Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review
Open accessThe DOI matches its Crossref record; 125 of the 127 references it cites were found in a registry or an index; 5 of the 167 sources its claims cite were checked for support.
Abstract
Background: Language is a valuable source of clinical information in Alzheimer’s disease, as it declines concurrently with neurodegeneration. Consequently, speech and language data have been extensively studied in connection with its diagnosis. Objective: Firstly, to summarize the existing findings on the use of artificial intelligence, speech, and language processing to predict cognitive decline in the context of Alzheimer’s disease. Secondly, to detail current research procedures, highlight their limitations, and suggest strategies to address them. Methods: Systematic review of original research between 2000 and 2019, registered in PROSPERO (reference CRD42018116606). An interdisciplinary search covered six databases on engineering (ACM and IEEE), psychology (PsycINFO), medicine (PubMed and Embase), and Web of Science. Bibliographies of relevant papers were screened until December 2019. Results: From 3,654 search results, 51 articles were selected against the eligibility criteria. Four tables summarize their findings: study details (aim, population, interventions, comparisons, methods, and outcomes), data details (size, type, modalities, annotation, balance, availability, and language of study), methodology (pre-processing, feature generation, machine learning, evaluation, and results), and clinical applicability (research implications, clinical potential, risk of bias, and strengths/limitations). Conclusion: Promising results are reported across nearly all 51 studies, but very few have been implemented in clinical research or practice. The main limitations of the field are poor standardization, limited comparability of results, and a degree of disconnect between study aims and clinical applications. Active attempts to close these gaps will support translation of future research into clinical practice.
Overview
CoChat Summary
## TL;DR
This systematic review synthesized 51 studies on AI, speech, and language processing for Alzheimer’s disease (AD) monitoring from 2000–2019 and found promising performance but very limited clinical translation. The authors argue the field is held back by poor standardization, small and imbalanced datasets, and weak comparability across studies.
## Key findings
- The search yielded 3,654 records in total, and 51 articles met eligibility criteria for the review.
- Across the reviewed literature, 41 of 51 papers focused on distinguishing cognitive impairment categories, while only 7 attempted three-way or four-way classification, 2 examined longitudinal cognitive change, and 1 used cluster analysis to describe discourse patterns.
- The review reports that 31% of studies trained on fewer than 50 participants, only 27% had 100 or more participants, and just 6 studies had 100+ participants/speech samples in each experimental group; all six used the Pitt Corpus.
- Reported performance ranged widely: some classifiers were 50% or lower, while others exceeded 90% accuracy; examples highlighted include 60% to 93.79% accuracy in one small, imbalanced study, 78.7% accuracy in a larger balanced dataset, 85.4% accuracy on the same dataset using text features, and 86.6% as the state-of-the-art accuracy for dialogue data.
## Methods
- Systematic review of original research from 2000–2019, registered in PROSPERO (CRD42018116606).
- Searched six databases: ACM, IEEE, PsycINFO, PubMed, Embase, and Web of Science, plus bibliographies and citation tracking through December 2019.
- Two reviewers independently screened records and extracted data using four tables covering study details, data details, methodology, and clinical applicability; synthesis was narrative, with no meta-analysis.
## Limitations
- The paper’s stated limitations of the field are poor standardization, limited comparability of results, and a disconnect between study aims and clinical applications.
- The review itself notes that most studies use relatively small datasets, class imbalance is common, and many studies lack held-out testing or external validation, which limits generalizability; this is stated in the paper and is also evident from the reporting.
- A further limitation evident from the text is that many studies report incomplete pre-processing and feature-generation details, making replication difficult; the authors explicitly say these procedures are often not fully described.
