From September 8–11, Datavant hosted its first annual Future of Healthcare Hackathon. Over 200 attendees spent the weekend developing innovative solutions to improve the future of healthcare. We are excited to announce the various winners of the Hackathon, starting with our Grand Prize Winner: PostOp | Physical Therapy AI. Read on to learn more about this fantastic project, which uses artificial intelligence to help physical therapy patients continue their recoveries at home.
Grand Prize Project: PostOp | Physical Therapy AI
PostOp Team Members
- Ian Matlak — Software Engineer, Fiserv
- Julian Hecker — Software Engineer, LexisNexis
- Geraldine Turcios — iOS Software Engineer, Trello at Atlassian
- Alex Monteverde — Software Engineer, Applied Visions, Inc.
Project Summary: PostOp | Physical Therapy AI
Physical therapy is one of the best ways to heal after an injury, surgery, or accident. However, up to 50% of patients do not adhere to their recovery plan at home. PostOp makes it easier for physical therapy patients to maintain their routine. It also provides feedback while executing recovery exercise programs to help prevent further injury.
Project Inspiration
Two of the developers on this Hackathon team are about to undergo surgery that will require physical therapy during recovery. The team wanted to develop an app that would help them through their own recovery process.
Ian Matlak, a software engineer on the team, noted: “As a national athlete, I am worried about forgetting or not keeping up with my physical therapy routine. Doing PT for 4 years on and off has shown me how hard it is to stay consistent. Before my upcoming procedure, my goal has been to build a solution that makes it easier for me to stay on top of my post-op care.”
Approaching the Problem
The team pointed out that helping users track their physical therapy routine was the easy part of building this app. The hard part was visually analyzing body movements and providing feedback relevant to the exercise being performed. Implementing a full body region-of-interest (ROI) was no small task.
To track a user’s movements, the team used Google’s MediaPipe Pose Estimation AI solution, which provides the coordinates of the user’s joints and limbs within the image. There are many different machine learning models available to track a user’s movements, but the team chose MediaPipe because it can be used on most consumer devices, including mobile smartphones. They also designed a user interface with Figma and developed the app for iOS using SwiftUI.
In order to develop this app, the team undertook a hackathon stretch goal to learn about Pose Estimation and full body region-of-interest (ROI) using AI.
Implications
In a future state, PostOp could be used by physical therapy clinics across the world as an at-home supplement to regular sessions to improve patient care and treatment effectiveness.
Future Steps
The team acknowledges that the position analysis and feedback systems in PostOp | Physical Therapy AI are still very rudimentary. Currently, all analysis and feedback are hard-coded for each exercise. To be a useful product, the team would need to develop a solution that can adapt to any exercise. This would likely require a second layer of machine learning.
Hackathon Grand Prize Announcement from Ritida Nanda on Vimeo.
The PostOp Team
The PostOp Team at the Datavant Future of Health Data Summit 2022 Left to Right: Alex Monteverde, Ian Matlak, Geraldine Turcios, Julian Hecker
The PostOp Team met at Farmingdale State College as members of the Computer Technology Club.
Left to Right: Alex Monteverde, Ian Matlak, Geraldine Turcios, Julian Hecker
The PostOp Team met at Farmingdale State College as members of the Computer Technology Club.
Congratulations to the PostOp Team for developing this project!
Considering joining the Datavant team? Check out our careers page and see us listed on the 2022 Forbes top startup employers in America. We’re currently hiring remotely across teams and would love to speak with any new potential Datavanters who are nice, smart, and get things done and want to build the future tools for securely connecting health data and improving patient outcomes.
.svg)
