{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T14:01:40Z","timestamp":1789740100003,"version":"4.0.1"},"reference-count":30,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000},"content-version":"unspecified","delay-in-days":17,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Noncommunicable Chronic Diseases-National Science and Technology Major Project","award":["2023ZD0501700"],"award-info":[{"award-number":["2023ZD0501700"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["72474010, 72622003, 72125009"],"award-info":[{"award-number":["72474010, 72622003, 72125009"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Health Care Inform"],"accepted":{"date-parts":[[2026,8,19]]},"published-print":{"date-parts":[[2026,9]]},"abstract":"<jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Chronic kidney disease (CKD) presents a growing public health challenge in China, exacerbated by low patient awareness and limited nephrology resources. This study evaluated the potential of large language models (LLMs) to support CKD diagnosis in primary care using time-series electronic health record (EHR) data.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Longitudinal data were extracted from the EHR system of Weinan, China. Among 29\u2009963 adults meeting inclusion criteria, 2300 participants were randomly sampled to generate clinical vignettes. CKD status was determined using diagnosis data, and individuals with early kidney injury indicated by laboratory abnormalities who lack a formal diagnosis were also identified. Using the DeepSeek LLM with prompt engineering, we generated binary CKD diagnoses and probability scores based on EHR data. Diagnostic performance was evaluated using accuracy, sensitivity, specificity, F1 score, area under the curve (AUC) and the detection rate for early kidney injury and compared with traditional machine learning (ML) models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Using 1-month EHR data, the LLM achieved an accuracy of 87.0%, AUC of 0.921, F1 score of 0.579, sensitivity of 89.1%, specificity of 86.8%, and detection rate for early kidney injury of 42.2%. With 1-year data, sensitivity increased to 92.2% and early detection to 52.6%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The LLM outperformed ML models across most metrics, achieving diagnostic performance comparable to clinicians and further enhancing detection for early kidney injury.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>LLMs applied to time-series EHR data may serve as a clinical decision support tool to improve CKD diagnosis in primary care, with particular value in resource-limited settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjhci-2025-101830","type":"journal-article","created":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T13:05:25Z","timestamp":1789736725000},"page":"e101830","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Leveraging time-series electronic health records with large language models for chronic kidney disease diagnosis in primary care"],"prefix":"10.1136","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3519-3952","authenticated-orcid":false,"given":"Jingyi","family":"Wu","sequence":"first","affiliation":[{"name":"National Institute of Health Data Science, Peking University, Beijing, China"},{"name":"Advanced Institute of Information Technology, Peking University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-9807-1802","authenticated-orcid":false,"given":"Yuewen","family":"Zheng","sequence":"additional","affiliation":[{"name":"Advanced Institute of Information Technology, Peking University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-2004-8645","authenticated-orcid":false,"given":"Zhijun","family":"He","sequence":"additional","affiliation":[{"name":"Independent Researcher, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-1977-6711","authenticated-orcid":false,"given":"Qianlin","family":"Zuo","sequence":"additional","affiliation":[{"name":"Advanced Institute of Information Technology, Peking University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-8303-3543","authenticated-orcid":false,"given":"Weidong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Weinan Municipal Health Commission, Weinan, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-0873-5992","authenticated-orcid":false,"given":"Pengfei","family":"Li","sequence":"additional","affiliation":[{"name":"National Institute of Health Data Science, Peking University, Beijing, China"},{"name":"Advanced Institute of Information Technology, Peking University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2349-2936","authenticated-orcid":false,"given":"Zhang","family":"Luxia","sequence":"additional","affiliation":[{"name":"National Institute of Health Data Science, Peking University, Beijing, China"},{"name":"Advanced Institute of Information Technology, Peking University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,9,18]]},"reference":[{"key":"2026091806051478000_33.1.e101830.1","doi-asserted-by":"publisher","DOI":"10.1016\/s0140-6736(24)00933-4"},{"key":"2026091806051478000_33.1.e101830.2","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2021-057500","article-title":"MicroRNAs associated with chronic kidney disease in the general population and high-risk subgroups: protocol for a systematic review and meta-analysis","volume":"12","author":"Motshwari","year":"2022","journal-title":"BMJ Open"},{"key":"2026091806051478000_33.1.e101830.