A prototype Apple Watch app for realtime heart monitoring.
We had a great weekend at the NuPIC (Numenta Platform for Intelligent Computing) Hackathon.
Many cool apps were built, but I centered on how I could use the machine learning on a wearable app.
I figured that we could use the predictions made by the CLA (cortical learning algorithm) to detect patterns and anomalies. When the CLA learns the pattern of the heart rate, the app shows your calculated beats per minute. When the CLA can't predict your heart rate, the anomaly score is high, and we alert you by showing an "Erratic" status. You could then share the rhythm strip history with your doctor.
I tailored the animations and the user experience for the Apple Watch.
I'm excited about how we can use AI (artificial intelligence) like this in healthcare. I think it will enable us all to understand our bodies and stay healthy.
The Watch Simulator is up on GitHub.
If you're inspired, snag your spot at the first ever WatchKit hackathon.