{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,10]],"date-time":"2025-12-10T16:03:05Z","timestamp":1765382585511,"version":"3.46.0"},"reference-count":25,"publisher":"Oxford University Press (OUP)","issue":"12","license":[{"start":{"date-parts":[[2025,10,8]],"date-time":"2025-10-08T00:00:00Z","timestamp":1759881600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Optum AI"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Longitudinal data in electronic health records (EHRs) represent an individual\u2019s clinical history through a sequence of codified concepts, including diagnoses, procedures, medications, and laboratory tests. Generative pretrained transformers (GPT) can leverage this data to predict future events. While fine-tuning of these models can enhance task-specific performance, it becomes costly when applied to many clinical prediction tasks. In contrast, a pretrained foundation model can be used in zero-shot forecasting setting, offering a scalable alternative to fine-tuning separate models for each outcome.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>This study presents the first comprehensive analysis of zero-shot forecasting with GPT-based foundational models in EHRs, introducing a novel pipeline that formulates medical concept prediction as a generative modeling task. Unlike supervised approaches requiring extensive labeled data, our method enables the model to forecast the next medical event purely from a pretraining knowledge. We evaluate performance across multiple time horizons and clinical categories, demonstrating model\u2019s ability to capture latent temporal dependencies and complex patient trajectories without task supervision.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The model\u2019s performance in predicting the next medical concept was evaluated using precision and recall metrics, achieving an average top-1 precision of 0.614 and recall of 0.524. For 12 major diagnostic conditions, the model demonstrated strong zero-shot performance, achieving high true positive rates while maintaining low false positives.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>We demonstrate the power of a foundational EHR GPT model in capturing diverse phenotypes and enabling robust, zero-shot forecasting of clinical outcomes. This capability highlights both its versatility across conditions like liver cancer and SLE, and its limitations in more ambiguous settings such as depression, while also revealing meaningful latent clinical structure.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This capability enhances the versatility of predictive healthcare models and reduces the need for task-specific training, enabling more scalable applications in clinical settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/jamia\/ocaf160","type":"journal-article","created":{"date-parts":[[2025,9,19]],"date-time":"2025-09-19T12:05:07Z","timestamp":1758283507000},"page":"1833-1842","source":"Crossref","is-referenced-by-count":1,"title":["Zero-shot medical event prediction using a generative pretrained transformer on electronic health records"],"prefix":"10.1093","volume":"32","author":[{"given":"Ekaterina","family":"Redekop","sequence":"first","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Bioengineering, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Zichen","family":"Wang","sequence":"additional","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Bioengineering, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Rushikesh","family":"Kulkarni","sequence":"additional","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Mara","family":"Pleasure","sequence":"additional","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Bioinformatics, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Aaron","family":"Chin","sequence":"additional","affiliation":[{"name":"Division of Immunology, Allergy and Rheumatology, Department of Pediatrics, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Hamid Reza","family":"Hassanzadeh","sequence":"additional","affiliation":[{"name":"Optum AI , Eden Prairie, MN 55344,","place":["United States"]}]},{"given":"Brian L","family":"Hill","sequence":"additional","affiliation":[{"name":"Optum AI , Eden Prairie, MN 55344,","place":["United States"]}]},{"given":"Melika","family":"Emami","sequence":"additional","affiliation":[{"name":"Optum AI , Eden Prairie, MN 55344,","place":["United States"]}]},{"given":"William F","family":"Speier","sequence":"additional","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]},{"given":"Corey W","family":"Arnold","sequence":"additional","affiliation":[{"name":"Biomedical AI Research Lab, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Radiology, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]},{"name":"Department of Bioengineering, University of California, Los Angeles (UCLA) , Los Angeles, CA 90024,","place":["United States"]}]}],"member":"286","published-online":{"date-parts":[[2025,10,8]]},"reference":[{"key":"2025112514465763200_ocaf160-B1","doi-asserted-by":"publisher","first-page":"7857","DOI":"10.1038\/s41467-023-43715-z","article-title":"TransformEHR: transformer-based encoder-decoder generative model to enhance prediction of disease outcomes using electronic health records","volume":"14","author":"Yang","year":"2023","journal-title":"Nat Commun"},{"key":"2025112514465763200_ocaf160-B2","doi-asserted-by":"crossref","first-page":"e281","DOI":"10.1016\/S2589-7500(24)00025-6","article-title":"Foresight\u2014a generative pretrained transformer for modelling of patient timelines using electronic health records: a retrospective modelling study","volume":"6","author":"Kraljevic","year":"2024","journal-title":"Lancet Digit Health"},{"year":"2024","author":"Pang","key":"2025112514465763200_ocaf160-B3"},{"key":"2025112514465763200_ocaf160-B4","doi-asserted-by":"crossref","first-page":"3121","DOI":"10.1109\/JBHI.2021.3063721","article-title":"Bidirectional representation learning from transformers using multimodal electronic health record data to predict depression","volume":"25","author":"Meng","year":"2021","journal-title":"IEEE J Biomed Health Inform"},{"key":"2025112514465763200_ocaf160-B5","doi-asserted-by":"crossref","first-page":"86","DOI":"10.1038\/s41746-021-00455-y","article-title":"Med-BERT: pretrained contextualized embeddings