{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T23:17:11Z","timestamp":1781306231153,"version":"3.54.1"},"reference-count":79,"publisher":"Springer Science and Business Media LLC","issue":"3","license":[{"start":{"date-parts":[[2025,9,10]],"date-time":"2025-09-10T00:00:00Z","timestamp":1757462400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,9,10]],"date-time":"2025-09-10T00:00:00Z","timestamp":1757462400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"University of Bern"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Large language models (LLMs) have been successfully used for data extraction from free-text radiology reports. Most current studies were conducted with LLMs accessed via an application programming interface (API). We evaluated the feasibility of using open-source LLMs, deployed on limited local hardware resources for data extraction from free-text mammography reports, using a common data element (CDE)-based structure. Seventy-nine CDEs were defined by an interdisciplinary expert panel, reflecting real-world reporting practice. Sixty-one reports were classified by two independent researchers to establish ground truth. Five different open-source LLMs deployable on a single GPU were used for data extraction using the\n                    <jats:italic>general-classifier<\/jats:italic>\n                    Python package. Extractions were performed for five different prompt approaches with calculation of overall accuracy, micro-recall and micro-F1. Additional analyses were conducted using thresholds for the relative probability of classifications. High inter-rater agreement was observed between manual classifiers (Cohen\u2019s kappa 0.83). Using default prompts, the LLMs achieved accuracies of 59.2\u201372.9%. Chain-of-thought prompting yielded mixed results, while few-shot prompting led to decreased accuracy. Adaptation of the default prompts to precisely define classification tasks improved performance for all models, with accuracies of 64.7\u201385.3%. Setting certainty thresholds further improved accuracies to\u2009&gt;\u200990% but reduced the coverage rate to\u2009&lt;\u200950%. Locally deployed open-source LLMs can effectively extract information from mammography reports, maintaining compatibility with limited computational resources. Selection and evaluation of the model and prompting strategy are critical. Clear, task-specific instructions appear crucial for high performance. Using a CDE-based framework provides clear semantics and structure for the data extraction.\n                  <\/jats:p>","DOI":"10.1007\/s10278-025-01659-4","type":"journal-article","created":{"date-parts":[[2025,9,10]],"date-time":"2025-09-10T19:53:08Z","timestamp":1757533988000},"page":"2737-2751","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Implementing a Resource-Light and Low-Code Large Language Model System for Information Extraction from Mammography Reports: A Pilot Study"],"prefix":"10.1007","volume":"39","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5374-8720","authenticated-orcid":false,"given":"Fabio","family":"Dennst\u00e4dt","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Simon","family":"Fauser","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nikola","family":"Cihoric","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Max","family":"Schmerder","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Paolo","family":"Lombardo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Grazia Maria","family":"Cereghetti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sandro","family":"von