DreamNLP: Novel Data Streaming Natural Language Parser System

Case ID:
C14296
Disclosure Date:
9/30/2016
Description:
Unmet Need
Extracting important information from electronic health record (EHR) database systems is an important part of data analytics in clinical settings. Performing this operation quickly is computationally expensive in a real-time setting.
 
Technology Overview 
This software is a novel natural language processing analytic platform that performs the identification and extraction of different words, sentences, phrases and ontologies to quickly visualize the key features of any medical record. The system uses an unsupervised count sketch method that can determine the relationship between structure, clinical variables, and histological features that may link a given electronic record. This feature can provide important clinical associations to improve diagnosis.
 
Stage of Development
Prototype in use.   
 
Publications
None.
 
Patent Information:
Title App Type Country Serial No. Patent No. File Date Issued Date Expire Date Patent Status
Method and System for a Natural Language Processing Using Data Streaming PRO: Provisional United States 62/441,619 1/3/2017     Expired
Method and System for a Natural Language Processing Using Data Streaming ORD: Ordinary Utility United States 15/845,842 12/18/2017     Pending
Inventors:
Category(s):
For Information, Contact:
Jon Gottlieb
jgottl10@jhu.edu
410-614-0300
Save This Technology:
2017 © Johns Hopkins Technology Ventures. All Rights Reserved. Powered by Inteum