LLM and CQL computable phenotypes for myocarditis adverse event feature and outcome determinations using patient FHIR data: a feasibility study
Fraser W. Gaspar, Seth B. Roberts, Francis X. Campion, Lynda Hoeksema, Jessica Kasirsky, Raymond Duncan, Matthew Sonesen, Renier Estiandan, Khyzer B. Aziz, Julie A. Riddler, Jessica Zhou, Hussein Ezzeldin · 2026
Read LLM and CQL computable phenotypes for myocarditis adverse event feature and outcome determinations using patient FHIR data: a feasibility study by Fraser W. Gaspar, Seth B. Roberts, Francis X. Campion, Lynda Hoeksema, Jessica Kasirsky, Raymond Duncan, Matthew Sonesen, Renier Estiandan, Khyzer B. Aziz, Julie A. Riddler, Jessica Zhou, Hussein Ezzeldin (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13593815.txt.
Description
Fraser Gaspar and colleagues tested in 2026 whether a large language model and a rule-based method could find signs of myocarditis in patient data. Safety checks after vaccination need facts that often sit in free-text notes, not in neat coded fields. The team used FHIR patient records from two health systems.
Records were turned into plain text, and a language model answered yes-or-no questions based on Brighton criteria for myocarditis features. Three methods were compared against judgments by three clinicians. Simple rules did well for lab markers but poorly for symptoms. The model did well for symptoms and imaging, and a combined method did well across most areas and got the outcome right for most test patients.
Catalog record
| CREATOR | Fraser W. Gaspar, Seth B. Roberts, Francis X. Campion, Lynda Hoeksema, Jessica Kasirsky, Raymond Duncan, Matthew Sonesen, Renier Estiandan, Khyzer B. Aziz, Julie A. Riddler, Jessica Zhou, Hussein Ezzeldin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |