biostudies-literatureFeng RNHLBI NIH HHSe013023https://www.ebi.ac.uk/biostudies/studies/S-EPMC11771986biostudies-literatureUnknown18(1)<h4>Background</h4>Large language models (LLMs) such as Chat Generative Pre-trained Transformer (ChatGPT) excel at interpreting unstructured data from public sources, yet are limited when responding to queries on private repositories, such as electronic health records (EHRs). We hypothesized that prompt engineering could enhance the accuracy of LLMs for interpreting EHR data without requiring domain knowledge, thus expanding their utility for patients and personalized diagnostics.<h4>Methods</h4>We designed and systematically tested prompt engineering techniques to improve the ability of LLMs to interpret EHRs for nuanced diagnostic questions, referenced to a panel of medical experts. In 490 full-text EHR notes from 125 patients with prior life-threatening heart rhythm disorders, we asked Circulation. Arrhythmia and electrophysiologyEngineering of Generative Artificial Intelligence and Natural Language Processing Models to Accurately Identify Arrhythmia Recurrence.PMC11771986R01 HL162260R01 HL149134R01 HL083359F32 HL144101K23 HL145017Narayan SMDe Larochelliere HDesai YBClopton PRogers AJGanesan PRuiperez-Campillo SDeb BRodrigo MPerez MVBaykaner TChang HJFeng RAzizi ZGoyal JPedron MBrennan KAfalseEngineering of Generative Artificial Intelligence and Natural Language Processing Models to Accurately Identify Arrhythmia Recurrence.<h4>Background</h4>Large language models (LLMs) such as Chat Generative Pre-trained Transformer (ChatGPT) excel at interpreting unstructured data from public sources, yet are limited when responding to queries on private repositories, such as electronic health records (EHRs). We hypothesized that prompt engineering could enhance the accuracy of LLMs for interpreting EHR data without requiring domain knowledge, thus expanding their utility for patients and personalized diagnostics.<h4>Methods</h4>We designed and systematically tested prompt engineering techniques to improve the ability of LLMs to interpret EHRs for nuanced diagnostic questions, referenced to a panel of medical experts. In 490 full-text EHR notes from 125 patients with prior life-threatening heart rhythm disorders, we asked 2025-01-01T00:00:00Z2025 Jan2026-06-06T07:42:35.31Z2026-05-27T03:11:49.022ZS-EPMC117719863967664210.1161/CIRCEP.124.01302310.1161/circep.124.013023