Cardiovascular disease kills 18 million people per year worldwide. It's for many countries either already or soon the leading cause of death. Prevention is better than curing the disease, so we wanted to apply our technical skills to make a lasting positive impact on the world.
What it does
Wavy Assistant monitors your heart rate continuously in real-time and can intervene when necessary. Thanks to its speech interface it's usable for the most vulnerable target audience: elderly. It also allows you to track the symptoms you can measure with a watch by using the speech interface.
How we built it
We connected a $25 off-the-shelf smart band to our cross platform app. The heart rate data is sent to an AI model in python, where it's analyzed for disturbing deviations. We connected a smart home speaker as a speech interface. The speaker sends the spoken text to the AI to analyse how the patient feels over time.
Challenges we ran into
Connecting the separate pieces of the puzzle, getting the heart rate analysis AI model working with our app, getting the spoken text analysis working with our app.
Accomplishments that we're proud of
We made it to the legendary Y Combinator Hackathon, the last one on our bucket list. We came to the US with only a smartband that we knew was hackable and our skills. And in 24 hours we built features that no other health smartwatch company offers in one.
What we learned
Time boxing is the key, Deep Learning is not always the answer, Y Combinator gets cold at night.
What's next for Wavy Assistant
We're going test with beta users by allowing a subscription form in 3 weeks.