D\u00e4niken","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas","family":"Minder","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaro","family":"Meyer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lawrence","family":"Chiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roberto","family":"Gaio","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luc","family":"Lerch","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Irina","family":"Filchenko","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Reichenpfader","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kerstin","family":"Denecke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Caslav","family":"Vojvodic","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Igor","family":"Tatalovic","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andr\u00e9","family":"Sander","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Janna","family":"Hastings","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel M","family":"Aebersold","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hendrik","family":"von Tengg-Kobligk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Knud","family":"Nairz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,9,10]]},"reference":[{"key":"1659_CR1","doi-asserted-by":"crossref","unstructured":"Neri E, Granata V, Montemezzi S, Belli P, Bernardi D, Brancato B, et al. Structured reporting of X-ray mammography in the first diagnosis of breast cancer: a Delphi consensus proposal. Radiol med. 2022 May;127(5):471\u201383.","DOI":"10.1007\/s11547-022-01478-5"},{"key":"1659_CR2","doi-asserted-by":"crossref","unstructured":"European Society of Radiology (ESR), Dos Santos DP, Kotter E, Mildenberger P, Mart\u00ed-Bonmat\u00ed L. ESR paper on structured reporting in radiology\u2014update 2023. Insights Imaging. 2023 Nov 23;14(1):199.","DOI":"10.1186\/s13244-023-01560-0"},{"key":"1659_CR3","doi-asserted-by":"crossref","unstructured":"Dromain C, Vullierme M, Hicks RJ, Prasad V, O\u2019Toole D, De Herder WW, et al. ENETS standardized (synoptic) reporting for radiological imaging in neuroendocrine tumours. J Neuroendocrinology. 2022 Mar;34(3):e13044.","DOI":"10.1111\/jne.13044"},{"key":"1659_CR4","doi-asserted-by":"crossref","unstructured":"Woller IA, Woller SC, Stevens SM, Lloyd JF, Conner KE, Gordon BH, et al. Synoptic reporting accuracy for computed tomography pulmonary arteriography among patients suspected of pulmonary embolism. J Am Coll Emerg Physicians Open. 2022 Oct;3(5):e12801.","DOI":"10.1002\/emp2.12801"},{"key":"1659_CR5","doi-asserted-by":"crossref","unstructured":"Hassanpour S, Langlotz CP. Information extraction from multi-institutional radiology reports. Artificial Intelligence in Medicine. 2016 Jan;66:29\u201339.","DOI":"10.1016\/j.artmed.2015.09.007"},{"key":"1659_CR6","doi-asserted-by":"crossref","unstructured":"M\u00e9tais E. Enhancing information systems management with natural language processing techniques. Data & Knowledge Engineering. 2002 Jun;41(2\u20133):247\u201372.","DOI":"10.1016\/S0169-023X(02)00043-5"},{"key":"1659_CR7","doi-asserted-by":"crossref","unstructured":"Casey A, Davidson E, Poon M, Dong H, Duma D, Grivas A, et al. A systematic review of natural language processing applied to radiology reports. BMC Med Inform Decis Mak. 2021 Dec;21(1):179.","DOI":"10.1186\/s12911-021-01533-7"},{"key":"1659_CR8","doi-asserted-by":"crossref","unstructured":"Kuling G, Curpen B, Martel AL. BI-RADS BERT and using section segmentation to understand radiology reports. J Imaging. 