CoChat
Sign Up
Cited by 76Dementia and Cognitive Impairment ResearchArtificial Intelligence in HealthcareNeurobiology of Language and Bilingualism

Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach

Open access
Amini, Samad, Hao, Boran, Zhang, Lifu, Song, Mengting, Gupta, Aman, Karjadi, Cody, Kolachalama, Vijaya B., Au, Rhoda +1 more·2023·Alzheimer's & Dementia
DOI 10.1002/alz.12721PMC PMC10148688
Open this paper in CoChat — free accountOpen PDF

Abstract

Introduction: Automated computational assessment of neuropsychological tests would enable widespread, cost‐effective screening for dementia. Methods: A novel natural language processing approach is developed and validated to identify different stages of dementia based on automated transcription of digital voice recordings of subjects’ neuropsychological tests conducted by the Framingham Heart Study (n = 1084). Transcribed sentences from the test were encoded into quantitative data and several models were trained and tested using these data and the participants’ demographic characteristics. Results: Average area under the curve (AUC) on the held‐out test data reached 92.6%, 88.0%, and 74.4% for differentiating Normal cognition from Dementia, Normal or Mild Cognitive Impairment (MCI) from Dementia, and Normal from MCI, respectively. Discussion: The proposed approach offers a fully automated identification of MCI and dementia based on a recorded neuropsychological test, providing an opportunity to develop a remote screening tool that could be adapted easily to any language.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1002/alz.12721

Cites

6 of 6 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Speech features have been used to automatically differentiate people with amnestic AD presentation from healthy controls.

    “Indeed, speech has been studied as an objective tool for automatically differentiating people with amnestic AD presentation from healthy controls (HC).”

    From Automatic classification of AD pathology in FTD phenotypes using natural speech (2024) · cites this paper as [16]

    The Methods and Results (pages not provided) describe automated analysis of voice recordings that differentiated participants with dementia from those with normal cognition. Participants were classified as normal, having MCI, or having dementia; the paper does not identify the dementia group specifically as people with an amnestic AD presentation. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • The same study distinguished individuals with mild cognitive impairment (MCI) from HC with 71.2% accuracy.

    “A more recent study used a large dataset and a natural language processing (NLP) approach to differentiate individuals with dementia from HC with 87.1% accuracy and to distinguish individuals with mild cognitive impairment (MCI) from HC with 71.2% accuracy.”

    From Automatic classification of AD pathology in FTD phenotypes using natural speech (2024) · cites this paper as [16]

    The Methods identify a normal-versus-MCI classification task, so the study did distinguish MCI from healthy controls. The Results report 74.4% AUC for that task, not 71.2% classification accuracy; the passages found do not substantiate the claimed accuracy figure. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • A recent study using a natural language processing (NLP) approach differentiated individuals with dementia from HC with 87.1% accuracy.

    “A more recent study used a large dataset and a natural language processing (NLP) approach to differentiate individuals with dementia from HC with 87.1% accuracy and to distinguish individuals with mild cognitive impairment (MCI) from HC with 71.2% accuracy.”

    From Automatic classification of AD pathology in FTD phenotypes using natural speech (2024) · cites this paper as [16]

    The Methods describe an NLP model for distinguishing normal cognition from dementia, so the study and comparison are correctly identified. The Results report 92.6% AUC for that comparison, not 87.1% accuracy; the 87.1% figure is an AUC for a different comparison (normal cognition or MCI versus dementia). No retrieved passage establishes the cited 71.2% accuracy for MCI versus healthy controls. Page numbers were not available in our copy. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 41

