PCOS is a condition in which a woman's hormones are out of balance. It is a very common metabolic condition, around 1 in 4 women are said to suffer from this issue. However, going through doctor appointments is a very tedious task and individuals may fall into the loophole of unnecessary medical tests.
What it does
To help with this issue we have built an app that can help one to check if there is a chance of them having PCOS using the elaborate questionnaire provided in our app. Utopia, with its predictor, schedule tracker, daily tips provides a tight schedule on their lifestyle to keep symptoms in control!
How we built it
We made an ML random forest-based classifier to predict the possibility of PCOS and hosted it on Heroku using Flask. We built our app on Android Studio using Java and XML and integrated it with our machine learning model. We made a UI for the app too!
Challenges we ran into
We faced difficulty in hosting the ML model on Heroku using Flask and integrating it with our app. Also, creating the UI and making the app within the time period was difficult. We also found the preprocessing of data a bit challenging but we overcame it.
Accomplishments that we're proud of
Despite the difficulties we faced, we were able to successfully host the ML model on Heroku using Flask and integrate it with our app. We are proud to say that we were able to make 'Utopia' into a fully functioning application in the given time period!
What we learned
We learnt how to host ML models on Heroku with the help of flask. We also learnt how to integrate ML model with apps. We explored a lot of UI themes to come up with our own UI for Utopia. Of course, time management is an important skill we learnt.
What's next for Utopia
We want to introduce a virtual platform for users to connect with doctors. We also want to schedule lab tests for users and to autofill, the data from lab reports with the app itself. And introduce a calendar along with a reminder system to help PCOS patients.
Please note ML model is in the main branch and app Code is in the master branch in GitHub repository.
Log in or sign up for Devpost to join the conversation.