2022 May 9;8(5):131.","DOI":"10.3390\/jimaging8050131"},{"key":"1659_CR9","doi-asserted-by":"crossref","unstructured":"Reichenpfader D, Knupp J, Von D\u00e4niken SU, Gaio R, Dennst\u00e4dt F, Cereghetti GM, et al. Enhancing bidirectional encoder representations from transformers (BERT) with frame semantics to extract clinically relevant information from German mammography reports: algorithm development and validation. J Med Internet Res. 2025 Apr 25;27:e68427.","DOI":"10.2196\/68427"},{"key":"1659_CR10","doi-asserted-by":"crossref","unstructured":"Busch F, Hoffmann L, Dos Santos DP, Makowski MR, Saba L, Prucker P, et al. Large language models for structured reporting in radiology: past, present, and future. Eur Radiol. 2024 Oct 23","DOI":"10.1007\/s00330-024-11107-6"},{"key":"1659_CR11","doi-asserted-by":"crossref","unstructured":"Ali SR, Dobbs TD, Hutchings HA, Whitaker IS. Using ChatGPT to write patient clinic letters. The Lancet Digital Health. 2023 Apr;5(4):e179\u201381.","DOI":"10.1016\/S2589-7500(23)00048-1"},{"key":"1659_CR12","doi-asserted-by":"crossref","unstructured":"Benary M, Wang XD, Schmidt M, Soll D, Hilfenhaus G, Nassir M, et al. Leveraging large language models for decision support in personalized oncology. JAMA Netw Open. 2023 Nov 17;6(11):e2343689.","DOI":"10.1001\/jamanetworkopen.2023.43689"},{"key":"1659_CR13","doi-asserted-by":"crossref","unstructured":"Safranek CW, Sidamon-Eristoff AE, Gilson A, Chartash D. The role of large language models in medical education: applications and implications. JMIR Med Educ. 2023 Aug 14;9:e50945.","DOI":"10.2196\/50945"},{"key":"1659_CR14","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Zink J, Putora PM, Hastings J, Cihoric N. Title and abstract screening for literature reviews using large language models: an exploratory study in the biomedical domain. Syst Rev. 2024 Jun 15;13(1):158.","DOI":"10.1186\/s13643-024-02575-4"},{"key":"1659_CR15","doi-asserted-by":"crossref","unstructured":"Bhayana R, Nanda B, Dehkharghanian T, Deng Y, Bhambra N, Elias G, et al. Large language models for automated synoptic reports and resectability categorization in pancreatic cancer. Fowler K, editor. Radiology. 2024 Jun 1;311(3):e233117.","DOI":"10.1148\/radiol.233117"},{"key":"1659_CR16","unstructured":"OpenAI: GPT-4. Available at https:\/\/openai.com\/index\/gpt-4\/. Accessed 21 July 2025"},{"key":"1659_CR17","unstructured":"Anthropic: Claude 3.5. Available at https:\/\/www.anthropic.com\/claude\/sonnet. Accessed 21 July 2025"},{"key":"1659_CR18","doi-asserted-by":"crossref","unstructured":"Riedemann L, Labonne M, Gilbert S. The path forward for large language models in medicine is open. npj Digit Med. 2024 Nov 27;7(1):339.","DOI":"10.1038\/s41746-024-01344-w"},{"key":"1659_CR19","doi-asserted-by":"crossref","unstructured":"Wo\u017anicki P, Laqua C, Fiku I, Hekalo A, Truhn D, Engelhardt S, et al. Automatic structuring of radiology reports with on-premise open-source large language models. Eur Radiol 2024 Oct 10","DOI":"10.1007\/s00330-024-11074-y"},{"key":"1659_CR20","doi-asserted-by":"crossref","unstructured":"Mahapatra J, Garain U. Impact of model size on fine-tuned LLM performance in data-to-text generation: a state-of-the-art investigation. arXiv; 2024. Available at http:\/\/arxiv.org\/abs\/2407.14088. Accessed 21 July 2025","DOI":"10.2139\/ssrn.4916435"},{"key":"1659_CR21","unstructured":"Zhou Z, Ning X, Hong K, Fu T, Xu J, Li S, et al. A survey on efficient inference for large language models. arXiv; 2024. Available at http:\/\/arxiv.org\/abs\/2404.14294. Accessed 21 July 2025"},{"key":"1659_CR22","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Hastings J, Putora PM, Schmerder M, Cihoric N. Implementing large language models in healthcare while balancing control, collaboration, costs and security. npj Digit Med. 2025 Mar 6;8(1):143.","DOI":"10.1038\/s41746-025-01476-7"},{"key":"1659_CR23","unstructured":"NIH CDE repository. Available at https:\/\/cde.nlm.nih.gov\/home. Accessed 21 July 2025"},{"key":"1659_CR24","doi-asserted-by":"crossref","unstructured":"Snyder JM, Pawloski JA, Poisson LM. Developing real-world evidence-ready datasets: time for clinician engagement. Curr Oncol Rep. 2020 May;22(5):45.","DOI":"10.1007\/s11912-020-00904-z"},{"key":"1659_CR25","unstructured":"The SMARAGD project. Available at https:\/\/smaragd-switzerland.ch\/. Accessed 21 July 2025"},{"key":"1659_CR26","doi-asserted-by":"crossref","unstructured":"Renshaw AA, Mena-Allauca M, Gould EW, Sirintrapun SJ. Synoptic reporting: evidence-based review and future directions. JCO Clinical Cancer Informatics. 