{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T18:54:48Z","timestamp":1791831288610,"version":"4.3.4"},"posted":{"date-parts":[[2023,12,26]]},"group-title":"Neurology","reference-count":72,"publisher":"openRxiv","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2023,12,26]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                <jats:p>Neurodegenerative dementia syndromes, such as Primary Progressive Aphasias (PPA), have traditionally been diagnosed based in part on verbal and nonverbal cognitive profiles. Debate continues about whether PPA is best subdivided into three variants and also regarding the most distinctive linguistic features for classifying PPA variants. In this study, we harnessed the capabilities of artificial intelligence (AI) and natural language processing (NLP) to first perform unsupervised classification of concise, connected speech samples from 78 PPA patients. Large Language Models discerned three distinct PPA clusters, with 88.5% agreement with independent clinical diagnoses. Patterns of cortical atrophy of three data-driven clusters corresponded to the localization in the clinical diagnostic criteria. We then used NLP to identify linguistic features that best dissociate the three PPA variants. Seventeen features emerged as most valuable for this purpose, including the observation that separating verbs into high and low-frequency types significantly improves classification accuracy. Using these linguistic features derived from the analysis of brief connected speech samples, we developed a classifier that achieved 97.9% accuracy in predicting PPA subtypes and healthy controls. Our findings provide pivotal insights for refining early-stage dementia diagnosis, deepening our understanding of the characteristics of these neurodegenerative phenotypes and the neurobiology of language processing, and enhancing diagnostic evaluation accuracy.<\/jats:p>\n                <jats:sec>\n                  <jats:title>One sentence summary<\/jats:title>\n                  <jats:p>Computational linguistic analyses of naturalistic speech samples can classify the aphasic variant of patients similarly to expert clinicians and identify well-established and novel linguistic features crucial for classification.<\/jats:p>\n                <\/jats:sec>","DOI":"10.1101\/2023.12.22.23300470","type":"posted-content","created":{"date-parts":[[2023,12,26]],"date-time":"2023-12-26T13:05:10Z","timestamp":1703595910000},"source":"Crossref","is-referenced-by-count":4,"title":["Using Generative Artificial Intelligence to Classify Primary Progressive Aphasia from Connected Speech"],"prefix":"10.64898","author":[{"given":"Neguine","family":"Rezaii","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Megan","family":"Quimby","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bonnie","family":"Wong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daisy","family":"Hochberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Brickhouse","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexandra","family":"Touroutoglou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5958-3445","authenticated-orcid":false,"given":"Bradford C.","family":"Dickerson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Phillip","family":"Wolff","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2023122908000688000_2023.12.22.23300470v1.1","doi-asserted-by":"publisher","DOI":"10.1002\/ana.91"},{"key":"2023122908000688000_2023.12.22.23300470v1.2","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0b013e31821103e6"},{"key":"2023122908000688000_2023.12.22.23300470v1.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.neurol.2020.12.001"},{"key":"2023122908000688000_2023.12.22.23300470v1.4","doi-asserted-by":"publisher","DOI":"10.1038\/nrneurol.2014.159"},{"key":"2023122908000688000_2023.12.22.23300470v1.5","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0b013e3182574f79"},{"key":"2023122908000688000_2023.12.22.23300470v1.6","doi-asserted-by":"crossref","first-page":"1127","DOI":"10.1212\/WNL.0000000000000271","article-title":"Logopenic, mixed, or Alzheimer-related aphasia?","volume":"82","year":"2014","journal-title":"Neurology"},{"key":"2023122908000688000_2023.12.22.23300470v1.7","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0000000000000261"},{"key":"2023122908000688000_2023.12.22.23300470v1.8","doi-asserted-by":"crossref","first-page":"1832","DOI":"10.1212\/01.wnl.0000436070.28137.7b","article-title":"Classification