Model-written summary of a shortened version of the paper’s full text (it is longer than we read). Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
5 Valid · 2 Not valid · 70 Not checked · 122 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Out of 51 reviewed papers, only seven did attempt three-way [50, 62–64] or four-way [12, 65, 66] classification.”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Study aim and design (Supplementary Table 4: SPICMO)
What the sources say
- Not checked yetcites [50] Mirzaei, 2018, Two-stage feature selection of voice parameters for early Alzheimer’s disease prediction, IRBM source
- Not checked yetcites [62] Egas López, 2019, Assessing Alzheimer’s disease from speech using the i-vector approach, International Conference on Speech and Computer source
- Not checked yetcites [63] Gosztolya, 2019, Identifying mild cognitive impairment and mild Alzheimer’s disease based on spontaneous speech using ASR and linguistic features, Comput Speech Lang source
- Not checked yetcites [64] Kato, 2013, Early detection of cognitive impairment in the elderly based on Bayesian mining using speech prosody and cerebral blood flow activation, Conference proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference
- Not checked yetcites [12] Mirheidari, 2019, Computational Cognitive Assessment: Investigating the Use of an Intelligent Virtual Agent for the Detection of Early Signs of Dementia, 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Supportedcites [65] Nasrolahzadeh, 2018, Higher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease, Cogn Neurodyn source
The source supports its use as an example of four-way classification: Table 8 (Conclusion) lists its proposed method as having four classes, and the Experimental results and discussion describe discrimination among four groups. The count of 51 reviewed papers and seven studies is the citing authors’ review finding, not a count attributed to this individual study. (LLM, conf=0.92)
- Not checked yetcites [66] Thomas, 2005, Automatic detection and rating of dementia of Alzheimer type through lexical analysis of spontaneous speech, IEEE International Conference Mechatronics and Automation, 2005 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Outside of those, seven papers are concerned with three or four disease stages [12, 50, 62–66], two explore longitudinal cognitive changes (although still through binary classification) [67, 68], and one describes discourse patterns [69].”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Study aim and design (Supplementary Table 4: SPICMO)
What the sources say
- Not checked yetcites [12] Mirheidari, 2019, Computational Cognitive Assessment: Investigating the Use of an Intelligent Virtual Agent for the Detection of Early Signs of Dementia, 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Not checked yetcites [50] Mirzaei, 2018, Two-stage feature selection of voice parameters for early Alzheimer’s disease prediction, IRBM source
- Not checked yetcites [62] Egas López, 2019, Assessing Alzheimer’s disease from speech using the i-vector approach, International Conference on Speech and Computer source
- Not checked yetcites [63] Gosztolya, 2019, Identifying mild cognitive impairment and mild Alzheimer’s disease based on spontaneous speech using ASR and linguistic features, Comput Speech Lang source
- Not checked yetcites [64] Kato, 2013, Early detection of cognitive impairment in the elderly based on Bayesian mining using speech prosody and cerebral blood flow activation, Conference proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference
- Supportedcites [65] Nasrolahzadeh, 2018, Higher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease, Cogn Neurodyn source
The Abstract of Nasrolahzadeh et al. (2018) states that four classifiers were used to classify subjects into three different levels of AD plus a healthy group, and the Data collection section documents the three AD levels (First, Second, Third Stage; FS=6, SS=15, TS=9) alongside 30 healthy controls, with Table 1 giving their MMSE/CDR characteristics. The Results (p. on Table 7a) likewise report per-stage sensitivity, specificity and accuracy for FS, SS and TS. This matches the citing review's characterization of the paper as one concerned with three or four disease stages. (LLM, conf=0.90)
- Not checked yetcites [66] Thomas, 2005, Automatic detection and rating of dementia of Alzheimer type through lexical analysis of spontaneous speech, IEEE International Conference Mechatronics and Automation, 2005 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“However, only [75, 90] share the exact specification of the subset of Pitt Corpus used for their analysis, in order for other researchers to be able to replicate their findings”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Data availability (Supplementary Table 5: Data Details)
What the sources say
- Supportedcites [75] Orimaye, 2017, Predicting probable Alzheimer’s disease using linguistic deficits and biomarkers, BMC Bioinformatics source
The Methods (Datasets and Participants; page unavailable) identify the University of Pittsburgh DementiaBank dataset and specify the analyzed subset: the first 99 probable-AD patients, 99 healthy controls, and transcripts from each participant’s last visit. This supports citing the paper as an example of a study that specifies its subset; the comparison with other studies is the citing authors’ assessment. (LLM, conf=0.89)
