Inspiration
Physiotherapists usually complain that kids are not very receptive to touch during exercises. I'm creating this app to tackle that exact problem. Idea is to create an engaging, gamified experience for kids, so they feel compelled to do the exercises. It could be an actual game (like xbox kinect) or, a cartoon character they like. I went with the latter option because it was easier to implement at this stage. (P.S. My sister is a physiotherapist. I got the inspiration from her.)
What it does
Move Buddy turns physiotherapy exercises into an interactive experience for children. A virtual character motivates the child while an AI chatbot gives them someone to talk to and interact with. Using the browser's camera and MediaPipe, the app tracks the child's movements and automatically counts exercise repetitions. The app generates a report of the completed exercises, which can be shared with the physiotherapist. Based on the child's progress, the physiotherapist can then adjust or assign new exercises.
How we built it
I used replit to create the base of the app. Then constructed individual backend components such as the ai chatbot (google cloud), the humanoid voice agent, the 3d model and the animation of exercises. Then used replit again to combine all the individual components to the final app.
Challenges we ran into
Creating a gamified and engaging experience for kids isn't exactly straightforward. We need to consider why a child would be compelled to do the exercises. An idea of the current version of this app is that a superhero (spiderman) motivates the child, says that if you perform these exercises, you'll be as strong as me (not really but you get the point). Or have in app friend (the chatbot) that a child could talk to normally. and within the normal talks, the ai would encourage the child to do exercises. Or, create a literal game, however the app gets more complex the more exercises we consider. So, spiderman for now.
Accomplishments that we're proud of
Creating the app wasn't the hard part. The hard part was the thought process. The question of why a child would use this. The app creates a base for the "why", I will build upon this "why" later. And, with a week of thinking and developing, the product is really good for an initial test. So, I'm proud of the fact that I could think and create this within one week. With more time and resources, this could be a really viable product, and something not just kids, but also adults could use.
What we learned
Development is easy. Thinking from an end user perspective is the actual challenge.
What's next for Move Buddy
A voice chat, not a chatbot. Kids will feel more engaged with an actual voice talking. More real exercises, more gamified experience, more 3d models kids could choose from. Building animations using motion capture for exercises. Developing and connecting the doctor portal with the app.
Log in or sign up for Devpost to join the conversation.