LLM Validates Cancer Patient’s Pan-Cancer Clinical Data Elements

Poster Number: P123
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Natural Language Processing, Machine Learning, Controlled Terminologies, Ontologies, and Vocabularies, Real-World Evidence Generation, Workflow, Clinical Guidelines
Primary Track: Applications
Programmatic Theme: Consumer Health Informatics

There are 100,000 MSKCC cancer patients with genomic data that is an asset for researchers but, can be enhanced if complemented with clinical data. MSKCC employs data curators to manually obtain clinical data which takes 1-1.5 days for each patient. To tackle this, we collaborate with Realyze Intelligence to harness large language models (LLM) to automate clinical concepts. The advantage is speed and efficiency in extracting data for many patients in real time.

Speaker(s):
Andrew Niederhausern, BS
MSKCC

Author(s):
Nadia Bahadur, Masters of Clinical Research - Memorial Sloan Kettering Cancer Center; Andrew Niederhausern - MSKCC; Gary Wallace, Bachelor of Science - Realyze Intelligence; Carlos Martinez, Master of Science - Memorial Sloan Kettering Cancer Center; Gilan Saadawi, Doctor of Philosophy, Doctor of Medicine - Realyze Intelligence; John Philip, MS - Memorial Sloan Kettering Cancer Center;