{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T18:54:49Z","timestamp":1791831289905,"version":"4.3.4"},"reference-count":68,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Alzheimer's disease and related dementias (ADRD) are progressive neurodegenerative conditions where early detection is critical for timely intervention and care planning. However, current diagnostic methods are often inaccessible, costly, and delayed, especially for underserved populations. There is a growing need for scalable, noninvasive tools that can support timely diagnosis. Spontaneous speech contains rich acoustic and linguistic markers that can serve as noninvasive behavioral markers for cognitive decline. Foundation models, pretrained on large-scale audio or text data, generate high-dimensional embeddings that encode rich contextual and acoustic information.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study benchmarks open-source foundation language and speech models to evaluate their effectiveness in detecting ADRD from spontaneous speech as a potential solution for early, noninvasive, and scalable ADRD detection.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this study, we used the Pioneering Research for Early Prediction of Alzheimer\u2019s and Related Dementias EUREKA (PREPARE) Challenge dataset, which consists of audio recordings from over 1600 participants with 3 distinct categories of cognitive decline: healthy control (HC), mild cognitive impairment (MCI), and Alzheimer's disease (AD). We further excluded samples that are non-English, nonspontaneous speech, or of poor quality. Our final samples included 703 (59.13%) HC, 81 (6.81%) MCI, and 405 (34.06%) AD cases. We systematically benchmarked 18 open-source foundation speech and language models to classify cognitive status into 3 categories (HC, MCI, or AD). Post hoc interpretability analysis was performed for the best-performing model using Shapley additive explanations linking high-dimensional embeddings with explainable acoustic and linguistic markers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Whisper-medium model achieved the highest performance among speech models at 0.731 accuracy and 0.802 area under the curve, while Bidirectional Encoder Representations from Transformers with pause annotation achieved the top accuracy of 0.662 and 0.744 area under the curve among language models. Overall, ADRD detection based on state-of-the-art automatic speech recognition model-generated audio-embeddings outperformed other models, and the inclusion of nonsemantic information, such as pause patterns, consistently improved the classification performance of text-embedding\u2013based models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our work presents a comprehensive comparative evaluation of state-of-the-art speech and language models for AD and MCI detection on a large, clinically relevant dataset. Embeddings derived from acoustic models, which capture both semantic and acoustic information, show promising performance and highlight the potential for developing a more scalable, noninvasive, and cost-effective early detection tool for ADRD.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/79411","type":"journal-article","created":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T12:20:09Z","timestamp":1766406009000},"page":"e79411","source":"Crossref","is-referenced-by-count":2,"title":["Early Detection of Alzheimer's Disease and Related Dementias From Spontaneous Speech Using Foundation Speech and Language Models: Comparative Evaluation"],"prefix":"10.2196","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8900-3528","authenticated-orcid":false,"given":"Jingyu","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9550-2393","authenticated-orcid":false,"given":"Lingchao","family":"Mao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6922-9135","authenticated-orcid":false,"given":"Xi","family":"Mao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3941-721X","authenticated-orcid":false,"given":"Hairong","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8782-0162","authenticated-orcid":false,"given":"Zhendong","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0634-0025","authenticated-orcid":false,"given":"Xuelei Sherry","family":"Ni","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,5,13]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2018.02.018"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1001\/archneurol.2009.266"},{"issue":"1","key":"ref3","doi-asserted-by":"crossref","first-page":"9","DOI":"10.1056\/NEJMoa2212948","volume":"388","author":"Van Dyck, CH","year":"2023","journal-title":"New Engl J Med"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2023.21109"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.14283\/jpad.2024.24"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.3233\/ADR230007"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.14283\/jpad.2023.131"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1111\/j.1600-0447.2008.01326.x"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1186\/s13195-023-01272-z"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1016\/j.trci.2019.09.018"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/s11065-008-9064-z"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.3389\/fnagi.2018.00369"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1002\/dad2.70082"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/icassp43922.2022.9747006"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2023-1734"},{"key":"ref16","unstructured":"RadfordAKimJWXuTBrockmanGRobust speech recognition via large-scale weak supervision2023International Conference on Machine LearningDec 6, 2022Seoul, South Korea"},{"key":"ref17","author":"Goodglass, H","year":"1983","journal-title":"Boston Diagnostic Aphasia Examination Booklet"},{"key":"ref18","unstructured":"EsterMKriegelHPSanderJA density-based algorithm for discovering clusters in large spatial databases with noiseAssociation for the Advancement of Artificial