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41581-020-00363-6"},{"key":"2026091806051478000_33.1.e101830.4","doi-asserted-by":"crossref","DOI":"10.3390\/diagnostics10060415","article-title":"Comparison between Statistical Models and Machine Learning Methods on Classification for Highly Imbalanced Multiclass Kidney Data","volume":"10","author":"Jeong","year":"2020","journal-title":"Diagnostics (Basel)"},{"key":"2026091806051478000_33.1.e101830.5","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-025-02889-w","article-title":"Bio inspired feature selection and graph learning for sepsis risk stratification","volume":"15","author":"Siri","year":"2025","journal-title":"Sci Rep"},{"key":"2026091806051478000_33.1.e101830.6","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1038\/s41551-021-00745-6","article-title":"Deep-learning models for the detection and incidence prediction of chronic kidney disease and type 2 diabetes from retinal fundus images","volume":"5","author":"Zhang","year":"2021","journal-title":"Nat Biomed Eng"},{"key":"2026091806051478000_33.1.e101830.7","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-025-94905-2","article-title":"Fine-tuned deep learning models for early detection and classification of kidney conditions in CT imaging","volume":"15","author":"Pimpalkar","year":"2025","journal-title":"Sci Rep"},{"key":"2026091806051478000_33.1.e101830.8","doi-asserted-by":"crossref","first-page":"1342","DOI":"10.1007\/s12149-021-01676-7","article-title":"Artificial neural network for the prediction model of glomerular filtration rate to estimate the normal or abnormal stages of kidney using gamma camera","volume":"35","author":"Hossain","year":"2021","journal-title":"Ann Nucl Med"},{"key":"2026091806051478000_33.1.e101830.9","doi-asserted-by":"crossref","DOI":"10.1093\/database\/baaa010","article-title":"Artificial intelligence with multi-functional machine learning platform development for better healthcare and precision medicine","volume":"2020","author":"Ahmed","year":"2020","journal-title":"Database (Oxford)"},{"key":"2026091806051478000_33.1.e101830.10","doi-asserted-by":"crossref","DOI":"10.1186\/s12967-021-02714-8","article-title":"A theoretical model of health management using data-driven decision-making: the future of precision medicine and health","volume":"19","author":"Kriegova","year":"2021","journal-title":"J Transl Med"},{"key":"2026091806051478000_33.1.e101830.11","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1007\/s00296-023-05464-6","article-title":"Diagnostic accuracy of a large language model in rheumatology: comparison of physician and ChatGPT-4","volume":"44","author":"Krusche","year":"2024","journal-title":"Rheumatol Int"},{"key":"2026091806051478000_33.1.e101830.12","doi-asserted-by":"crossref","DOI":"10.1111\/bpa.13207","article-title":"Evaluating the performance of large language models: ChatGPT and Google Bard in generating differential diagnoses in clinicopathological conferences of neurodegenerative disorders","volume":"34","author":"Koga","year":"2024","journal-title":"Brain Pathol"},{"key":"2026091806051478000_33.1.e101830.13","doi-asserted-by":"crossref","unstructured":"Lv X , Liu D , Chen X , et al . Machine learning for the prediction of mortality in patients with sepsis-associated acute kidney injury: a systematic review and meta-analysis. BMC Infect Dis 2024;24. doi:10.1186\/s12879-024-10380-6","DOI":"10.1186\/s12879-024-10380-6"},{"key":"2026091806051478000_33.1.e101830.14","doi-asserted-by":"crossref","unstructured":"Mei Z-F , Yan J , Qian L , et al . Enrichment of Circulating Tumor Cells of Lung Cancer and Correlation With Serum Leukomonocyte and Tumor Biomarkers: A Retrospective Study. Technol Cancer Res Treat 2023;22. doi:10.1177\/15330338231167827","DOI":"10.1177\/15330338231167827"},{"key":"2026091806051478000_33.1.e101830.15","doi-asserted-by":"crossref","first-page":"99","DOI":"10.2147\/DMSO.S228335","article-title":"Serum Ferritin Independently Predicts the Incidence of Chronic Kidney Disease in Patients with Type 2 Diabetes Mellitus","volume":"13","author":"Wu","year":"2020","journal-title":"Diabetes Metab Syndr Obes"},{"key":"2026091806051478000_33.1.e101830.16","doi-asserted-by":"crossref","unstructured":"Sandmann S , Hegselmann S , Fujarski M , et al . Benchmark evaluation of DeepSeek large language models in clinical decision-making. Nat Med 2025. doi:10.1038\/s41591-025-03727-2","DOI":"10.1038\/s41591-025-03727-2"},{"key":"2026091806051478000_33.1.e101830.17","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1016\/j.aopr.2025.05.001","article-title":"DeepSeek-R1 outperforms Gemini 2.0 Pro, OpenAI o1, and o3-mini in bilingual complex