and pathology of primary progressive aphasia","volume":"81","year":"2013","journal-title":"Neurology"},{"key":"2023122908000688000_2023.12.22.23300470v1.9","doi-asserted-by":"publisher","DOI":"10.1002\/ana.24885"},{"key":"2023122908000688000_2023.12.22.23300470v1.10","doi-asserted-by":"crossref","first-page":"3121","DOI":"10.1093\/brain\/awaa245","article-title":"Graded, multidimensional intra- and intergroup variations in primary progressive aphasia and post-stroke aphasia","volume":"143","year":"2020","journal-title":"Brain J. Neurol"},{"key":"2023122908000688000_2023.12.22.23300470v1.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.cortex.2012.12.006"},{"key":"2023122908000688000_2023.12.22.23300470v1.12","doi-asserted-by":"crossref","first-page":"1185","DOI":"10.3233\/JAD-201101","article-title":"Automatic Subtyping of Individuals with Primary Progressive Aphasia","volume":"79","year":"2021","journal-title":"J. Alzheimers Dis. JAD"},{"key":"2023122908000688000_2023.12.22.23300470v1.13","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.90.11.4957"},{"key":"2023122908000688000_2023.12.22.23300470v1.14","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.59.5.720"},{"key":"2023122908000688000_2023.12.22.23300470v1.15","doi-asserted-by":"publisher","DOI":"10.1080\/026432900410766"},{"key":"2023122908000688000_2023.12.22.23300470v1.16","doi-asserted-by":"publisher","DOI":"10.1016\/0028-3932(94)00066-2"},{"key":"2023122908000688000_2023.12.22.23300470v1.17","doi-asserted-by":"publisher","DOI":"10.1038\/349788a0"},{"key":"2023122908000688000_2023.12.22.23300470v1.18","doi-asserted-by":"publisher","DOI":"10.1126\/science.270.5233.102"},{"key":"2023122908000688000_2023.12.22.23300470v1.19","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1016\/j.cortex.2021.05.006","article-title":"Dissociating nouns and verbs in temporal and perisylvian networks: Evidence from neurodegenerative diseases","volume":"142","year":"2021","journal-title":"Cortex J. Devoted Study Nerv. Syst. Behav"},{"key":"2023122908000688000_2023.12.22.23300470v1.20","doi-asserted-by":"crossref","unstructured":"M. Montembeault , S. M. Brambati , M. L. Gorno-Tempini , R. Migliaccio , Clinical, Anatomical , and Pathological Features in the Three Variants of Primary Progressive Aphasia: A Review. Front. Neurol. 9 (2018).","DOI":"10.3389\/fneur.2018.00692"},{"key":"2023122908000688000_2023.12.22.23300470v1.21","doi-asserted-by":"publisher","DOI":"10.1002\/ana.10812"},{"key":"2023122908000688000_2023.12.22.23300470v1.22","doi-asserted-by":"publisher","DOI":"10.1016\/j.cogbrainres.2005.08.017"},{"key":"2023122908000688000_2023.12.22.23300470v1.23","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2006.04.179"},{"key":"2023122908000688000_2023.12.22.23300470v1.24","doi-asserted-by":"publisher","DOI":"10.1016\/j.bandl.2007.09.003"},{"key":"2023122908000688000_2023.12.22.23300470v1.25","doi-asserted-by":"publisher","DOI":"10.1162\/jocn.2007.19.9.1542"},{"key":"2023122908000688000_2023.12.22.23300470v1.26","doi-asserted-by":"crossref","first-page":"655","DOI":"10.1080\/02643290126377","article-title":"A neural basis for the retrieval of words for actions","volume":"18","year":"2001","journal-title":"Cogn. Neuropsychol"},{"key":"2023122908000688000_2023.12.22.23300470v1.27","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-444-62630-1.00010-X"},{"key":"2023122908000688000_2023.12.22.23300470v1.28","doi-asserted-by":"publisher","DOI":"10.1162\/08989290260138681"},{"key":"2023122908000688000_2023.12.22.23300470v1.29","first-page":"68","article-title":"Verb retrieval in aphasia","volume":"56","year":"1997","journal-title":"Characterizing single word impairments. Brain Lang"},{"key":"2023122908000688000_2023.12.22.23300470v1.30","doi-asserted-by":"publisher","DOI":"10.1006\/brln.1997.1923"},{"key":"2023122908000688000_2023.12.22.23300470v1.31","doi-asserted-by":"publisher","DOI":"10.1016\/S0093-934X(03)00147-0"},{"key":"2023122908000688000_2023.12.22.23300470v1.32","doi-asserted-by":"crossref","first-page":"266","DOI":"10.1016\/j.bandl.2005.11.002","article-title":"Semantic weight and verb retrieval in aphasia","volume":"97","year":"2006","journal-title":"Brain Lang"},{"key":"2023122908000688000_2023.12.22.23300470v1.33","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1016\/0911-6044(96)00006-1","article-title":"Cinderella revisited: A comparison of fluent and nonfluent aphasic speech","volume":"9","year":"1996","journal-title":"J. Neurolinguistics"},{"key":"2023122908000688000_2023.12.22.23300470v1.34","doi-asserted-by":"crossref","unstructured":"N. Rezaii , An information-theoretic analysis of agrammatism in Broca\u2019s aphasia. 