Inspiration
PhizzIO was born out of a personal experience that struck a chord with our teammate, Joe. After tearing his ACL, Joe embarked on a challenging journey of physiotherapy, which presented various accessibility and adherence issues. Moreover, he often found himself lacking confidence in his recovery, battling the temptation to give up. As friends and roommates, it was very painful to see him struggle with his treatment and suffer through various physical and mental challenges it posed.
Additionally, something that made this period of constant pain and discomfort so much more difficult for him was that he only received a piece of paper to guide him through his at-home exercises. This approach to the at-home treatment didn't give him the confidence or support to push through the pain and achieve the recovery he needed.
Joe's physiotherapy journey helped us recognize a need for a supportive and interactive at-home physiotherapy assistant who guides, monitors, and motivates patients throughout their physiotherapy journey. We were able to gather further validation for our claims after talking to other patients and physiotherapists along with various peer-reviewed research articles, allowing us to understand the need for an interactive physiotherapy assistant. Additionally, we also aim to engage with physiotherapists to ensure the presence of a medical expert throughout the patient's journey.
Our vision for PhizzIO extends beyond just an exercise support application; it's a holistic approach to personalized physiotherapy, fostering confidence, accessibility, and motivation for users throughout their recovery journey.
What it does
The various features of PhizzIO fall into two categories: the patient view and the physiotherapist view. The presence of both parties allows for PhizzIO to be a is an interactive platform that enables seamless interaction between patients and physiotherapists.
PhizzIO enables the patient to perform their exercises with confidence and precision through our advanced form guide that tracks the patients' movements and provides them with real-time feedback according to the treatment plan and goals set by their physiotherapist. Our patient analytics dashboard will allow patients to view their current treatment goals, reports from previous exercise sessions, and upcoming appointments and interact with their physiotherapist through an in-built chat feature. These features will ensure the increase in accessibility and treatment adherence for patients by eliminating hindrances such as wait time, physiotherapy center location, and treatment costs.
Through PhizzIO, the physiotherapist would be able to track the patient's progress and guide them through their treatment journey through a more accurate and personalized approach. We also aim to provide various center analytics features to help the therapist better manage their patients' treatment plans and optimize the time and cost efficiency of the physiotherapy center. Physiotherapists will be able to view the patients' adherence rates and consistency of at-home sessions as well as update their at-home treatment plans with highly specific metrics and progress goals. These features will provide the physiotherapy centers with a platform to track the entire patient recovery journey by now being able to access their at-home exercise data. Through the newfound data, the therapy centers will be able to provide personalized plans for patients and expedite their recovery; this will, in turn, increase the reputation and the treatment quality of the physiotherapy center.
How we built it
PhizzIO has several facets, and its integral part is the form guide. We build the exercise form guide in Python using the media pipe pose estimation library. Leveraging real pose estimation, we determined biomechanical angles for each exercise that is performed by physiotherapy patients in general and built the guide. The guide also has an interactive UI with circles that fill up and turn green with the right posture and red with the wrong posture. We used the Tembo stack as our database and used the Tembo standard stack to store patient, physiotherapist, exercise, and log data from the application. The application was built with flask as the backend connecting the database, the form guide, and the UI. For the UI we used HTML, Javascript, and CSS to stylize and make the frontend easy to use.
Challenges we ran into
During the course of our project, we encountered several notable challenges, each requiring careful consideration and strategic navigation to overcome. One such challenge arose from the necessity to reevaluate our project objectives in light of time constraints and practical feasibility. This process led to the difficult decision to abandon a feature that had been under development for a significant duration, despite the considerable investment of effort and resources. Our team grappled with the task of realigning our goals with the project's requirements, ensuring that our endeavors remained focused and conducive to progress.
Furthermore, the integration of new libraries and technologies (Tailwind REACT) posed a unique set of challenges. As we sought to incorporate these innovations into our existing framework, we encountered complexities in compatibility, functionality, and implementation. The process of adapting to unfamiliar technologies demanded thorough research, experimentation, and collaboration among team members. Despite these hurdles, we remained steadfast in our commitment to delivering a robust and refined solution, leveraging our collective expertise and problem-solving skills to address each challenge head-on.
Accomplishments that we're proud of
We are proud of completing this project and getting a working application in 24 hours. This application is a passion project for us as Joe, one of the team members tore his ACL and was finding it extremely difficult to adhere to this treatment. We built PhizzIO so that no one has to go through what Joe had to go through during his painful recovery process. We were able to learn a lot about software development, web development, pose estimation, and tembo! and we will apply what we learned into our careers. We are also proud to get the form guide working as it has several use cases and will definitely help several patients and individuals.
What we learned
During this hackathon, we learned a very important lesson, which was to never give up and pursue what we truly believe in. There were several moments when the Flask library was not working and the pose estimation gave unreliable results, but we stuck true our vision and had an iterative debugging process that helped us overcome all of those hurdles. We also learned a lot about new technology like Tembo which is built on top of Postgres and it was cool to use it in our application.
What's next for PhizzIO
Moving forward, we would like to integrate an AI chatbot into the patient view that is able to assist the patient with inquiries about their treatment plans, and exercise routines and motivate them through audio outputs as they do their exercises. Additionally, we would also like to create a large data set of physiotherapy exercises with detailed descriptions of their uses and techniques.
Log in or sign up for Devpost to join the conversation.