ophthalmology reasoning","volume":"5","author":"Xu","year":"2025","journal-title":"Adv Ophthalmol Pract Res"},{"key":"2026091806051478000_33.1.e101830.18","doi-asserted-by":"crossref","unstructured":"Sundstr\u00f6m J , Bodegard J , Bollmann A , et al . Prevalence, outcomes, and cost of chronic kidney disease in a contemporary population of 2.4 million patients from 11 countries: The CaReMe CKD study. Lancet Regional Health-Europe 2022;20. doi:10.1016\/j.lanepe.2022.100438","DOI":"10.1016\/j.lanepe.2022.100438"},{"key":"2026091806051478000_33.1.e101830.19","doi-asserted-by":"crossref","unstructured":"Dong Z , Wang Q , Ke Y , et al . Prediction of 3-year risk of diabetic kidney disease using machine learning based on electronic medical records. J Transl Med 2022;20. doi:10.1186\/s12967-022-03339-1","DOI":"10.1016\/j.ekir.2022.01.346"},{"key":"2026091806051478000_33.1.e101830.20","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1001\/jama.2023.8288","article-title":"Accuracy of a Generative Artificial Intelligence Model in a Complex Diagnostic Challenge","volume":"330","author":"Kanjee","year":"2023","journal-title":"JAMA"},{"key":"2026091806051478000_33.1.e101830.21","doi-asserted-by":"crossref","DOI":"10.2196\/59782","article-title":"Evaluating Medical Entity Recognition in Health Care: Entity Model Quantitative Study","volume":"12","author":"Liu","year":"2024","journal-title":"JMIR Med Inform"},{"key":"2026091806051478000_33.1.e101830.22","doi-asserted-by":"crossref","DOI":"10.2196\/63190","article-title":"Leveraging Large Language Models for Infectious Disease Surveillance\u2014Using a Web Service for Monitoring COVID-19 Patterns From Self-Reporting Tweets: Content Analysis","volume":"27","author":"Xie","year":"2025","journal-title":"J Med Internet Res"},{"key":"2026091806051478000_33.1.e101830.23","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-025-07628-9","article-title":"Enhancing breast cancer diagnosis through machine learning algorithms","volume":"15","author":"Amraei","year":"2025","journal-title":"Sci Rep"},{"key":"2026091806051478000_33.1.e101830.24","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-021-00428-1","article-title":"Medical records-based chronic kidney disease phenotype for clinical care and \u201cbig data\u201d observational and genetic studies","volume":"4","author":"Shang","year":"2021","journal-title":"NPJ Digit Med"},{"key":"2026091806051478000_33.1.e101830.25","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1016\/j.ekir.2023.10.019","article-title":"Clinical Decision Support Tools in the Electronic Medical Record","volume":"9","author":"Alexiuk","year":"2024","journal-title":"Kidney Int Rep"},{"key":"2026091806051478000_33.1.e101830.26","doi-asserted-by":"crossref","first-page":"e107","DOI":"10.1097\/ICL.0000000000001158","article-title":"Evaluation of Responses to Questions About Keratoconus Using ChatGPT-4.0, Google Gemini and Microsoft Copilot: A Comparative Study of Large Language Models on Keratoconus","volume":"51","author":"Demir","year":"2025","journal-title":"Eye Contact Lens"},{"key":"2026091806051478000_33.1.e101830.27","doi-asserted-by":"crossref","first-page":"1660","DOI":"10.1093\/jamia\/ocab051","article-title":"Using nursing notes to improve clinical outcome prediction in intensive care patients: A retrospective cohort study","volume":"28","author":"Huang","year":"2021","journal-title":"J Am Med Inform Assoc"},{"key":"2026091806051478000_33.1.e101830.28","doi-asserted-by":"crossref","first-page":"4622","DOI":"10.1109\/JBHI.2025.3578197","article-title":"Temporal Integrative Machine Learning for Early Detection of Diabetic Retinopathy Using Fundus Imaging and Electronic Health Records","volume":"30","author":"Messica","year":"2025","journal-title":"IEEE J Biomed Health Inform"},{"key":"2026091806051478000_33.1.e101830.29","doi-asserted-by":"crossref","DOI":"10.1177\/1479973120985594","article-title":"Spotlight on primary care management of COPD: Electronic health records","volume":"18","author":"Harries","year":"2021","journal-title":"Chron Respir Dis"},{"key":"2026091806051478000_33.1.e101830.30","doi-asserted-by":"crossref","unstructured":"Sun M , Ni Y , Wu X , et al . OM\u201085, a Bacterial Lysate, Reduces Pulmonary Nodule Malignant Probability: A Retrospective Study. Clinical Respiratory J 2025;19. doi:10.1111\/crj.70109","DOI":"10.1111\/crj.70109"}],"container-title":["BMJ Health &amp; Care Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjhci-2025-101830","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T13:05:33Z","timestamp":1789736733000},"score":1,"resource":{"primary":{"URL":"https:\/\/informatics.bmj.com\/lookup\/doi\/10.1136\/bmjhci-2025-101830"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":30,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,9,18]]},"published-print":{"date-parts":[[2026,9]]}},"alternative-id":["10.1136\/bmjhci-2025-101830"],"URL":"https:\/\/doi.org\/10.1136\/bmjhci-2025-101830","relation":{},"ISSN":["2632-1009"],"issn-type":[{"value":"2632-1009","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9]]}}}