2023.04.23.23288999 (2023).","DOI":"10.1101\/2023.04.23.23288999"},{"key":"2023122908000688000_2023.12.22.23300470v1.35","doi-asserted-by":"crossref","first-page":"647","DOI":"10.1002\/ana.26744","article-title":"Measuring Sentence Information via Surprisal: Theoretical and Clinical Implications in Nonfluent Aphasia","volume":"94","year":"2023","journal-title":"Ann. Neurol"},{"key":"2023122908000688000_2023.12.22.23300470v1.36","doi-asserted-by":"crossref","unstructured":"A. Kertesz , A. Kertesz , J. C. Raven , PsychCorp (Firm), WAB-R: Western Aphasia Battery-Revised (PsychCorp, 2007).","DOI":"10.1037\/t15168-000"},{"key":"2023122908000688000_2023.12.22.23300470v1.37","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2110.08207"},{"key":"2023122908000688000_2023.12.22.23300470v1.38","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1910.10683"},{"key":"2023122908000688000_2023.12.22.23300470v1.39","doi-asserted-by":"publisher","DOI":"10.2307\/2532178"},{"key":"2023122908000688000_2023.12.22.23300470v1.40","doi-asserted-by":"publisher","DOI":"10.1007\/BF02289263"},{"key":"2023122908000688000_2023.12.22.23300470v1.41","doi-asserted-by":"crossref","unstructured":"J. J. Godfrey , E. C. Holliman , J. McDaniel , SWITCHBOARD: telephone speech corpus for research and development in Proceedings of the 1992 IEEE International Conference on Acoustics, Speech and Signal Processing - Volume 1, ICASSP\u201992., (IEEE Computer Society, 1992), pp. 517\u2013520.","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"2023122908000688000_2023.12.22.23300470v1.42","unstructured":"J. W. Du Bois , Santa Barbara Corpus of Spoken American English Part I LDC2000S85 in Linguistic Data Consortium., (2000)."},{"key":"2023122908000688000_2023.12.22.23300470v1.43","unstructured":"M. Davies , The Corpus of Contemporary American English (COCA) (2008)."},{"key":"2023122908000688000_2023.12.22.23300470v1.44","unstructured":"F. Pedregosa , et al., Scikit-learn: Machine Learning in Python. Mach. Learn. PYTHON."},{"key":"2023122908000688000_2023.12.22.23300470v1.45","doi-asserted-by":"crossref","unstructured":"P. Qi , Y. Zhang , Y. Zhang , J. Bolton , C. D. Manning , Stanza: A Python Natural Language Processing Toolkit for Many Human Languages in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, (Association for Computational Linguistics, 2020), pp. 101\u2013108.","DOI":"10.18653\/v1\/2020.acl-demos.14"},{"key":"2023122908000688000_2023.12.22.23300470v1.46","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0188709"},{"key":"2023122908000688000_2023.12.22.23300470v1.47","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/73.3.751"},{"key":"2023122908000688000_2023.12.22.23300470v1.48","first-page":"2069","article-title":"Connected speech production in three variants of primary progressive aphasia","volume":"133","year":"2010","journal-title":"Brain Lond. Engl"},{"key":"2023122908000688000_2023.12.22.23300470v1.49","doi-asserted-by":"publisher","DOI":"10.1016\/j.jphysparis.2008.03.008"},{"key":"2023122908000688000_2023.12.22.23300470v1.50","doi-asserted-by":"publisher","DOI":"10.1016\/j.bandl.2008.09.010"},{"key":"2023122908000688000_2023.12.22.23300470v1.51","unstructured":"A. E. Goldberg , Constructions: A Construction Grammar Approach to Argument Structure (University of Chicago Press, 1995) (August 13, 2023)."},{"key":"2023122908000688000_2023.12.22.23300470v1.52","doi-asserted-by":"crossref","unstructured":"A. E. Goldberg , Constructions at Work: The Nature of Generalization in Language (Oxford University Press, 2006).","DOI":"10.1093\/acprof:oso\/9780199268511.001.0001"},{"key":"2023122908000688000_2023.12.22.23300470v1.53","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2011.09.014"},{"key":"2023122908000688000_2023.12.22.23300470v1.54","doi-asserted-by":"crossref","first-page":"1018","DOI":"10.1080\/02687038.2014.912744","article-title":"Grammatical Impairments in PPA","volume":"28","year":"2014","journal-title":"Aphasiology"},{"key":"2023122908000688000_2023.12.22.23300470v1.55","unstructured":"J. Grimshaw , \u201cSemantic structure and semantic content in lexical representation\u201d in (1993)."