app deems that the life of the user is in danger, it automatically alerts emergency responders for immediate medical assistance.

How I built it

The app was designed on Sketch and coded using Swift. The HealthKit API is used in order to retrieve data regarding Steps, Heart Rate, etc., and the WatchKit API allows an Apple Watch to communicate with the app. Firebase is used as the backend service to manage profiles and store user data. The app utilizes the NEXMO API to send status updates to users and text message notifications to emergency responders. The supervised machine learning algorithm that screens for any abnormalities is based on the M4CVD model by Boursalie et. al..

What's next for

I am currently in the process of developing a wearable device that acts as a continuous, clinical-grade ECG monitor for patients afflicted with heart disease. Although information about heart rate, respiration rate, etc. is valuable, ECG data provides the most accurate/ reliable information about a patient's heart beat. Currently, a patient's ECG chart is approximated in the app based on the heart rate information measured by the Apple Watch/ iPhone. However, there currently exists no FDA-approved device that can monitor ECG data in real time. This is why is poised to enter a very valuable space. Stay tuned for more updates in the future!

Pitch Deck

Share this project: