Accuracy of Large Language Models to Identify Stroke Subtypes Within Unstructured Electronic Health Record Data.

Dylan Owens¹, Danh Q Nguyen¹, Michael Dohopolski^2,3, Justin F Rousseau^4,5, Eric D Peterson¹, Ann Marie Navar¹

¹Department of Medicine (D.O., D.Q.N., E.D.P., A.M.N.), UT Southwestern Medical Center, Dallas, TX.

²Department of Radiation Oncology (M.D.), UT Southwestern Medical Center, Dallas, TX.

³Medical Artificial Intelligence and Automation Lab (M.D.), UT Southwestern Medical Center, Dallas, TX.

⁴Biostatistics and Clinical Informatics Section, Department of Neurology (J.F.R.), UT Southwestern Medical Center, Dallas, TX.

⁵Peter O'Donnell Jr. Brain Institute (J.F.R.), UT Southwestern Medical Center, Dallas, TX.

PubMed PMC DOI PDF

Abstract

BACKGROUND: While International Classification of Diseases, Tenth Revision codes suffice for identifying stroke events in surveillance, accurately classifying stroke types and subtypes using electronic health records remains challenging due to limitations in structured data. This often necessitates manual review of clinical documentation. This study evaluated whether a large language model, Genera…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.