Intelligence2026-04-09https:\/\/cdn.aaai.org\/KDD\/1996\/KDD96-037.pdf?source=post_page"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.3233\/JAD-150520"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1159\/000356726"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1016\/j.dadm.2014.11.012"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1159\/000510820"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2020.100583"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1016\/j.jamda.2014.04.015"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1017\/s1355617714000873"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.5209\/rev_sjop.2012.v15.n2.38859"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2009-282"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2020.624137"},{"key":"ref29","doi-asserted-by":"crossref","first-page":"624488","DOI":"10.3389\/fcomp.2020.624488","volume":"2","author":"Yuan, J","year":"2021","journal-title":"Front Comput Sci"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-332"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/taffc.2015.2457417"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2013-56"},{"key":"ref33","unstructured":"PennebakerJWFrancisMEBoothRJLinguistic inquiry and word count: LIWCMahway: Lawrence Erlbaum Associates20012026-04-21http:\/\/downloads.liwc.net.s3.amazonaws.com\/LIWC2015_OperatorManual.pdf"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.25080\/majora-7b98e3ed-003"},{"key":"ref35","unstructured":"PREPARE challenge - phase 2: model arena (acoustic track)DrivenData2024-04-15https:\/\/www.drivendata.org\/competitions\/299\/competition-nih-alzheimers-acoustic-2\/"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1044\/2022_ajslp-22-00281"},{"key":"ref37","unstructured":"DementiaBank20252026-04-09https:\/\/talkbank.org\/dementia\/"},{"key":"ref38","unstructured":"DevlinJChangMWLeeKToutanovaKBERT: pre-training of deep bidirectional transformers for language understanding2019Proceedings of the 2019 Conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers)June 1, 2019Minneapolis, Minnesota"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/access.2022.3223444"},{"key":"ref40","unstructured":"LuzSHaiderFFuente GarciaSDLFrommDAlzheimer's dementia recognition through spontaneous speechFrontiers Media SA20212026-04-09https:\/\/www.frontiersin.org\/journals\/computer-science\/articles\/10.3389\/fcomp.2021.780169\/pdf"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2024.104598"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2020-2557"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/icassp43922.2022.9747055"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btz682"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/w19-1909"},{"key":"ref46","unstructured":"BaevskiAZhouYMohamedAAuliMWav2Vec 2.0: a framework for self-supervised learning of speech representationsAdvances in Neural Information Processing Systems2020NIPS'20: Proceedings of the 34th International Conference on Neural Information Processing SystemsDecember 6-12, 2020Virtual1244912460"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/taslp.2021.3122291"},{"key":"ref48","unstructured":"WangCWuYQianYKumataniKUnispeech: unified speech representation learning with labeled and unlabeled dataInternational Conference on Machine Learning2021Proceedings of the 38th International Conference on Machine Learning, PMLRJuly 18-24, 2021Virtual event"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/jstsp.2022.3188113"},{"key":"ref50","unstructured":"BaevskiAHsuWNXuQBabuAData2vec: A general framework for self-supervised learning in speech, vision and language2022Proceedings of the 39 th International Conference on Machine LearningJuly 17-23, 2022Baltimore, Maryland, USA"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/icassp.2015.7178964"},{"key":"ref52","unstructured":"LundbergSMLeeSIA unified approach to interpreting model predictions2017Advances in Neural Information Processing SystemsDecember 4-9, 2017California, USA"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/iscslp63861.2024.10800483"},{"key":"ref54","unstructured":"LittleMMcSharryPHunterESpielmanJSuitability of dysphonia measurements for telemonitoring of Parkinson's diseaseNature Precedings20082026-04-09https:\/\/www.nature.com\/articles\/npre.2008.2298.1.pdf"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/jstsp.2019.2955022"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2015-568"},{"key":"ref57","unstructured":"FraserKCRudziczFGrahamNRochonEAutomatic speech recognition in the diagnosis of primary progressive aphasia2013Proceedings of the Fourth Workshop on Speech and Language Processing for Assistive TechnologiesApril 8, 2026Grenoble, FranceAssociation for Computational Linguistics4754"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.3109\/02699206.2010.521612"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1002\/alz.13731"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1080\/13825585.2024.2315774"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalz.2015.02.003"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.14283\/jpad.2024.114"},{"key":"ref63","unstructured":"American perspectives on early detection of Alzheimer's disease in the era of treatmentAlzheimer's Association20252026-04-09https:\/\/www.alz.org\/getmedia\/3d226bf2-0690-48d0-98ac-d790384f4ec2\/alzheimers-facts-and-figures-special-report.pdf"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1002\/trc2.70102"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.3390\/brainsci13010028"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-90460-y"},{"key":"ref67","unstructured":"DementiaBank2026-04-10https:\/\/talkbank.org\/dementia\/DrivenData\/"},{"key":"ref68","unstructured":"DrivenData2026-04-10https:\/\/www.drivendata.org\/"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e79411","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:12:28Z","timestamp":1782843148000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e79411"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,13]]},"references-count":68,"URL":"https:\/\/doi.org\/10.2196\/79411","relation":{"has-preprint":[{"id-type":"doi","id":"10.2196\/preprints.79411","asserted-by":"object"}]},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,13]]}}}