Extracting Disease Phenotype Information for Patients with Rheumatic Disease from RISE Registry Clinical Notes
ABOUT THE PROJECT
Project Description
The RISE registry includes complete EHR data for over 300 rheumatology practices across the United States, including over 400 million clinical notes. In this project, we will use natural language processing to extract key information from these clinical notes regarding the patient's phenotype and patient-reported outcomes. Initial concepts include rheumatoid arthritis patient-reported outcomes that are routinely recorded by rheumatologists in clinical notes, and organ manifestations of systemic lupus Erythematosus.
Funding
Probably yes
LOOKING FOR
Required Skills
Natural Language Processing tools (NLTK, CTAKES, PyTorch-NLP, etc)
Required Course Work or Level of Knowledge
- Machine Learning, intermediate - advanced
- Acceptable Level of Education (eg. Undergrad, Grad Students, Post Docs, MD, PhD)
- Graduate students,Postgraduates,Full-time workers who volunteer time
Additional Considerations
We are collaborating with the Stanford NLP team (Suzanne Tamang et al) but would like to also partner with someone at UCSF if available.
CONTACT
PI/Research group
Yazdany/Rheumatology Quality and Informatics Lab
Contact
Jinoos Yazdanym, [email protected]