Come check out our fun Demo near the Google Cloud Booth in the West Atrium!! Could you use a physiotherapy exercise?
The problem
A specific application of physiotherapy is that joint movement may get limited through muscle atrophy, surgery, accident, stroke or other causes. Reportedly, up to 70% of patients give up physiotherapy too early — often because they cannot see the progress. Automated tracking of ROM via a mobile app could help patients reach their physiotherapy goals.
Insurance studies showed that 70% of the people are quitting physiotherapy sessions when the pain disappears and they regain their mobility. The reasons are multiple, and we can mention a few of them: cost of treatment, the feeling that they recovered, no more time to dedicate for recovery and the loss of motivation. The worst part is that half of them are able to see the injury reappear in the course of 2-3 years.
Current pose tracking technology is NOT realtime and automatic, requiring the need for physiotherapists on hand and expensive tracking devices. Although these work well, there is a HUGE room for improvement to develop a cheap and scalable solution.
Additionally, many seniors are unable to comprehend current solutions and are unable to adapt to current in-home technology, let alone the kinds of tech that require hours of professional setup and guidance, as well as expensive equipment.
Our Solution!
- Our solution only requires a device with a working internet connection!! We aim to revolutionize the physiotherapy industry by allowing for extensive scaling and efficiency of physiotherapy clinics and businesses. We understand that in many areas, the therapist to patient ratio may be too high to be profitable, reducing quality and range of service for everyone, so an app to do this remotely is revolutionary.
We collect real-time 3D position data of the patient's body while doing exercises for the therapist to adjust exercises using a machine learning model directly implemented into the browser, which is first analyzed within the app, and then provided to a physiotherapist who can further analyze the data. It also asks the patient for subjective feedback on a pain scale
This makes physiotherapy exercise feedback more accessible to remote individuals WORLDWIDE from their therapist
Inspiration
- The growing need for accessible physiotherapy among seniors, stroke patients, and individuals in third-world countries without access to therapists but with a stable internet connection
- The room for AI and ML innovation within the physiotherapy market for scaling and growth
How I built it
- Firebase hosting
- Google cloud services
- React front-end
- Tensorflow PoseNet ML model for computer vision
- Several algorithms to analyze 3d pose data.
Challenges I ran into
- Testing in React Native
- Getting accurate angle data
- Setting up an accurate timer
- Setting up the ML model to work with the camera using React
Accomplishments that I'm proud of
- Getting real-time 3D position data
- Supporting multiple exercises
- Collection of objective quantitative as well as qualitative subjective data from the patient for the therapist
- Increasing the usability for senior patients by moving data analysis onto the therapist's side
- finishing this within 48 hours!!!! We did NOT think we could do it, but we came up with a working MVP!!!
What I learned
- How to implement Tensorflow models in React
- Creating reusable components and styling in React
- Creating algorithms to analyze 3D space
What's next for Physio-Space
- Implementing the sharing of the collected 3D position data with the therapist
- Adding a dashboard onto the therapist's side
Built With
- computer-vision
- firebase
- javascript
- machine-learning
- node.js
- posenet
- react
- tensorflow
Log in or sign up for Devpost to join the conversation.