},{"key":"2023122908000688000_2023.12.22.23300470v1.56","unstructured":"A. Marantz , No escape from syntax: Don\u2019t try morphological analysis in the privacy of your own lexicon in (1997) (August 13, 2023)."},{"key":"2023122908000688000_2023.12.22.23300470v1.57","unstructured":"S. Pinker , Words and Rules: The Ingredients Of Language (Basic Books, 2015)."},{"key":"2023122908000688000_2023.12.22.23300470v1.58","unstructured":"B. Levin , English Verb Classes and Alternations (The University of Chicago Press, 1993) (July 16, 2021)."},{"key":"2023122908000688000_2023.12.22.23300470v1.59","unstructured":"R. S. Jackendoff , Semantic Structures (MIT Press, 1990)."},{"key":"2023122908000688000_2023.12.22.23300470v1.60","doi-asserted-by":"crossref","unstructured":"P. Indefrey , The Spatial and Temporal Signatures of Word Production Components: A Critical Update. Front. Psychol. 2 (2011).","DOI":"10.3389\/fpsyg.2011.00255"},{"key":"2023122908000688000_2023.12.22.23300470v1.61","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2002.06.001"},{"key":"2023122908000688000_2023.12.22.23300470v1.62","doi-asserted-by":"crossref","unstructured":"N. Rezaii , B. Ren , M. Quimby , D. Hochberg , B. C. Dickerson , Less is more in language production: an information-theoretic analysis of agrammatism in primary progressive aphasia. Brain Commun. 5, fcad136 (2023).","DOI":"10.1093\/braincomms\/fcad136"},{"key":"2023122908000688000_2023.12.22.23300470v1.63","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2120203119"},{"key":"2023122908000688000_2023.12.22.23300470v1.64","doi-asserted-by":"crossref","unstructured":"S. Josephy-Hernandez , et al., Automated analysis of written language in the three variants of primary progressive aphasia. Brain Commun. 5, fcad202 (2023).","DOI":"10.1093\/braincomms\/fcad202"},{"key":"2023122908000688000_2023.12.22.23300470v1.65","doi-asserted-by":"publisher","DOI":"10.1016\/S1364-6613(00)01463-7"},{"key":"2023122908000688000_2023.12.22.23300470v1.66","doi-asserted-by":"crossref","first-page":"43","DOI":"10.2217\/nmt.11.2","article-title":"Monitoring progression of primary progressive aphasia: current approaches and future directions","volume":"1","year":"2011","journal-title":"Neurodegener. Dis. Manag"},{"key":"2023122908000688000_2023.12.22.23300470v1.67","doi-asserted-by":"publisher","DOI":"10.1080\/02687038.2014.931563"},{"key":"2023122908000688000_2023.12.22.23300470v1.68","unstructured":"T. Wolf , et al., Transformers: State-of-the-Art Natural Language Processing in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations , (Association for Computational Linguistics, 2020), pp. 38\u201345."},{"key":"2023122908000688000_2023.12.22.23300470v1.69","doi-asserted-by":"publisher","DOI":"10.1109\/42.906426"},{"key":"2023122908000688000_2023.12.22.23300470v1.70","doi-asserted-by":"crossref","unstructured":"B. S. Everitt , The Analysis of Contingency Tables 1977 Edition, First Edition (Chapman and Hall, 1977).","DOI":"10.1007\/978-1-4899-2927-3"},{"key":"2023122908000688000_2023.12.22.23300470v1.71","doi-asserted-by":"crossref","unstructured":"A. Frick , A. Ludwig , H. Mehldau , A fast adaptive layout algorithm for undirected graphs (extended abstract and system demonstration) in Graph Drawing, Lecture Notes in Computer Science., R. Tamassia, I. G. Tollis, Eds. (Springer, 1995), pp. 388\u2013403.","DOI":"10.1007\/3-540-58950-3_393"},{"key":"2023122908000688000_2023.12.22.23300470v1.72","doi-asserted-by":"publisher","DOI":"10.2139\/ssrn.3378438"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2023.12.22.23300470","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T19:47:20Z","timestamp":1768506440000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2023.12.22.23300470"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,26]]},"references-count":72,"URL":"https:\/\/doi.org\/10.1101\/2023.12.22.23300470","relation":{},"subject":[],"published":{"date-parts":[[2023,12,26]]},"subtype":"preprint"}}