  1. [2]Scheltens P, Blennow K, Breteler M, Alzheimer’s disease. Lancet. 2016; 388: 505– 517. 26921134
  2. [3]Turner RS, Stubbs T, Davies DA, Albensi BC. Potential new approaches for diagnosis of Alzheimer’s disease and related dementias. Front Neurol. 2020; 11: 496. 32582013 PMC7290039
  3. [4]Thomas JA, Burkhardt HA, Chaudhry S, Assessing the utility of language and voice biomarkers to predict cognitive impairment in the Framingham heart study cognitive aging cohort data, J Alzheimers Dis. 2020; 76(3): 905– 922. 32568190
  4. [5]Filiou RP, Bier N, Slegers A, Houze B, Belchior P, Brambati SM. Connected speech assessment in the early detection of Alzheimer’s disease and mild cognitive impairment: a scoping review. Aphasiology. 2020; 34(6): 723– 755.
  5. [6]Downer B, Fardo DW, Schmitt FA. A summary score for the Framingham heart study neuropsychological battery. J Aging Health. 2015; 27(7): 1199– 1222. 25804903 PMC4603385
  6. [7]Ashendorf L, Jefferson AL, Green RC, Stern RA. Test–retest stability on the WRAT-3 reading subtest in geriatric cognitive evaluations. J Clin Exp Neuropsychol. 2009; 31(5): 605– 610. 18821160 PMC2728046
  7. [8]Stuck D, Signorini A, Alhanai T, Novel digital voice biomarkers of dementia from the Framingham study. Alzheimers Deme. 2018; 14(7): P778– P779.
  8. [9]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. 28321196 PMC5337522
  9. [10]Hernandez-Domınguez L, Ratte S, Sierra-Martınez G, Roche-Bergua A. Computer-based evaluation of Alzheimer’s disease and mild cognitive impairment patients during a picture description task, Alzheimers Dement (Amst). 2018; 10: 260– 268. 29780871 PMC5956933
  10. [11]Liu L, Zhao S, Chen H, Wang A. A new machine learning method for identifying Alzheimer’s disease. Simul Model Pract Theory. 2020; 99: 102023.
  11. [12]Pulido MLB, Hernandez JBA, Ballester MAF, Gonzalez CMT, Mekyska J, Smekal Z. Alzheimer’s disease and automatic speech analysis: a review. Expert Syst Appl. 2020; 150: 113213.
  12. [13]Berisha V, Wang S, LaCross A, Liss J. Tracking discourse complexity preceding alzheimer’s disease diagnosis: a case study comparing the press conferences of presidents Ronald Reagan and George Herbert Walker Bush. J Alzheimers Dis. 2015; 45(3): 959– 963. 25633673 PMC6922000
  13. [14]Konig A, Satt A, Sorin A, Automatic speech analysis for the assessment of patients with predementia and Alzheimer’s disease. Alzheimers Dement (Amst). 2015; 1(1): 112– 124. 27239498 PMC4876915
  14. [15]Fraser KC, Meltzer JA, Rudzicz F. Linguistic features identify Alzheimer’s disease in narrative speech. J Alzheimers Dis. 2016; 49(2): 407– 422. 26484921
  15. [16]Asgari M, Kaye J, Dodge H. Predicting mild cognitive impairment from spontaneous spoken utterances. Alzheimers Dement (N Y). 2017; 3(2): 219– 228. 29067328 PMC5651423
  16. [17]Lin H, Karjadi C, Ang TFT, Identification of digital voice biomarkers for cognitive health. Explor Med. 2020; 1: 406. 33665648 PMC7929495
  17. [20]Iroju OG, Olaleke JO. A systematic review of natural language processing in healthcare. Int J Comput Sci Inf Technol Res. 2015; 7(8): 44– 50.
  18. [21]Srivastava SK, Singh SK, Suri JS. A healthcare text classification system and its performance evaluation: a source of better intelligence by characterizing healthcare text. In: Cognitive Informatics, Computer Modelling, and Cognitive Science. Elsevier; 2020: 319– 369. 29654417
  19. [22]Murdoch TB, Detsky AS. The inevitable application of big data to health care. JAMA. 2013; 309(13): 1351– 1352. 23549579
  20. [23]Hao B, Zhu H, Paschalidis IC. Enhancing clinical BERT embedding using a biomedical knowledge base, In: Proceedings of the 28th International Conference on Computational Linguistics, 2020; 657– 661.
  21. [25]Au R, Piers RJ, Devine S. How technology is reshaping cognitive assessment: lessons from the Framingham heart study. Neuropsychology. 2017; 31(8): 846– 861. 29376667 PMC5791752
  22. [26]Jak AJ, Preis SR, Beiser AS, Neuropsychological criteria for mild cognitive impairment and dementia risk in the Framingham heart study. J Int Neuropsychol Soc. 2016; 22(9): 937– 943. 27029348 PMC5045758
  23. [30]Kepuska V, Bohouta G. Comparing speech recognition systems (microsoft api, google api and cmu sphinx). Int J Eng Res Appl. 2017; 7(03): 20– 24.
  24. [31]Lan Z, Chen M, Goodman S, Gimpel K, Sharma P, Soricut R. ALBERT: a lite bert for self-supervised learning of language representations. arXiv preprint arXiv:1909. 11942.
  25. [32]Devlin J, Chang M-W, Lee K, Toutanova K. BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. · arXiv
  26. [35]Rios A, Kavuluru R. Neural transfer learning for assigning diagnosis codes to EMRs. Artif Intell Med. 2019; 96: 116– 122. 31164204
  27. [36]Lehr M, Prudhommeaux E, Shafran I, Roark B. Fully automated neuropsychological assessment for detecting mild cognitive impairment. In: Thirteenth Annual Conference of the International Speech Communication Association, 2012.
  28. [37]Vaswani A, Shazeer N, Parmar N, Attention is all you need. arXiv preprint arXiv:1706.03762. · arXiv
  29. [38]Cer D, Yang Y, Kong S-Y, Universal sentence encoder. arXiv preprint arXiv:1803.11175. · arXiv
  30. [39]Hao B, Sotudian S, Wang T, Early prediction of level-of-care requirements in patients with COVID-19. Elife. 2020; 9: e60519. 33044170 PMC7595731
  31. [40]Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982; 143(1): 29– 36. 7063747
  32. [41]Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975; 12(3): 189– 198. 1202204
  33. [42]Yang Y, Cer D, Ahmad A,, Multilingual universal sentence encoder for semantic retrieval. arXiv preprint arXiv:1907.04307. · arXiv
  34. [43]Conneau A, Khandelwal K, Goyal N, Unsupervised cross-lingual representation learning at scale. arXiv preprint arXiv:1911.02116. · arXiv
  35. [44]Deshpande G, Viraraghavan VS, Duggirala M, Patel S. Detecting emotional valence using time- domain analysis of speech signals. In: 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE; 2019: 3605– 3608. 31946657
  36. [45]Kopru B, Erzin E. Multimodal continuous emotion recognition using deep multi-task learning with correlation loss. arXiv preprint arXiv:2011.00876. · arXiv
  37. [46]Xue C, Karjadi C, Paschalidis IC, Au R, Kolachalama VB. Detection of dementia on voice recordings using deep learning: a framingham heart study. Alzheimers Res Ther. 2021; 13(1): 146. 34465384 PMC8409004
  38. [47]Berisha V, Krantsevich C, Hahn PR, Digital medicine and the curse of dimensionality. NPJ Digit Med. 2021; 4(1): 1– 8. 34711924 PMC8553745
  39. [48]Uzuner O, South BR, Shen S, DuVall SL. 2010 i2b2/VA challenge on concepts, assertions, and relations in clinical text. J Am Med Inform Assoc. 2011; 18(5): 552– 556. 21685143 PMC3168320
  40. [49]Blodgett SL, Barocas S, Daume III H, Wallach H. Language (technology) is power: a critical survey of “bias” in NLP. arXiv preprint arXiv:2005.14050 · arXiv
  41. [50]Liang PP, Li IM, Zheng E, Lim YC, Salakhutdinov R, Morency L-P. Towards debiasing sentence representations. arXiv preprint arXiv:2007.08100. · arXiv

