Description
- Abstract:
- Analyzing large clinical note collections using natural language processing (NLP) requires computational resources in excess of that typically available internal to academic health centers. External high performance computing (HPC) center have sufficient resources; however, strict HIPAA compliance requirements limit their use for processing Protected Health Information (PHI). We have developed a protocol for using HPC resources while minimizing security risks to PHI by using encrypted tunnels, ensuring that PHI data is never written to shared disk space, and by isolating PHI processing from other users. Formal independent evaluation confirmed the protocol poses minimal risk to PHI and is compliant with HIPAA. This protocol and approach can significantly expand the ability of medical centers to leverage computationally intensive processing securely and cost-effectively.
- Notes:
- Source: Electronic Health Record
- Method: Natural Language Processing
- This research was supported by Agency for Healthcare Research & Quality grant (#R01HS022085) (GM) and National Institutes of Health (#R01LM011364 (GM), #R01GM102282 (SP) and #8UL1TR000114 (Blazer))
- Podium abstract from the 2015 AMIA Annual Symposium
Access Conditions
- Use and Reproduction
-
- This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Citation
Bill, Robert, Melton, Genevieve B., Gottschalk, Bart, et al.,
"Enabling use of high performance computing for secure processing of protected health information"
(2015).
SFHERE Publications and Presentations.
Brown Digital Repository. Brown University Library.
https://repository.library.brown.edu/studio/item/bdr:697491/
Relations
Collection:
-
SFHERE Publications and Presentations
This collection houses journal articles, conference papers and presentations, and abstracts and posters authored by members of the SFHERE team....