What it does
Our project uses optical character recognition (OCR) software to process photographs of handwritten forms to streamline the data entry process. Once data is securely stored in the database, it allows for the ability to run models and analysis. This eliminates the barrier for sharing data between doctor and patient, saving time and money for the clinic and potentially decreasing the hospitalization rate by catching unhealthy trends in data.
How we built it
We developed full-stack web application using the Django framework and a Bootstrap front-end. This way, the application can be automatically scaled and utilized for mobile devices as well. The OCR software was implemented using Google Cloud's Vision API and processes handwritten numbers with upwards of 95% confidence.
Challenges we ran into
The main challenge we faced when implementing this project was trying to connect the OCR software to the website. There are many packages available to use, but not all were updated or accurate and powerful enough to our standard. However, we discovered Google Cloud's Vision library, which worked great for us, and we slowly experimented with it and got it to work with our sample form.
What we learned
Neither one of us had any experience embedding OCR technologies or data visualization libraries into a web application before. Especially since we had to transfer all the processes in Python to display it visually within our HTML templates, it was a great learning experience to be able apply these new technologies when building a web app.
Log in or sign up for Devpost to join the conversation.