2018 Dec;(2):1\u20139.","DOI":"10.1200\/CCI.17.00088"},{"key":"1659_CR27","unstructured":"Common data elements (CDE) program. Available at https:\/\/heal.nih.gov\/data\/common-data-elements. Accessed 21 July 2025"},{"key":"1659_CR28","unstructured":"American College of Radiology, D\u2019Orsi CJ, Sickles EA, Mendelson EB, Morris EA, editors. ACR BI-RADS atlas: breast imaging reporting and data system\u202f; mammography, ultrasound, magnetic resonance imaging, follow-up and outcome monitoring, data dictionary. 5th edition. Reston, Va.: ACR, American College of Radiology; 2013."},{"key":"1659_CR29","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Putora PM, Heuser M, Vlaskou Badra E, Baumert BG, Leiser D, et al. Extraction of interoperable data from healthcare documents by identifying common data elements: an analysis of radiation therapy planning CT physician order entry records. Oncology. 2024;102(4):327\u201336.","DOI":"10.1159\/000534204"},{"key":"1659_CR30","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Schmalfuss M, Zink J, Hastings J, Gaio R, Schmerder M, et al. A CDE-based data structure for radiotherapeutic decision-making in breast cancer. BMC Med Inform Decis Mak. 2025 Jul 1.","DOI":"10.1186\/s12911-025-03036-1"},{"key":"1659_CR31","unstructured":"NIH: CDE forms. Available at https:\/\/cde.nlm.nih.gov\/form\/search. Accessed 21 July 2025"},{"key":"1659_CR32","doi-asserted-by":"crossref","unstructured":"Reichenpfader D, M\u00fcller H, Denecke K. A scoping review of large language model based approaches for information extraction from radiology reports. npj Digit Med. 2024 Aug 24;7(1):222.","DOI":"10.1038\/s41746-024-01219-0"},{"key":"1659_CR33","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Windisch P, Filchenko I, Zink J, Putora PM, Shaheen A, et al. Application of a general LLM-based classification system to retrieve information about oncological trials. Oncology. 2025 Jun 13;1\u201318.","DOI":"10.1101\/2024.12.03.24318390"},{"key":"1659_CR34","unstructured":"PyPI: general-classifier package. Available at https:\/\/pypi.org\/project\/general-classifier\/. Accessed 21 July 2025"},{"key":"1659_CR35","unstructured":"GitHub: low-code-LLM-CDE-extraction-mammography. Available at https:\/\/github.com\/Smart-Radiology-Goes-Digital-SMARAGD\/low-code-LLM-CDE-extraction-mammography. Accessed 21 July 2025"},{"key":"1659_CR36","unstructured":"PyTorch. Available at https:\/\/pytorch.org\/. Accessed 21 July 2025"},{"key":"1659_CR37","unstructured":"Hendrycks D, Burns C, Basart S, Zou A, Mazeika M, Song D, et al. Measuring massive multitask language understanding. arXiv; 2021. Available at http:\/\/arxiv.org\/abs\/2009.03300. Accessed 21 July 2025"},{"key":"1659_CR38","unstructured":"Zhou J, Lu T, Mishra S, Brahma S, Basu S, Luan Y, et al. Instruction-following evaluation for large language models. arXiv; 2023. Available at http:\/\/arxiv.org\/abs\/2311.07911. Accessed 21 July 2025"},{"key":"1659_CR39","unstructured":"Rein D, Hou BL, Stickland AC, Petty J, Pang RY, Dirani J, et al. GPQA: a graduate-level Google-proof Q&A benchmark. arXiv; 2023. Available at http:\/\/arxiv.org\/abs\/2311.12022. Accessed 21 July 2025"},{"key":"1659_CR40","unstructured":"Huggingface: