Intelligent Chart Summarization

feature

(Brady Miller) #41

looking forward to yaml :slight_smile:


(Matthew Vita) #42

Shane and I just used some of the JSON from the parser for the friendly web ui. That project is hard but coming along nicely. We noted a couple very minor improvements for the parser, but all in all it is testing well!


(Matthew Vita) #43

cTAKES: How to Create an ICD10 Dictionary: https://www.youtube.com/watch?v=4aOnafv-NQs

This is the process in which we can create ICD10/CPT code mappings. SNOMED/RXNORM are included out of the box, but there is no good documentation on doing anything else (until this video, that is!)


(Matthew Vita) #44

Very early screen of web ui:

There are a few bugs, but overall it’s coming along. Missing features include:
Multi colored backgrounds (if multiple concept mention types are present, “stack” the background colors)

  • Overall codes view
  • Deletion of codes
  • Test out with ICD10 and CPT
  • Documentation on the API (it’s pretty easy… just fetch the raw text and data from an endpoint and pass it into the JS code’s default constructor)

(Brady Miller) #45

Very nice and makes sense clinically!


(Matthew Vita) #46

w00t! I’m going to try to make it more user friendly as I go…


(Matthew Vita) #47

latest-q.pdf (92.0 KB)


(Matthew Vita) #48

Phew! I spent my Sunday revamping the algorithm to support multi color and fixed a big bug


(Matthew Vita) #49

Multi line and escaped character support!


(Matthew Vita) #50

Working on a formal solution to get millions and millions of records into MySQL: https://github.com/GoTeamEpsilon/cTAKES-HSQLDB-to-MySQL-Dictionary

The community has been very helpful as there are multiple people looking to do this as well!

-m


(Kirk Hadley) #51

Is this project still a going concern?


(Matthew Vita) #52

@KirkHadley yes, we are reaching our first milestone. I’ll post back here soon :slight_smile: