{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T08:17:39Z","timestamp":1790669859645,"version":"4.1.0"},"reference-count":37,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T00:00:00Z","timestamp":1783641600000},"content-version":"vor","delay-in-days":3,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001652","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001652","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Inform Decis Mak"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Analyzing semi-spontaneous speech is a promising direction for supporting Alzheimer\u2019s disease (AD) assessment, yet progress is limited by the scarcity of annotated clinical data. Large Language Models (LLMs) offer new opportunities to generate synthetic narratives that may resemble speech patterns of both patients with AD and healthy controls during cognitive evaluation tasks such as the Cookie Theft Picture description.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study evaluates whether models including GPT, T5\/Flan-T5, LLaMA, Mistral, and Qwen can generate clinically plausible picture-description narratives under two configurations: Human-to-Bot, where an LLM responds directly to real interviewer prompts, and Bot-to-Bot, where two LLMs simulate both interviewer and participant roles. Models were fine-tuned on transcripts from the DementiaBank Pitt Corpus and assessed using lexical and semantic metrics, as well as human expert ratings. Generated narratives were further used to augment training data for an AD vs. healthy control classifier based on BERT embeddings and an MLP architecture.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>LLMs differed substantially in their ability to reproduce clinically meaningful and semantically coherent narratives of patient-interviewer interactions. Mistral, LLaMA, and Qwen achieved the strongest automatic evaluation metrics, e.g., BERTScores above 0.90 in the Human-to-Bot condition\u2014and produced narratives rated by human experts as fluent, plausible, and diagnostically informative. When combining real and synthetic narratives for classifier training, the highest F1-score reached 0.84, outperforming models trained on real data alone (F1\u2009=\u20090.74). Synthetic data generated in Human-to-Bot settings contributed most to diagnostic improvements, whereas Bot-to-Bot interactions exhibited greater variability and reduced clinical realism.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>LLMs can generate high-quality synthetic narratives that enhance downstream AD classification and show promising clinical plausibility in cognitive assessment contexts. Incorporating LLM-generated data provides a scalable strategy for mitigating data scarcity in dementia research. Future work should focus on improving fully synthetic dialogue quality, expanding multilingual capabilities, and refining evaluation frameworks to better capture clinically relevant linguistic features.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12911-026-03653-4","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T11:25:45Z","timestamp":1783423545000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Generating Alzheimer\u2019s narratives using large language models"],"prefix":"10.1186","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2727-2116","authenticated-orcid":false,"given":"Paula Andrea","family":"Perez-Toro","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mahmoud","family":"Almizel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Elmar","family":"N\u00f6th","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andreas","family":"Maier","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tomas","family":"Arias-Vergara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"issue":"1","key":"3653_CR1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s44385-024-00004-8","volume":"1","author":"WU Hasan","year":"2024","unstructured":"Hasan WU, Zaman KT, Wang X, Li J, Xie B, Tao C. Empowering Alzheimer\u2019s caregivers with conversational AI: a novel approach for enhanced communication and personalized support. npj Biomed Innov. 2024;1(1):1\u201310. https:\/\/doi.org\/10.1038\/s44385-024-00004-8.","journal-title":"npj Biomed Innov"},{"issue":"5","key":"3653_CR2","doi-asserted-by":"publisher","first-page":"79","DOI":"10.9781\/ijimai.2021.02.010","volume":"6","author":"PR Sb","year":"2021","unstructured":"Sb PR, Agnihotri M, Jayagopi DB. Improving asynchronous interview interaction with follow-up question generation. IJIMAI. 2021;6(5):79\u201389. https:\/\/doi.org\/10.9781\/ijimai.2021.02.010.","journal-title":"IJIMAI"},{"key":"3653_CR3","doi-asserted-by":"crossref","unstructured":"Liao Y, Meng Y, Wang Y, Liu H, Wang Y, Wang Y. Automatic interactive evaluation for large language models with state aware patient simulator. arXiv preprint arXiv:2403.08495. 2024.","DOI":"10.2139\/ssrn.4890649"},{"key":"3653_CR4","doi-asserted-by":"publisher","first-page":"105401","DOI":"10.1016\/j.ebiom.2024.105401","volume":"109","author":"X Du","year":"2024","unstructured":"Du X, Novoa-Laurentiev J, Plasek JM, Chuang Y-W, Wang L, Marshall GA, et al. Enhancing early detection of cognitive decline in the elderly: a comparative study utilizing large language models in clinical notes. EBioMedicine. 2024;109:105401. https:\/\/doi.org\/10.1016\/j.ebiom.2024.105401.","journal-title":"EBioMedicine"},{"issue":"12","key":"3653_CR5","doi-asserted-by":"publisher","first-page":"7531","DOI":"10.1109\/JBHI.2024.3435085","volume":"28","author":"Y Wu","year":"2024","unstructured":"Wu Y, Mao K, Zhang Y, Chen J. Callm: enhancing clinical interview analysis through data augmentation with large language models. IEEE J Biomed Health Inf. 2024;28(12):7531\u201342. https:\/\/doi.org\/10.1109\/JBHI.2024.3435085.","journal-title":"IEEE J Biomed Health Inf"},{"issue":"1","key":"3653_CR6","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1038\/s41572-021-00269-y","volume":"7","author":"DS Knopman","year":"2021","unstructured":"Knopman DS, Amieva H, Petersen RC, Ch\u00e9telat G, Holtzman DM, Hyman BT, et al. Alzheimer disease. Nat Rev Dis Primers. 2021;7(1):33. https:\/\/doi.org\/10.1038\/s41572-021-00269-y.","journal-title":"Nat Rev Dis Primers"},{"key":"3653_CR7","doi-asserted-by":"publisher","first-page":"269","DOI":"10.3389\/fpsyg.2017.00269","volume":"8","author":"V Boschi","year":"2017","unstructured":"Boschi V, Catricala E, Consonni M, Chesi C, Moro A, Cappa SF. Connected speech in neurodegenerative language disorders: a review. Front Psychol. 2017;8:269. https:\/\/doi.org\/10.3389\/fpsyg.2017.00269.","journal-title":"Front Psychol"},{"key":"3653_CR8","volume-title":"Cookie Theft picture. Boston diagnostic aphasia examination","author":"H Goodglass","year":"1983","unstructured":"Goodglass H, et al. Cookie Theft picture. Boston diagnostic aphasia examination. Philadelphia, PA: Lea & Febiger; 1983."},{"key":"3653_CR9","volume-title":"BDAE: the Boston diagnostic aphasia examination","author":"H Goodglass","year":"2001","unstructured":"Goodglass H, Kaplan E, Weintraub S. BDAE: the Boston diagnostic aphasia examination. Philadelphia, PA: Lippincott Williams & Wilkins; 2001."},{"key":"3653_CR10","doi-asserted-by":"crossref","unstructured":"Feng Y, Xu X, Zhuang Y, Zhang M. Large language models improve Alzheimer\u2019s disease diagnosis using multi-modality data. 2023 IEEE International Conference on Medical Artificial Intelligence (MedAI). IEEE; 2023, pp. 61\u201366.","DOI":"10.1109\/MedAI59581.2023.00016"},{"issue":"12","key":"3653_CR11","doi-asserted-by":"publisher","first-page":"0000168","DOI":"10.1371\/journal.pdig.0000168","volume":"1","author":"F Agbavor","year":"2022","unstructured":"Agbavor F, Liang H. Predicting dementia from spontaneous speech using large language models. PLoS Digit Health. 2022;1(12):0000168. https:\/\/doi.org\/10.1371\/journal.pdig.0000168.","journal-title":"PLoS Digit Health"},{"key":"3653_CR12","doi-asserted-by":"crossref","unstructured":"Cummins N, Pan Y, Ren Z, Fritsch J, Nallanthighal VS, Christensen H, et al. A comparison of acoustic and linguistics methodologies for alzheimer\u2019s dementia recognition. In: Interspeech 2020. ISCA-International Speech Communication Association; 2020. pp. 2182\u201386.","DOI":"10.21437\/Interspeech.2020-2635"},{"key":"3653_CR13","doi-asserted-by":"crossref","unstructured":"Koo J, Lee JH, Pyo J, Jo Y, Lee K. Exploiting multi-modal features from pre-trained networks for alzheimer\u2019s dementia recognition. Interspeech. 2020;2217\u201321.","DOI":"10.21437\/Interspeech.2020-3153"},{"key":"3653_CR14","doi-asserted-by":"crossref","unstructured":"Farzana S, Parde N. Exploring mmse score prediction using verbal and non-verbal cues. 2020;2207\u201311.","DOI":"10.21437\/Interspeech.2020-3085"},{"key":"3653_CR15","doi-asserted-by":"crossref","unstructured":"P\u00e9rez-Toro PA, Bayerl SP, Arias-Vergara T, V\u00e1squez-Correa JC, Klumpp P, Schuster M, et al. Influence of the interviewer on the automatic assessment of Alzheimer\u2019s disease in the context of the ADReSSo challenge. Interspeech. 2021;3785\u201389.","DOI":"10.21437\/Interspeech.2021-1589"},{"key":"3653_CR16","doi-asserted-by":"publisher","unstructured":"P\u00e9rez-Toro PA, Klumpp P, Hernandez A, Arias T, Lillo P, Slachevsky A, et al. Alzheimer\u2019s detection from English to Spanish using acoustic and linguistic embeddings. In Interspeech 2022. 2022. pp. 2483\u201387. https:\/\/doi.org\/10.21437\/Interspeech.2022-10883.","DOI":"10.21437\/Interspeech.2022-10883"},{"key":"3653_CR17","doi-asserted-by":"crossref","unstructured":"Peled-Cohen L, Reichart R. A systematic review of nlp for dementia-tasks, datasets and opportunities. arXiv preprint arXiv:2409.19737. 2024.","DOI":"10.1162\/TACL.a.35"},{"issue":"6","key":"3653_CR18","doi-asserted-by":"publisher","first-page":"585","DOI":"10.1001\/archneur.1994.00540180063015","volume":"51","author":"JT Becker","year":"1994","unstructured":"Becker JT, Boiler F, Lopez OL, Saxton J, McGonigle KL. The natural history of Alzheimer\u2019s disease: description of study cohort and accuracy of diagnosis. Arch Neurol. 1994;51(6):585\u201394. https:\/\/doi.org\/10.1001\/archneur.1994.00540180063015.","journal-title":"Arch Neurol"},{"key":"3653_CR19","doi-asserted-by":"publisher","DOI":"10.3389\/fcomp.2021.780169","volume-title":"Alzheimer\u2019s dementia recognition through spontaneous speech","author":"S Luz","year":"2021","unstructured":"Luz S, Haider F, Fuente Garcia S, Fromm D, MacWhinney B. Alzheimer\u2019s dementia recognition through spontaneous speech. Frontiers Media SA; 2021."},{"key":"3653_CR20","doi-asserted-by":"publisher","unstructured":"Balagopalan A, Eyre B, Rudzicz F, Novikova J. To bert or not to bert: comparing speech and language-based approaches for alzheimer\u2019s disease detection. In: Interspeech 2020. 2020. pp. 2167\u201371. https:\/\/doi.org\/10.21437\/Interspeech.2020-2557.","DOI":"10.21437\/Interspeech.2020-2557"},{"key":"3653_CR21","doi-asserted-by":"crossref","unstructured":"Devlin J, Chang M, Lee K, Toutanova K. BERT: pre-training of deep bidirectional transformers for language understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Minneapolis, Minnesota: Association for Computational Linguistics; 2019, pp. 4171\u201386. https:\/\/doi.org\/10.18653\/v1\/N19-1423. https:\/\/www.aclweb.org\/anthology\/N19-1423.","DOI":"10.18653\/v1\/N19-1423"},{"key":"3653_CR22","doi-asserted-by":"crossref","unstructured":"Li R, Wang X, Yu H. Two directions for clinical data generation with large language models: data-to-label and label-to-data. Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing. 2023, p. 7129, vol. 2023.","DOI":"10.18653\/v1\/2023.findings-emnlp.474"},{"key":"3653_CR23","doi-asserted-by":"crossref","unstructured":"Yan LK, Niu Q, Li M, Zhang Y, Yin CH, Fei C, et al. Large language model benchmarks in medical tasks. arXiv preprint arXiv:2410.21348. 2024.","DOI":"10.31219\/osf.io\/8j7d3"},{"key":"3653_CR24","doi-asserted-by":"crossref","unstructured":"Chen Z, Deng J, Zhou J, Wu J, Qian T, Huang M. Depression detection in clinical interviews with LLM-empowered structural element graph. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024, pp. 8174\u201387.","DOI":"10.18653\/v1\/2024.naacl-long.452"},{"key":"3653_CR25","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. Adv Neural Inf Process Syst. 2017."