Inspiration - we saw this need where people were skipping physical therapy because it was inaccessible.
What it does - use the accelerometer on the phone to detect the motion of the different leg segment and use the differences in this signal to predict the cause and recommend relevant excises.
How we built it - we build the proof of concept UI in adobe and android studio. The data processing/ analysis was done (for now in matlab) but eventually will be integrated.
Challenges we ran into - human locomotion is a complex problem with multiple facets of interesting challenges. One challenge in particular we faced was figuring out the contribution of changes in acceleration that is correlated with walking speed to just have an abnormal gait. To solve this problem, we normalize the results using velocity, but this fix maybe be feasible on the application side, and further thought must be put it to solve this problem.
Accomplishments that we're proud of - the data we collected for healthy gait using or phone was well correlated to the data available in scientific literature.
What we learned - we about technical changes that need to overcome before physia can become a real application. We have some ideas to fix it, but due to the shortage of time, ad-hoc fixes were implemented.
What's next for physia - we have thought a lot about this aspect of the project: 1) Use machine learning to reliable gait detection and excision suggestion. 2) using physia to remind patients to do the excises on a daily bases. 3) track progress on an improvement metric to motivate the user to get better.
Log in or sign up for Devpost to join the conversation.