MediScan

Inspiration

We wanted to create a project that would assist in taking prescriptions properly and habitually because with a lot of medications comes confusion. It can be easy to forget when to take certain medications and how much of it to take when you have to keep track of multiple. To fix this, we wanted to come up with a data-driven solution that would allow users to have more ease in their life when it comes to taking medication and being reminded.

What it does

MediScan is a web application that allows users to upload an image of their medical prescription. Our program would then be able to store the information, such as name and dosage, to allow users to automatically keep track of all their medication digitally in one place, with just one snap of a camera each time. Our program would also allow users to set up reminders sent to their phone based on the necessary frequency of the dosage.

How we built it

Using Google Vision and Google Cloud we are easily able to send pictures uploaded to our website to a computer vision program to read the prescription and targeted the third row to find the dosage requirements with Google Natural Language Processing. We then cross referenced that with a database we pulled of all prescription drugs and then run a search of dosage data to see when the user needs to take the medication. Finally using Twilio we would be able to allow the user an option to receive text messages as reminders to take their medicine. The web application was built using Django in Python and React.js working together to make a full-stack application and making API calls.

Challenges we ran into

We originally tried running our own computer vision program to read the images uploaded by patients but ran into issues with processing power. But we then found Google Vision which allows us to have an easy automated reading process of images. Another challenge we ran into was utilizing the Twilio API without the need of a personal phone number to send text messages from.

Accomplishments that we're proud of

We’re very proud to have pushed a minimal viable product in such a short period of time that could be used by people today.

What we learned

We learned a great deal about using data to draw up solutions to real-world healthcare problems and using services such as Google Vision and Google Cloud.

What's next for MediScan

Next we hope to move to properly integrate automated text message reminders as well as an app format in order to make it easier for users to upload their images by simply taking a picture in the app.

Built With

• django
• google-cloud
• google-vison
• python
• react