PhysioQuest

Gamifying physiotherapy with computer vision to encourage consistency and accuracy when completing quests (of exercises)

Comment

Inspiration

This project came about because a member of our team had recently had a personal experience on how difficult it is to actually stay on top and keep track of exercises that physiotherapists assign. About a week ago, I went to the physiotherapist to check up on my jaw (it was misaligned), and the physiotherapist gave me a list of exercises for me to do every evening until our next appointment. I told myself I would do the exercises, but when the next day came, I forgot, and the day after that, I was too lazy to do it. PhysioQuest is meant to be an answer to this problem of being unable to complete daily exercises, a problem that I'm sure is experienced by more than just me.

After the initial inspiration, we did research to see how common this issue is. According to a study, approximately 75% of patients do not adhere to their prescribed home exercises. Additionally, completing the exercises are proven to be effective in preventing/delaying mobility issues. Studies have shown that the reason why patients (especially teens) do not do their exercises is because of the complexity/confusion, boringness, and time it takes. However, PhysioQuest can solve this!

What it does

PhysioQuest turns the mundane physiotherapist-assigned exercises into a game, encouraging you to complete your exercises every day by creating streaks and providing points for accuracy, ratings. It also contains other features such as a ranking system, badges, an admin interface for physiotherapists to assign exercises to their clients, and much, much more.

How we built it

PhysioQuest utilizes the Mediapipe Pose Landmarks library to detect and track human poses from a live camera feed. It then measures the angle of each joint to ensure the user is correctly completing the given quest and rewards them based on accuracy.

Challenges we ran into

One challenge we encountered is integration. Although we were able to get each aspect of our project working individually, we struggled with integrating them all together. Another challenge we ran into was time management. Given a 36 hour time frame, we struggled with keeping on task to complete the project on time and left almost all of the web development part to the very end. Another challenge was that some of our teammates needed to sleep.

Accomplishments that we're proud of

We are proud of our use of AI considering the fact that none of us had any knowledge on how to create an AI model going into this project. We are also proud of our effective communication and teamwork, utilizing each other's strengths and recognizing weaknesses to effectively create this project.

What we learned

Going into this project, we had little to no knowledge on the use of OpenCV. Throughout the course of this project, we were able to learn how to use facial detection and body movements to complete complex tasks.

What's next for PhysioQuest

The first thing to do would be to fully integrate OpenCV with our program to get full functionality. Then, we would implement some of the following:

  • notifications to remind patients to do their quests
  • real-life incentives (where medical centers partner with local businesses and restaurants to give away coupons or free products)
  • a better admin panel for physiotherapists that is more intuitive. Currently, we just use the template Django admin panel.
  • cooler calendar/history tracking
  • fleshing out the socialization feature, allowing users with similar issues to connect and work together to get better (using a Tinder-style matching system)
  • make the program into a mobile application, as it would probably be more applicable/practical there. Currently, our product is a PWA.
  • incorporate more game aspects like sprites and other graphics to make it a better user experience and more themed
Share this project:

Updates