on large-scale structured electronic health records for disease prediction","volume":"4","author":"Rasmy","year":"2021","journal-title":"NPJ Digit Med"},{"key":"2025112514465763200_ocaf160-B6","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"first-page":"301","year":"2016","author":"Choi","key":"2025112514465763200_ocaf160-B7"},{"year":"2023","author":"Bellamy","key":"2025112514465763200_ocaf160-B8"},{"key":"2025112514465763200_ocaf160-B9","doi-asserted-by":"crossref","first-page":"3880","DOI":"10.3390\/cancers15153880","article-title":"Advances in the early detection of hepatobiliary cancers","volume":"15","author":"Y\u0131ld\u0131r\u0131m","year":"2023","journal-title":"Cancers (Basel)"},{"key":"2025112514465763200_ocaf160-B10","doi-asserted-by":"crossref","first-page":"pkad034","DOI":"10.1093\/jncics\/pkad034","article-title":"Opportunities to address gaps in early detection and improve outcomes of liver cancer","volume":"7","author":"McMahon","year":"2023","journal-title":"JNCI Cancer Spectr"},{"key":"2025112514465763200_ocaf160-B11","doi-asserted-by":"crossref","first-page":"1348","DOI":"10.1016\/j.jhep.2022.01.014","article-title":"Artificial intelligence for the prevention and clinical management of hepatocellular carcinoma","volume":"76","author":"Calderaro","year":"2022","journal-title":"J Hepatol"},{"key":"2025112514465763200_ocaf160-B12","doi-asserted-by":"crossref","first-page":"890522","DOI":"10.3389\/fimmu.2022.890522","article-title":"Understanding the concept of pre-clinical autoimmunity: prediction and prevention of systemic lupus erythematosus: identifying risk factors and developing strategies against disease development","volume":"13","author":"Choi","year":"2022","journal-title":"Front Immunol"},{"key":"2025112514465763200_ocaf160-B13","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1177\/0961203320983445","article-title":"Delayed diagnosis adversely affects outcome in systemic lupus erythematosus: cross sectional analysis of the LuLa cohort","volume":"30","author":"Kernder","year":"2021","journal-title":"Lupus"},{"key":"2025112514465763200_ocaf160-B14","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.comppsych.2012.06.006","article-title":"The diagnosis of depression: current and emerging methods","volume":"54","author":"Smith","year":"2013","journal-title":"Compr Psychiatry"},{"key":"2025112514465763200_ocaf160-B15","doi-asserted-by":"crossref","first-page":"544","DOI":"10.1017\/S2045796018000227","article-title":"Empirical evidence for definitions of episode, remission, recovery, relapse and recurrence in depression: a systematic review","volume":"28","author":"de Zwart","year":"2019","journal-title":"Epidemiol Psychiatr Sci"},{"year":"2018","author":"McInnes","key":"2025112514465763200_ocaf160-B16"},{"key":"2025112514465763200_ocaf160-B17","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1097\/CU9.0000000000000021","article-title":"Trimester-based changes in urogenital symptoms and their impact on the quality of life in pregnant women: a preliminary report","volume":"15","author":"Uzelpasaci","year":"2021","journal-title":"Curr Urol"},{"key":"2025112514465763200_ocaf160-B18","doi-asserted-by":"crossref","first-page":"549","DOI":"10.1016\/j.psc.2009.05.004","article-title":"The neurobiology of anxiety disorders: brain imaging, genetics, and psychoneuroendocrinology","volume":"32","author":"Martin","year":"2009","journal-title":"Psychiatr Clin North Am"},{"key":"2025112514465763200_ocaf160-B19","doi-asserted-by":"crossref","first-page":"927804","DOI":"10.1155\/2014\/927804","article-title":"Behavioral and psychological symptoms in Alzheimer\u2019s disease","volume":"2014","author":"Li","year":"2014","journal-title":"Biomed Res Int"},{"key":"2025112514465763200_ocaf160-B20","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1093\/brain\/awu335","article-title":"Vision and vision-related outcome measures in multiple sclerosis","volume":"138","author":"Balcer","year":"2015","journal-title":"Brain"},{"key":"2025112514465763200_ocaf160-B21","doi-asserted-by":"crossref","first-page":"e0222731","DOI":"10.1371\/journal.pone.0222731","article-title":"The frequency of bowel and bladder problems in multiple sclerosis and its relation to fatigue: a single centre experience","volume":"14","author":"Lin","year":"2019","journal-title":"PLoS One"},{"key":"2025112514465763200_ocaf160-B22","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1097\/WCO.0b013e328346a43b","article-title":"Cognition in multiple sclerosis","volume":"24","author":"Langdon","year":"2011","journal-title":"Curr Opin Neurol"},{"key":"2025112514465763200_ocaf160-B23","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1016\/j.cortex.2017.03.016","article-title":"Resting-state functional reorganization in Parkinson\u2019s disease: an activation likelihood estimation meta-analysis","volume":"92","author":"Tahmasian","year":"2017","journal-title":"Cortex"},{"key":"2025112514465763200_ocaf160-B24","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1111\/j.1365-2990.1989.tb01147.x","article-title":"The significance of the Lewy body in the diagnosis of idiopathic Parkinson\u2019s disease","volume":"15","author":"Gibb","year":"1989","journal-title":"Neuropathol Appl Neurobiol"},{"key":"2025112514465763200_ocaf160-B25","doi-asserted-by":"crossref","first-page":"1620","DOI":"10.1111\/j.1475-6773.2005.00444.x","article-title":"Measuring diagnoses: ICD code accuracy","volume":"40","author":"O'Malley","year":"2005","journal-title":"Health Serv Res"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/32\/12\/1833\/64549073\/ocaf160.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/32\/12\/1833\/64549073\/ocaf160.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T19:47:07Z","timestamp":1764100027000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/32\/12\/1833\/8277596"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,8]]},"references-count":25,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2025,10,8]]},"published-print":{"date-parts":[[2025,12,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocaf160","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"type":"print","value":"1067-5027"},{"type":"electronic","value":"1527-974X"}],"subject":[],"published-other":{"date-parts":[[2025,12]]},"published":{"date-parts":[[2025,10,8]]}}}