Inspiration
With the amount of deadlines and dates we have as students, it's easy to slip up and forget an assignment or two. This is no different in the field of medicine, patients are prone to sometimes forget to take their prescribed medication. We sought to solve this issue using by sending reminders to the one device everyone carries at all times, their phones. MediScan targets patients that are elderly or less abled, but it can be used by any patient to receive text reminders of their prescriptions.
What it does
MediScan allows elderly or less abled individuals to keep track of their prescriptions and improve their health by encouraging proper medication use. It does this by using OCR (Optical Character Recognition) tools to catalog information from the inputted Rx prescription image, then sending that information to the user's phone as a text message reminder.
How we built it
We built MediScan on python using the packages smtplib, ssl, and pytesseract. Smtplib and ssl was used for the automatic text sending program and pytesseract was used as a wrapper for the OCR engine Tesseract. Tesseract's pre trained models for text recognition was used in MediScan to read images of Rx prescriptions.
Challenges we ran into
The biggest challenge we ran into was finding the proper packages for our specific use case. There are numerous ones that will do similar things but we needed some that were streamlined for a proof of concept.
Accomplishments that we're proud of
We are proud that the OCR was able to be implemented with little fault in reading into the software and that the email-to-text from smtplib and ssl worked fairly efficiently.
What we learned
We learned of the effectiveness of OCR in reading text from images and how usefully it can be applied in many situations.
What's next for MediScan
To fully feature out MediScan we would need to include host servers, databases for prescriptions, more versatile deployment, and streamlined user interface/experience.
Log in or sign up for Devpost to join the conversation.