- Not checked yetcites [90] Dos Santos, 2017, Enriching complex networks with word embeddings for detecting mild cognitive impairment from speech transcripts, ACL 2017 – 55th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Only two papers rely on truly spontaneous and natural monologues, prompted with an open question instead of a picture description [60, 65].”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Data type (Supplementary Table 5: Data Details)
What the sources say
- Not checked yetcites [60] Martinez de Lizarduy U , Calvo Salomon P , Gomez Vilda P , Ecay Torres M , Lopez de Ipina K (2017) ALZUMERIC: A decision support system for diagnosis and monitoring of cognitive impairment. Loquens 4, https://doi.org/10.3989/loquens.2017.037. source
- Partly supportedcites [65] Nasrolahzadeh, 2018, Higher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease, Cogn Neurodyn sourceExperimental — accuracy not yet measured
The Data collection section describes recordings of spontaneous speech in a relaxed setting: participants were asked to converse, tell personal stories, and express their feelings. It does not say they responded to an open question or produced monologues. The 'only two papers' count is the citing review’s own characterization, not a finding established by this source. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Several other studies choose leave-pair-out CV instead [7, 70, 73, 75, 78, 97], since it produces unbiased estimates for AUC and also reduces potential size bias.”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Evaluation techniques (Supplementary Table 6: Methodology)
What the sources say
- Not checked yetcites [7] Fraser, 2019, Predicting MCI Status From Multimodal Language Data Using Cascaded Classifiers, Front Aging Neurosci source
- Not checked yetcites [70] Prud’hommeaux, 2015, Graph-based word alignment for clinical language evaluation, Comput Linguist source
- Not checked yetcites [73] Prud’Hommeaux, 2011, Alignment of spoken narratives for automated neuropsychological assessment, 2011 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2011 source
- Supportedcites [75] Orimaye, 2017, Predicting probable Alzheimer’s disease using linguistic deficits and biomarkers, BMC Bioinformatics source
In “Evaluations with ML algorithm,” the paper says it used leave-pair-out cross-validation to estimate AUC. The “Diagnostic task” section also states that its experiments used LPOCV, confirming that this is a study that chose the method. (LLM, conf=0.94)
- Not checked yetcites [78] Roark, 2011, Spoken language derived measures for detecting mild cognitive impairment, N Engl J Med
- Not checked yetcites [97] Yu, 2018, Speech-based automated cognitive impairment detection from remotely-collected cognitive test audio, IEEE Access source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
108 more claims
Create a free account to see all 199 claims and to verify the 162 references still open.
- It may state two separate claims as one. 57
- Probably not a substantial claim. 39
- The quote only partly states this claim. 1
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?
The cited source says what the claim attributes to it.
In the citing paper
“In a systematic review of the use of AI in assessing speech and language skills to predict cognitive decline in Alzheimer's disease, promising results were reported in almost all 51 studies”
From Artificial intelligence in assessment and intervention of speech and language disorders: A literature review (2025) · cites this paper as [32]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Initiatives like the Alzheimer’s disease imaging initiative (ADNI) have established best practices for acquisition and preprocessing, while large-scale studies such as the UK Biobank have advanced standardization for diverse MRI techniques.71–74”
From Review of Artificial Intelligence for Clinical Use in Alzheimer's Disease and Related Dementias (2026) · cites this paper as [72]
Experimental — accuracy not yet measuredIn the citing paper
“Also, as shown by systematic reviews, speech and language studies across neurodegenerative diseases span fewer than 20 languages, most of them tested in only a handful of papers. 4,5,47,81-83”
From Speech and language markers of neurodegeneration: a call for global equity (2023) · cites this paper as [47]
In the citing paper
“ASLA offers ob jective, examiner-independent, multidimensional results via brief oral production tasks, through measures of the acoustic speech sig nal (e.g. speech timing, pitch variability) and/or its transcription (e.g. syntactic complexity, semantic specificity). 47,48”
From Speech and language markers of neurodegeneration: a call for global equity (2023) · cites this paper as [47]
In the citing paper
“This language is also predominant in Alzheimer's disease research, accounting for 69% of word retrieval studies 83 and over 40% of ASLA reports 47”
From Speech and language markers of neurodegeneration: a call for global equity (2023) · cites this paper as [47]
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Data type (Supplementary Table 5: Data Details)
“Dialogue data are present less frequently, in 27% of the studies”
§ Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer’s Disease: A Systematic Review > DISCUSSION > Strengths/Limitations (Supplementary Table 7: Clinical applicability)
“Of the reviewed papers, only 18% report the use of dialogue (i.e., structured, semi-structured, or conversational).”
