{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T17:15:09Z","timestamp":1781716509371,"version":"3.54.5"},"reference-count":30,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T00:00:00Z","timestamp":1759190400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T00:00:00Z","timestamp":1759190400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"National Institutes of Health, The National Institute of Biomedical Imaging and Bioengineering (NIBIB), Public Trust of Artificial Intelligence in the Precision CDS Health Ecosystem","award":["1-RO1-EB030492"],"award-info":[{"award-number":["1-RO1-EB030492"]}]},{"name":"National Institutes of Health, The National Institute of Biomedical Imaging and Bioengineering (NIBIB), Public Trust of Artificial Intelligence in the Precision CDS Health Ecosystem","award":["1-RO1-EB030492"],"award-info":[{"award-number":["1-RO1-EB030492"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Application of large language models in healthcare continues to expand, specifically for medical free-text classification tasks. While foundation models like those from ChatGPT show potential, alternative models demonstrate superior accuracy and lower costs. This study underscores significant challenges, including computational costs and model reliability. Amidst rising healthcare expenditures and AI\u2019s perceived potential to reduce costs, a combination of local and commercial models might offer balanced solutions for healthcare systems.<\/jats:p>","DOI":"10.1038\/s41746-025-01971-x","type":"journal-article","created":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T12:07:09Z","timestamp":1759234029000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Generative AI costs in large healthcare systems, an example in revenue cycle"],"prefix":"10.1038","volume":"8","author":[{"given":"Michael L.","family":"Burns","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ssu-Ying","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chu-An","family":"Tsai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John","family":"Vandervest","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Balaji","family":"Pandian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Paige","family":"Nong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David A.","family":"Hanauer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew","family":"Rosenberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jodyn","family":"Platt","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,9,30]]},"reference":[{"key":"1971_CR1","doi-asserted-by":"crossref","unstructured":"Maliha G., Gerke S., Cohen I.G. & Parikh R.B. Artificial Intelligence and Liability in Medicine: Balancing Safety and Innovation. Milbank Q. 99, 629\u2013647 (2021).","DOI":"10.1111\/1468-0009.12504"},{"key":"1971_CR2","doi-asserted-by":"publisher","first-page":"103283","DOI":"10.1016\/j.inffus.2025.103283","volume":"123","author":"J Wu","year":"2025","unstructured":"Wu, J., He, K., Mao, R., Shang, X. & Cambria, E. Harnessing the potential of multimodal EHR data: a comprehensive survey of clinical predictive modeling for intelligent healthcare. Inf. Fusion 123, 103283 (2025).","journal-title":"Inf. Fusion"},{"key":"1971_CR3","doi-asserted-by":"crossref","unstructured":"He, K. et al. A survey of large language models for healthcare: from data, technology, and applications to accountability and ethics. Inform. Fusion. 118, C (2025).","DOI":"10.1016\/j.inffus.2025.102963"},{"key":"1971_CR4","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1001\/jama.2024.0555","volume":"331","author":"KE Goodman","year":"2024","unstructured":"Goodman, K. E., Yi, P. H. & Morgan, D. J. AI-generated clinical summaries require more than accuracy. JAMA 331, 637\u2013638 (2024).","journal-title":"JAMA"},{"key":"1971_CR5","unstructured":"ChatGPT sets record for fastest-growing user base - analyst note. https:\/\/www.reuters.com\/technology\/chatgpt-sets-record-fastest-growing-user-base-analyst-note-2023-02-01\/Reuters (2023)."