Rombos-LLM-V2.6-Qwen-14b. Available at https:\/\/huggingface.co\/rombodawg\/Rombos-LLM-V2.6-Qwen-14b. Accessed 21 July 2025"},{"key":"1659_CR41","unstructured":"Huggingface: SOLAR Pro Preview Instruct. Available at https:\/\/huggingface.co\/upstage\/solar-pro-preview-instruct. Accessed 21 July 2025"},{"key":"1659_CR42","unstructured":"Abdin M, Aneja J, Behl H, Bubeck S, Eldan R, Gunasekar S, et al. Phi-4 technical report [Internet]. arXiv; 2024. Available at http:\/\/arxiv.org\/abs\/2412.08905. Accessed 21 July 2025"},{"key":"1659_CR43","unstructured":"Huggingface: Li-14B v0.4. Available at https:\/\/huggingface.co\/wanlige\/li-14b-v0.4. Accessed 21 July 2025"},{"key":"1659_CR44","unstructured":"Huggingface: OpenLLM leaderboard. Available at https:\/\/huggingface.co\/spaces\/open-llm-leaderboard\/open_llm_leaderboard#\/. Accessed 21 July 2025"},{"key":"1659_CR45","unstructured":"Huggingface: Lamarck-14B-v0.6Lamarck 14B v0.6. Available at https:\/\/huggingface.co\/sometimesanotion\/Lamarck-14B-v0.6Lamarck 14B v0.6. Accessed 21 July 2025"},{"key":"1659_CR46","unstructured":"Huggingface: Virtuoso-small. Available at https:\/\/huggingface.co\/arcee-ai\/Virtuoso-Small. Accessed 21 July 2025"},{"key":"1659_CR47","unstructured":"DeepSeek-AI, Guo D, Yang D, Zhang H, Song J, Zhang R, et al. DeepSeek-R1: incentivizing reasoning capability in LLMs via reinforcement learning. arXiv; 2025. Available at http:\/\/arxiv.org\/abs\/2501.12948. Accessed 21 July 2025"},{"key":"1659_CR48","doi-asserted-by":"crossref","unstructured":"Wang J, Meng F, Liang Y, Zhou J. DRT: deep reasoning translation via long chain-of-thought. arXiv; 2025. Available at http:\/\/arxiv.org\/abs\/2412.17498. Accessed 21 July 2025","DOI":"10.18653\/v1\/2025.findings-acl.351"},{"key":"1659_CR49","doi-asserted-by":"crossref","unstructured":"Wang L, Chen X, Deng X, Wen H, You M, Liu W, et al. Prompt engineering in consistency and reliability with the evidence-based guideline for LLMs. npj Digit Med. 2024 Feb 20;7(1):41.","DOI":"10.1038\/s41746-024-01029-4"},{"key":"1659_CR50","unstructured":"Promptingguide.ai. Available at https:\/\/www.promptingguide.ai\/de. Accessed 21 July 2025"},{"key":"1659_CR51","unstructured":"Brown TB, Mann B, Ryder N, Subbiah M, Kaplan J, Dhariwal P, et al. Language models are few-shot learners. arXiv; 2020. Available at http:\/\/arxiv.org\/abs\/2005.14165. Accessed 21 July 2025"},{"key":"1659_CR52","unstructured":"Wei J, Wang X, Schuurmans D, Bosma M, Ichter B, Xia F, et al. Chain-of-thought prompting elicits reasoning in large language models. arXiv; 2023. Available at http:\/\/arxiv.org\/abs\/2201.11903. Accessed 21 July 2025"},{"issue":"3","key":"1659_CR53","doi-asserted-by":"publisher","first-page":"276","DOI":"10.11613\/BM.2012.031","volume":"22","author":"ML McHugh","year":"2012","unstructured":"McHugh ML. Interrater reliability: the kappa statistic. Biochem Med (Zagreb). 2012;22(3):276\u201382.","journal-title":"Biochem Med (Zagreb)."},{"key":"1659_CR54","doi-asserted-by":"crossref","unstructured":"Sokolova M, Lapalme G. A systematic analysis of performance measures for classification tasks. Information Processing & Management. 2009 Jul;45(4):427\u201337.","DOI":"10.1016\/j.ipm.2009.03.002"},{"key":"1659_CR55","unstructured":"Dror R, Baumer G, Shlomov S, Reichart R. The Hitchhiker\u2019s guide to testing statistical significance in natural language processing. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Melbourne, Australia: Association for Computational Linguistics; Available at http:\/\/aclweb.org\/anthology\/P18-1128. Accessed 21 July 2025"},{"key":"1659_CR56","unstructured":"Pandas.pydata.org. Available at https:\/\/pandas.pydata.org\/. Accessed 21 July 2025"},{"key":"1659_CR57","unstructured":"NumPy.org. Available at https:\/\/numpy.org\/. Accessed 21 July 2025"},{"key":"1659_CR58","unstructured":"Scikit-Learn. Available at https:\/\/scikit-learn.org\/stable\/. Accessed 21 July 2025"},{"key":"1659_CR59","doi-asserted-by":"crossref","unstructured":"Armstrong RA. When to use the Bonferroni correction. Ophthalmic Physiologic Optic. 