Citations
Cites
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975) · Journal of Psychiatric Research · doi:10.1016/0022-3956(75)90026-6 · Cited by 88,282 · cited 18×
“However, Mirheidari et al. [12] reported 62% accuracy and 0.815 AUC for a 4-way classifier in a slightly more balanced dataset and Thomas et al. [66], also 4-way, only 50%, on four groups of MMSE scores.”
MethodInfluentialPreferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & for the PRISMA Group (2009) · BMJ · doi:10.1136/bmj.b2535 · Cited by 83,823
Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & the PRISMA Group* (2009) · Annals of Internal Medicine · doi:10.7326/0003-4819-151-4-200908180-00135 · Cited by 37,554
(2003) · Journal of Machine Learning Research · doi:10.5555/944919.944937 · Cited by 32,337
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., & Stadlan, E. M. (1984) · Neurology · doi:10.1212/wnl.34.7.939 · Cited by 27,915 · cited 1×
“Based on the results of these tests, clinicians usually proceed to diagnose MCI, following Petersen’s criteria [38], or AD, following NINCDS-ADRDA criteria [39].”
MethodEfficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568
The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment
Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., Cummings, J. L., & Chertkow, H. (2005) · Journal of the American Geriatrics Society · doi:10.1111/j.1532-5415.2005.53221.x · Cited by 25,553 · cited 1×
“The most commonly used batteries are the Mini-Mental State Examination (MMSE) [27], the Montreal Cognitive Assessment (MoCA) [28], the Hierarchical Dementia ScaleRevised (HDS-R) [29], the Clinical Dementia Rating (CDR) [30], the Clock Drawing Test (CDT) [31], the Alzheimer’s disease Assessment Scale, Cognitive part (A…”
MethodQUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies
Whiting, P. F., Rutjes, A. W. S., Westwood, M. E., Mallett, S., Deeks, J. J., Reitsma, J. B., Leeflang, M. M. G., Sterne, J. A. C., Bossuyt, P. M. M., & the QUADAS-2 Group* (2011) · Annals of Internal Medicine · doi:10.7326/0003-4819-155-8-201110180-00009 · Cited by 14,199
The Clinical Dementia Rating (CDR): Current version and scoring rules
Morris, J. C. (1993) · Neurology · doi:10.1212/wnl.43.11.2412-a · Cited by 9,882
The Psychological Meaning of Words: LIWC and Computerized Text Analysis Methods
Tausczik, Y. R., & Pennebaker, J. W. (2010) · Journal of Language and Social Psychology · doi:10.1177/0261927x09351676 · Cited by 5,774 · cited 1×
“(NLP) Linguistic Inquiry and Word Count LIWC [21] Lexical diversity Type-Token Ratio (TTR), Moving Average TTR (MATTR), Simpson’s Diversity Index (SDI) Brunét’s Index (BI), Honoré’s Statistic (HS).”
BackgroundDavis, S., & Mermelstein, P. (1980) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1980.1163420 · Cited by 5,465 · cited 1×
“Mel Frequency Cepstral Coefficients MFCCs [26].”
A new rating scale for Alzheimer's disease
(1984) · American Journal of Psychiatry · doi:10.1176/ajp.141.11.1356 · Cited by 4,175 · cited 1×
“The most commonly used batteries are the Mini-Mental State Examination (MMSE) [27], the Montreal Cognitive Assessment (MoCA) [28], the Hierarchical Dementia ScaleRevised (HDS-R) [29], the Clinical Dementia Rating (CDR) [30], the Clock Drawing Test (CDT) [31], the Alzheimer’s disease Assessment Scale, Cognitive part (A…”
MethodAssessing Risk of Bias in Included Studies
Higgins, J. P., & Altman, D. G. (2008) · Cochrane Handbook for Systematic Reviews of Interventions · doi:10.1002/9780470712184.ch8 · Cited by 3,396 · cited 1×
“Additionally, the Cochrane Collaboration recently emphasized a preference for systematic reviews to focus on the performance of individual papers’ on the different risk of bias criteria [18].”