},{"key":"1971_CR6","doi-asserted-by":"publisher","first-page":"e240622","DOI":"10.1001\/jamahealthforum.2024.0622","volume":"5","author":"MM Mello","year":"2024","unstructured":"Mello, M. M. & Rose, S. Denial-artificial intelligence tools and health insurance coverage decisions. JAMA Health Forum 5, e240622 (2024).","journal-title":"JAMA Health Forum"},{"key":"1971_CR7","doi-asserted-by":"publisher","unstructured":"Johnson, M., Albizri, A. & Harfouche, A. Responsible artificial intelligence in healthcare: predicting and preventing insurance claim denials for economic and social wellbeing. Inf. Syst. Front. https:\/\/doi.org\/10.1007\/s10796-021-10137-5 (2021).","DOI":"10.1007\/s10796-021-10137-5"},{"key":"1971_CR8","doi-asserted-by":"crossref","unstructured":"Pal, S. et al. Driving impact in claims denial management using artificial intelligence. In International Conference on Advances in Computing and Data Sciences 107\u2013120 (Springer International Publishing, Cham, 2022).","DOI":"10.1007\/978-3-031-12638-3_10"},{"key":"1971_CR9","doi-asserted-by":"publisher","first-page":"e22461","DOI":"10.2196\/22461","volume":"5","author":"H Joo","year":"2021","unstructured":"Joo, H., Burns, M., Kalidaikurichi Lakshmanan, S. S., Hu, Y. & Vydiswaran, V. G. V. Neural machine translation-based automated current procedural terminology classification system using procedure text: development and validation study. JMIR Form. Res. 5, e22461 (2021).","journal-title":"JMIR Form. Res."},{"key":"1971_CR10","doi-asserted-by":"publisher","first-page":"738","DOI":"10.1097\/ALN.0000000000003150","volume":"132","author":"ML Burns","year":"2020","unstructured":"Burns, M. L. et al. Classification of current procedural terminology codes from electronic health record data using machine learning. Anesthesiology 132, 738\u2013749 (2020).","journal-title":"Anesthesiology"},{"key":"1971_CR11","unstructured":"Bird Guru Guruganesh, B. et al. Big bird: transformers for longer sequences. In: NIPS'20: Proceedings of the 34th International Conference on Neural Information Processing Systems 17283\u201317297 (2020)."},{"key":"1971_CR12","unstructured":"Li, Y., Wehbe, R. M., Ahmad, F. S., Wang, H. & Luo, Y. Clinical-longformer and clinical-bigbird: transformers for long clinical sequences. arXiv https:\/\/arxiv.org\/abs\/2201.11838 (2022)."},{"key":"1971_CR13","doi-asserted-by":"publisher","first-page":"826","DOI":"10.4338\/ACI-2017-03-CR-0046","volume":"08","author":"MP Sendak","year":"2017","unstructured":"Sendak, M. P., Balu, S. & Schulman, K. A. Barriers to achieving economies of scale in analysis of EHR data. Appl. Clin. Inform. 08, 826\u2013831 (2017).","journal-title":"Appl. Clin. Inform."},{"key":"1971_CR14","doi-asserted-by":"crossref","unstructured":"Nagarajan, R. et al. \u201cEconomics and equity of large language models: health care perspective.\u201d J. Med. Internet Res 26, e64226 (2024).","DOI":"10.2196\/64226"},{"key":"1971_CR15","unstructured":"Developer Bazaar Technologies. How much does it cost to hire ai developers? Developer Bazaar Technologies https:\/\/www.developerbazaar.com\/how-much-does-it-cost-to-hire-ai-developers\/ (2024)."},{"key":"1971_CR16","doi-asserted-by":"publisher","first-page":"117","DOI":"10.1038\/s41746-025-01479-4","volume":"8","author":"P Nong","year":"2025","unstructured":"Nong, P., Maurer, E. & Dwivedi, R. The urgency of centering safety-net organizations in AI governance. NPJ Digit. Med. 8, 117 (2025).","journal-title":"NPJ Digit. Med."},{"key":"1971_CR17","unstructured":"Alkhaldi, N. Assessing the cost of implementing AI in healthcare \u2014. ITRex https:\/\/itrexgroup.com\/blog\/assessing-the-costs-of-implementing-ai-in-healthcare\/ (2024)."},{"key":"1971_CR18","doi-asserted-by":"publisher","first-page":"320","DOI":"10.1038\/s41746-024-01315-1","volume":"7","author":"E Klang","year":"2024","unstructured":"Klang, E. et al. A strategy for cost-effective large language model use at health system-scale. NPJ Digit. Med. 7, 320 (2024).","journal-title":"NPJ Digit. Med."},{"key":"1971_CR19","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1055\/a-2491-3872","volume":"16","author":"A Simmons","year":"2025","unstructured":"Simmons, A. et al. Extracting international classification of diseases codes from clinical documentation using large language models. Appl. Clin. Inform. 