Cited by

Sorted by citations

Authors
Amini, S., Hao, B., Zhang, L., Song, M., Gupta, A., Karjadi, C., Kolachalama, V. B., Au, R., & Paschalidis, I. C.
Affiliations
Boston University (US)
Corresponding author
Paschalidis, Ioannis Ch.
ORCID
Venue
Alzheimer's & Dementia
Publisher
Wiley
Published
2022-07-07
Volume
19
Issue
3
Pages
946-955
ISSN
  • 1552-5260 (ISSN-L)
  • 1552-5279
DOI
10.1002/alz.12721
PMC
PMC10148688
OpenAlex
W4284667718
Cited by
76
References
6
Citations by year
2026: 16 · 2025: 27 · 2024: 19 · 2023: 11 · 2022: 3
FWCI
9.94
Type
journal article
Language
en
Open access
Licence
Locations
  • Alzheimer s & Dementia (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
  • OpenBU (Boston University) (submittedVersion)
  • OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
Funders
  • National Science Foundation (DMS‐1664644, CNS‐1645681, IIS‐1914792)
  • Office of Naval Research (N00014‐19‐1‐2571)
  • U.S. Department of Energy (DE‐AR‐0001282, DE‐EE0009696)
  • National Institutes of Health (R01 GM135930, UL54 TR004130, R01‐HL159620, R21‐CA253498)
  • National Institute on Aging (AG008122, AG016495, AG033040, AG054156, AG049810, AG062109, AG068753)
  • Alzheimer's Association (AARG‐NTF‐20‐643020)
  • Boston University
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neurobiology of Language and Bilingualism
Keywords
mild cognitive impairment, dementia
MeSH terms
Aged, Aged, 80 and over, Dementia, Female, Humans, Male, Natural Language Processing, Neuropsychological Tests, Voice, Cognitive Dysfunction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review