New NLP Datasets Added!

Natural Language Processing (NLP) tool SKAN now includes de-identified diagnostic imaging narratives.

December 4, 2020

New datasets added! We have added de-identified diagnostic imaging narratives to the list of free text records that Wake Forest researchers may now search in order to identify cohorts for research prior to obtaining IRB approval. This is in addition to the existing surgical pathology notes already available.

Did you know that greater than 80% of all clinical data is captured in an unstructured form?

Explore this wealth of data now!

Login to

To use this NLP product, visit within the network (or VPN) and login using your WFBMC credentials. Once you identified reports of interest, you should obtain an IRB approval to access full notes or any other related patient data.

Please email for any questions, requests, or problems and learn more at Data Extraction service page.