Inspiration
We primarily wanted to make use of the Google Vision API because extracting information from an image seemed really cool. The choice of a medical app was made because one of our teammates has a friend who had a bad reaction to a medication after different doctors prescribed conflicting medications. Some future evolution of our MedScanner could help doctors and patients to keep tabs on their data and make sure incidents like that do not occur.
What it does
It reads text from an image file and matches words against a large library of known medications, as well as extracting some other useful information such as the dosage and date of prescription. This data is stored and can be added to later.
How we built it
We coded in python on PyCharm.
Challenges we ran into
Getting the Google Vision API to work! Originally we tried to do our project on Java but after many hours of trying we just could not get it to work. Python proved to be much, much easier, and it was nice to get more experience with python since most of us have not used it extensively. Finally we had difficulty working together because every member of our team was working from very different time zones.
Accomplishments that we're proud of
We are proud that our application runs properly and works on many different labels.
What we learned
We learned how to use Google APIs and a lot about how to get different libraries on PyCharm and Intellij. We also learned more about how to use python, particularly dictionaries, the regex library, and json.
What's next for MedScanner
MedScanner needs a UI, we planned for it to be a mobile app to make it easy to scan labels. Eventually, it should sync your medical record with a centralized record that you can access from any device.
Log in or sign up for Devpost to join the conversation.