},{"issue":"140","key":"3653_CR26","first-page":"1","volume":"21","author":"C Raffel","year":"2020","unstructured":"Raffel C, Shazeer N, Roberts A, Lee K, Narang S, Matena M, et al. Exploring the limits of transfer learning with a unified text-to-text transformer. J Mach Learn Res. 2020;21(140):1\u201367.","journal-title":"J Mach Learn Res"},{"issue":"219","key":"3653_CR27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1561\/2200000050","volume":"25","author":"M Welandawe","year":"2024","unstructured":"Welandawe M, Andersen MR, Vehtari A, Huggins JH, et al. A framework for improving the reliability of black-box variational inference. J Mach Learn Res. 2024;25(219):1\u201371. https:\/\/doi.org\/10.1561\/2200000050.","journal-title":"J Mach Learn Res"},{"issue":"8","key":"3653_CR28","first-page":"9","volume":"1","author":"A Radford","year":"2019","unstructured":"Radford A, Wu J, Child R, Luan D, Amodei D, Sutskever I, et al. Language models are unsupervised multitask learners. OpenAI Blog. 2019;1(8):9.","journal-title":"OpenAI Blog"},{"key":"3653_CR29","unstructured":"Touvron H, Lavril T, Izacard G, Martinet X, Lachaux M-A, Lacroix T, et al. LLaMA: open and efficient foundation language models. arXiv e-prints, 2302. 2023."},{"key":"3653_CR30","unstructured":"Jiang AQ, Sablayrolles A, Mensch A, Bamford C, Chaplot DS, Casas DDL, et al.: Mistral 7B. 2023. arXiv preprint arXiv:2310.06825."},{"key":"3653_CR31","unstructured":"Yang A, Li A, Yang B, Zhang B, Hui B, Zheng B, et al. Qwen3 technical report. arXiv preprint arXiv:2505.09388. 2025."},{"key":"3653_CR32","unstructured":"Hu EJ, Shen Y, Wallis P, Allen-Zhu Z, Li Y, Wang S, et al. LoRA: low-rank adaptation of large language models. International Conference on Learning Representations. 2022. https:\/\/openreview.net\/forum?id=nZeVKeeFYf9."},{"key":"3653_CR33","unstructured":"Lavie A. Evaluating the output of machine translation systems. Proceedings of the 9th Conference of the Association for Machine Translation in the Americas: Tutorials. 2010."},{"key":"3653_CR34","unstructured":"Zhang T, Kishore V, Wu F, Weinberger KQ, Artzi Y. Bertscore: evaluating text generation with bert. International Conference on Learning Representations. 2020."},{"key":"3653_CR35","unstructured":"Springer JM, Goyal S, Wen K, Kumar T, Yue X, Malladi S, et al. Overtrained language models are harder to fine-tune. International Conference on Machine Learning. PMLR; 2025, pp. 56719\u201389."},{"key":"3653_CR36","doi-asserted-by":"crossref","unstructured":"Steed R, Panda S, Kobren A, Wick M. Upstream mitigation is not all you need: testing the bias transfer hypothesis in pre-trained language models. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022, pp. 3524\u201342.","DOI":"10.18653\/v1\/2022.acl-long.247"},{"key":"3653_CR37","unstructured":"Carlini N, Ippolito D, Jagielski M, Lee K, Tramer F, Zhang C. Quantifying memorization across neural language models. The Eleventh International Conference on Learning Representations. 2022."}],"container-title":["BMC Medical Informatics and Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12911-026-03653-4","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-026-03653-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-026-03653-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T07:14:37Z","timestamp":1783667677000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s12911-026-03653-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,7]]},"references-count":37,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["3653"],"URL":"https:\/\/doi.org\/10.1186\/s12911-026-03653-4","relation":{},"ISSN":["1472-6947"],"issn-type":[{"value":"1472-6947","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,7]]},"assertion":[{"value":"13 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Data collection procedures for the English dataset (the publicly available Pitt Corpus) were approved by the Institutional Review Board (IRB) of the University of Pittsburgh under the Alzheimer\u2019s and Related Dementias Study (Protocol IRG-88\u2013061). All participants (or their legal representatives) provided informed consent in accordance with this protocol. The project was supported by grants from the National Institutes of Health (NIH), specifically AG005133 and AG03705, awarded to the University of Pittsburgh Alzheimer\u2019s Disease Research Center. For subsequent reuse and broader research studies, the dataset is also governed by the Carnegie Mellon University IRB approval (ID STUDY2022_00000172). More information about the data can be found at\n                      \n                      . The DementiaBank Pitt Corpus was used under its academic license terms, and access was granted by TalkBank.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"258"}}