2014 Sep;34(5):502\u20138.","DOI":"10.1111\/opo.12131"},{"key":"1659_CR60","doi-asserted-by":"crossref","unstructured":"Keshavarz P, Bagherieh S, Nabipoorashrafi SA, Chalian H, Rahsepar AA, Kim GHJ, et al. ChatGPT in radiology: a systematic review of performance, pitfalls, and future perspectives. Diagnostic and Interventional Imaging. 2024 Jul;105(7\u20138):251\u201365.","DOI":"10.1016\/j.diii.2024.04.003"},{"key":"1659_CR61","doi-asserted-by":"crossref","unstructured":"Mallio CA, Bernetti C, Sertorio AC, Zobel BB. ChatGPT in radiology structured reporting: analysis of ChatGPT-3.5 Turbo and GPT-4 in reducing word count and recalling findings. Quant Imaging Med Surg. 2024 Feb 1;14(2):2096\u2013102.","DOI":"10.21037\/qims-23-1300"},{"key":"1659_CR62","doi-asserted-by":"crossref","unstructured":"Wiest IC, Ferber D, Zhu J, Van Treeck M, Meyer SK, Juglan R, et al. Privacy-preserving large language models for structured medical information retrieval. npj Digit Med. 2024 Sep 20;7(1):257.","DOI":"10.1038\/s41746-024-01233-2"},{"key":"1659_CR63","doi-asserted-by":"crossref","unstructured":"Le Guellec B, Lef\u00e8vre A, Geay C, Shorten L, Bruge C, Hacein-Bey L, et al. Performance of an open-source large language model in extracting information from free-text radiology reports. Radiol Artif Intell. 2024 Jul;6(4):e230364.","DOI":"10.1148\/ryai.230364"},{"key":"1659_CR64","unstructured":"Ding T, Chen T, Zhu H, Jiang J, Zhong Y, Zhou J, et al. The efficiency spectrum of large language models: an algorithmic survey. arXiv; 2024. Available at http:\/\/arxiv.org\/abs\/2312.00678. Accessed 21 July 2025"},{"key":"1659_CR65","doi-asserted-by":"crossref","unstructured":"Bozkurt S, Gimenez F, Burnside ES, Gulkesen KH, Rubin DL. Using automatically extracted information from mammography reports for decision-support. J Biomed Inform. 2016 Aug;62:224\u201331.","DOI":"10.1016\/j.jbi.2016.07.001"},{"key":"1659_CR66","doi-asserted-by":"crossref","unstructured":"Dehghani N, Saliba-Colombani V, Chick A, Heng M, Operto G, Fillard P. Efficient labeling of french mammogram reports with MammoBERT. Sci Rep. 2024 Oct 22 ;14(1). Available at https:\/\/www.nature.com\/articles\/s41598-024-76369-y. Accessed 21 July 2025","DOI":"10.1038\/s41598-024-76369-y"},{"key":"1659_CR67","unstructured":"Upadhayay B, Behzadan V, Karbasi A. Cognitive overload attack:prompt injection for long context. arXiv; 2024. Available from: http:\/\/arxiv.org\/abs\/2410.11272. Accessed 21 July 2025"},{"key":"1659_CR68","unstructured":"Nookala VPS, Verma G, Mukherjee S, Kumar S. Adversarial robustness of prompt-based few-shot learning for natural language understanding. arXiv; 2023. Available at http:\/\/arxiv.org\/abs\/2306.11066. Accessed 21 July 2025"},{"key":"1659_CR69","unstructured":"Long DX, Dinh D, Nguyen NH, Kawaguchi K, Chen NF, Joty S, et al. What makes a good natural language prompt?. arXiv; 2025. Available at http:\/\/arxiv.org\/abs\/2506.06950. Accessed 21 July 2025"},{"key":"1659_CR70","unstructured":"Jr EGS, Benjamin G, Araujo M, Santos H, Freitas D, Almeida E, et al. Which prompting technique should i use? An empirical investigation of prompting techniques for software engineering tasks [Internet]. arXiv; 2025. Available at http:\/\/arxiv.org\/abs\/2506.05614. Accessed 21 July 2025"},{"key":"1659_CR71","doi-asserted-by":"crossref","unstructured":"Bhayana R. Chatbots and large language models in radiology: a practical primer for clinical and research applications. Radiology. 