Epic Tesseract

We used Natural Language Processing to transform unstructured medical records into a well defined structure

Comment

Inspiration: Epic Systems challenged us to use Natural Language Processing to transform unstructured medical records into a set of data that can be injected into a template form.

What it does: GCP ingests History of Present Illness documents and outputs domain-specific properly labeled systemic data.

How we built it: caffeine, sweat, blood, and tears

Challenges we ran into: weakness of the body. Also not having enough data to train our dataset properly.

Accomplishments that I'm proud of: persisting through the hard times

What I learned: we learned more about the diverse feature set that GCP has to offer. Also learned about REST calls, web dev, and use cases of natural language processing/natural language understanding.

What's next for Epic Tesseract: world domination, usurping Epic, voice to text implementation

Built With

Share this project:

Updates