16, 337\u2013344 (2025).","journal-title":"Appl. Clin. Inform."},{"key":"1971_CR20","doi-asserted-by":"crossref","unstructured":"Soroush, A. et al. Assessing GPT-3.5 and GPT-4 in generating international classification of diseases billing codes. medRxiv 2023\u20132027 https:\/\/www.medrxiv.org\/content\/10.1101\/2023.07.07.23292391v2 (2023).","DOI":"10.1101\/2023.07.07.23292391"},{"key":"1971_CR21","doi-asserted-by":"publisher","DOI":"10.1007\/s10916-025-02149-4","volume":"49","author":"JM Roy","year":"2025","unstructured":"Roy, J. M. et al. Evaluating large language models for automated CPT code prediction in endovascular neurosurgery. J. Med. Syst. 49, 15 (2025).","journal-title":"J. Med. Syst."},{"key":"1971_CR22","doi-asserted-by":"crossref","unstructured":"Soroush, A. et al. Large language models are poor medical coders-benchmarking of medical code querying. NEJM AI 1 (2024).","DOI":"10.1056\/AIdbp2300040"},{"key":"1971_CR23","doi-asserted-by":"publisher","unstructured":"Isch, E. L. et al. Bridging the coding gap: assessing large language models for accurate modifier assignment in craniofacial operative notes. J. Craniofac. Surg. https:\/\/doi.org\/10.1097\/SCS.0000000000011390 (2025).","DOI":"10.1097\/SCS.0000000000011390"},{"key":"1971_CR24","doi-asserted-by":"publisher","first-page":"e60164","DOI":"10.2196\/60164","volume":"12","author":"M Nunes","year":"2024","unstructured":"Nunes, M., Bone, J., Ferreira, J. C. & Elvas, L. B. Health care language models and their fine-tuning for information extraction: scoping review. JMIR Med. Inform. 12, e60164 (2024).","journal-title":"JMIR Med. Inform."},{"key":"1971_CR25","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1016\/j.imed.2025.03.002","volume":"5","author":"X Chen","year":"2025","unstructured":"Chen, X. et al. Evaluating large language models and agents in healthcare: key challenges in clinical applications. Intell. Med. 5, 151\u2013163 (2025).","journal-title":"Intell. Med."},{"key":"1971_CR26","doi-asserted-by":"crossref","unstructured":"Kim, M. et al. Fine-tuning LLMs with medical data: can safety be ensured? NEJM AI 2 (2025).","DOI":"10.1056\/AIcs2400390"},{"key":"1971_CR27","unstructured":"State of AI in the Enterprise 2022. Deloitte United States https:\/\/www2.deloitte.com\/us\/en\/pages\/consulting\/articles\/state-of-ai-2022.html (2022)."},{"key":"1971_CR28","unstructured":"AI in Hospitals: Reducing Burnout, Improving Margins. Deloitte United States https:\/\/www2.deloitte.com\/us\/en\/pages\/consulting\/articles\/artificial-intelligence-in-hospitals-financial-performance-clinical-burnout.html (2024)."},{"key":"1971_CR29","unstructured":"Historical. https:\/\/www.cms.gov\/data-research\/statistics-trends-and-reports\/national-health-expenditure-data\/historical Accessed June 2025."},{"key":"1971_CR30","unstructured":"The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. https:\/\/www.equator-network.org\/reporting-guidelines\/strobe\/ Accessed June 2025."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01971-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01971-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01971-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T12:07:16Z","timestamp":1759234036000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01971-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,30]]},"references-count":30,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1971"],"URL":"https:\/\/doi.org\/10.1038\/s41746-025-01971-x","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,30]]},"assertion":[{"value":"19 May 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 August 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 September 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"M.B. and J.V. are co-inventors on patent No. 11,288,445 B2 entitled \u201cAutomated System and Method for Assigning Billing Codes to Medical Procedures,\u201d related to the use of ML techniques for medical procedural billing. M.B. and J.V. reported holding equity in the company, Decimal Code. No other disclosures were reported.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"579"}}