2024 Jan 1;310(1):e232756.","DOI":"10.1148\/radiol.232756"},{"key":"1659_CR72","doi-asserted-by":"crossref","unstructured":"Rubin DL, Kahn CE. Common data elements in radiology. Radiology. 2017 Jun;283(3):837\u201344.","DOI":"10.1148\/radiol.2016161553"},{"key":"1659_CR73","doi-asserted-by":"publisher","unstructured":"Mehdiratta G, Duda JT, Elahi A, Borthakur A, Chatterjee N, Gee J, et al. Automated integration of AI results into radiology reports using common data elements. J Digit Imaging Inform med. 2025 Jan 27; Available at https:\/\/link.springer.com\/https:\/\/doi.org\/10.1007\/s10278-025-01414-9. Accessed 21 July 2025","DOI":"10.1007\/s10278-025-01414-9"},{"key":"1659_CR74","unstructured":"Radelement.org. Available at https:\/\/radelement.org\/home. Accessed 21 July 2025"},{"key":"1659_CR75","unstructured":"International Collaboration on Cancer Reporting (ICCR). Availableat https:\/\/www.iccr-cancer.org\/. Accessed 21 July 2025"},{"key":"1659_CR76","unstructured":"NCI\/NIH: semantics series\u2014the role of common data elements and artificial intelligence. Available at https:\/\/datascience.cancer.gov\/news-events\/blog\/semantics-series-role-common-data-elements-and-artificial-intelligence. Accessed 21 July 2025"},{"key":"1659_CR77","unstructured":"Wang Y, Huang J, He H, Zhang V, Zhou Y, Hao X, et al. CDEMapper: enhancing NIH common data element normalization using large language models. arXiv; 2024. Available at http:\/\/arxiv.org\/abs\/2412.00491. Accessed 21 July 2025"},{"key":"1659_CR78","doi-asserted-by":"crossref","unstructured":"Dennst\u00e4dt F, Putora PM, Cihoric N. (Common) data elements in radiation oncology: a systematic literature review. JCO Clinical Cancer Informatics. 2023 Jun;(7):e2300008.","DOI":"10.1200\/CCI.23.00008"},{"key":"1659_CR79","unstructured":"GitHub: FacebookResearch: LLM transparency tool. Available at https:\/\/github.com\/facebookresearch\/llm-transparency-tool. Accessed 21 July 2025"}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-025-01659-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-025-01659-4","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-025-01659-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T09:06:19Z","timestamp":1780391179000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10278-025-01659-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,10]]},"references-count":79,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["1659"],"URL":"https:\/\/doi.org\/10.1007\/s10278-025-01659-4","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,10]]},"assertion":[{"value":"7 April 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"31 July 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 August 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 September 2025","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Ethics approval for the study was granted by the Ethics Committee of the Canton of Bern (BASEC number 2022\u201301621), in alignment with the principles outlined in the Declaration of Helsinki.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics Approval"}},{"value":"The data utilized in this study comprised anonymized mammography reports. All patients provided informed consent for the use of their data for research purposes, in accordance with ethical standards.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to Participate"}},{"value":"Dr. Cihoric is a technical lead for the\n                      SmartOncology\n                      project and a medical advisor for Wemedoo AG, Steinhausen AG, Switzerland. The authors declare no other conflicts of interest.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}