Theraflow | Devpost

Graphs of patient progression
Dashboard for doctors to view patients stats at a glace
Profile page + qr code for patient to connect with therapist
Page for doctors to view / manage all their patients
Easy to use scheduling for doctor to see future tasks of eahc patient
"Film Session" page for doctors to view workouts and provide additional feedback to patients
Profile screen for the user to connect with their doctor if applicable
Onboarding screen for users to share their symptoms and get a workout plan
Automatic recognition of both symptoms and places where user is hurt to identify key exercises
Filming screen where users can receive live feedback on form and send workout film to doctors

Inspiration

Whether it be through friends or our own personal experiences, each one of us has seen firsthand the strenuous process to recover from an injury, whether it be a small fracture or a much larger injury. Moreover, lower income members of soceity often have a higher chance of "relapsing" into injury because of the large, prohibitive cost of physical therapy. Our app, Theraflow, reorients our current understanding of physical therapy and allows patients to not only gain access to high quality care without economic status in mind, but also ensures more accountability in the physical therapy process.

What it does

There's two distinct platforms through which Theraflow functions - our iOS app and our web app. On the iOS app, patients are able to enter their symptoms and conditions and recieve a customized physical therapy regiment to ensure they can be back to 100% as quickly as possible. On top of this, once patients start using this plan, our app allows them to film their sessions and get live feedback from our AI model on their form and how they can improve. These features are crucial for a couple of reasons: 1) It ensures that patients are able to see the steps they can take to recover without having to go through a costly and potentially inaccessible doctors appointment and 2) it allows patients to have the most effective care possible as they get live feedback on how to improve their form - something previously only available through costly doctor appointments. On the web app, if a patient has a doctor, they can connect with them and the doctor can view important statistics on the progression of patients. Specifically, doctors can watch "film" from their patients that recorded workouts on the iOS app in order to provide any necessary additional feedback. On top of this, doctors are also offered a wide variety of statistics (on things like heart rate, reps, time spent, etc) on the website that allow them to see how each of their patients is progressing through physical therapy. Essentially, our app not only allows patients to receive care if they were previously unable to afford it, but it also enhances current care infrastrucutre, decreasing the amount of time it takes to heal from injuries and reducing the chance at repeat injuries.

How we built it

The front end of our website is built with html, css, bootstrap, and javascript. Our iOS app is written in swift/swiftui and makes use of machine learning powered through AWS Medical Comprehend, AWS Elastic Beanstalk, and Apple in order to provide accurate recovery plans for patients to follow, as well as check for form. For our database we used firebase.

Challenges we ran into

On the iOS squat, it was difficult to check for the form on squats and ensure that values being used would correspond to a good form. To counter this, we experimented (and did a lot of squats) with how we could best track form and ensure that the end user could get the best possible live feedback. On the website, it was difficult to integrate the numerous user schedules onto one page for the doctor to track. The large amount of data involved in storing these schedules allowed us to learn more about efficiency in reading data from the database as well as the importance of scalability.

Accomplishments that we're proud of

We're proud of how we our ML model is able to reccomend exercises for patients based on the various injuries that they may have. We're also proud of how we were able provide live feedback on user form so that people using the app can understnad how they can best improve

What we learned

We learned a lot more about AWS and the different services they provide / how to integrate these services with swift. We also learned more about the importance of having databases that are both horizontally and vertically scalable to allow for the optimum amount of integration between the website and app.

What's next for Theraflow

During the hackathon we received a lot of super helpful feedback on our app and the steps that we can take to help expedite processes for doctors even more. We're excited to continue working on this app and adding more functionality so that patients are truly able to recieve the best quality of care, regardless of their economic status. During the hackathon, we were able to reach out to some friends in the healthcare industry and get a better understanding of what it would take to get this product to market, so we're looking forward to using that feedback to work more on Theraflow over the next few weeks and beta launching in the coming month