BackgroundThe CHILDES project: tools for analyzing talk
MacWhinney, B. (1992) · Child Language Teaching and Therapy · doi:10.1177/026565909200800211 · Cited by 3,396
PROBAST: A Tool to Assess the Risk of Bias and Applicability of Prediction Model Studies
Wolff, R. F., Moons, K. G. M., Riley, R. D., Whiting, P. F., Westwood, M., Collins, G. S., Reitsma, J. B., Kleijnen, J., Mallett, S., & for the PROBAST Group† (2019) · Annals of Internal Medicine · doi:10.7326/m18-1376 · Cited by 2,984 · cited 1×
“Initially, we considered two tools for risk of bias assessment, namely the “QUADAS-2: Quality Assessment of Diagnosis Studies checklist - 2” [16] and the “PROBAST: Prediction model Risk Of Bias ASsessment Tool” [17].”
The well-built clinical question: A key to evidence-based decisions
(1995) · ACP Journal Club · doi:10.7326/acpjc-1995-123-3-a12 · Cited by 2,931 · cited 1×
“Our original intention was to rely on the PICO framework [14] for data collection.”
MethodToward a mechanistic psychology of dialogue.
Pickering, M. J., & Garrod, S. (2004) · Behavioral and Brain Sciences · doi:10.1017/s0140525x04000056 · Cited by 2,678 · cited 1×
“, we spend part of our day interacting with somebody else), as well as more representative of a broader range psycholinguistic aspects such as alignment and entrainment at different linguistic levels [121], which might be relevant to AD detection.”
BackgroundOpensmile: the munich versatile and fast open-source audio feature extractor
Eyben, F., Wöllmer, M., & Schuller, B. (2010) · Proceedings of the 18th ACM international conference on Multimedia · doi:10.1145/1873951.1874246 · Cited by 2,642
The Geneva Minimalistic Acoustic Parameter Set (GeMAPS) for Voice Research and Affective Computing
Eyben, F., Scherer, K. R., Schuller, B. W., Sundberg, J., Andre, E., Busso, C., Devillers, L. Y., Epps, J., Laukka, P., Narayanan, S. S., et al. (2016) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2015.2457417 · Cited by 1,818 · cited 1×
“With regards to methodology, we recommend a wider use of standardized feature sets, such as eGeMAPS [122], purposefully developed to detect physiological changes in voice production.”
Methodten Have, P. (2007) · SAGE Publications Ltd eBooks · doi:10.4135/9781849208895 · Cited by 1,453
Motor Speech Disorders: A Cross-Language Perspective
(2014) · Channel View Publications eBooks · doi:10.2307/jj.26931941 · Cited by 1,260
Systematic Reviews of Diagnostic Test Accuracy
Leeflang, M. M. G., Deeks, J. J., Gatsonis, C., & Bossuyt, P. M. M. (2008) · Annals of Internal Medicine · doi:10.7326/0003-4819-149-12-200812160-00008 · Cited by 1,124 · cited 1×
“The poor reporting of results and subsequent interpretation difficulties is a longstanding challenge of diagnostic test accuracy research [15].”
Clinical epidemiology of Alzheimer’s disease: assessing sex and gender differences
Mielke, M., Vemuri, P., & Rocca, W. (2014) · Clinical Epidemiology · doi:10.2147/clep.s37929 · Cited by 1,021 · cited 2×
“Demographic variables are established risk factors for AD [81], therefore demographics reporting is essential for this type of study.”
BackgroundAging, Memory, and Mild Cognitive Impairment
Petersen, R. C., Kokmen, E., Waring, S. C., Smith, G. E., Ivnik, R. J., & Tangelos, E. G. (1997) · International Psychogeriatrics · doi:10.1017/s1041610297004717 · Cited by 888
Linguistic Features Identify Alzheimer’s Disease in Narrative Speech
Fraser, K. C., Meltzer, J. A., & Rudzicz, F. (2015) · Journal of Alzheimer's Disease · doi:10.3233/jad-150520 · Cited by 797 · cited 8×
“There are only 6 studies where not only the dataset, but also each experimental group contains 100 or more participants/speech samples [6, 9, 11, 85, 92, 95].”
BackgroundMethodInfluentialClinical Dementia Rating (CDR)
(2001) · Management of Dementia · doi:10.3109/9780203213896-34 · Cited by 696
Discourse on the Move: Using corpus analysis to describe discourse structure
Biber, D., Connor, U., & Upton, T. A. (2007) · Studies in Corpus Linguistics · doi:10.1075/scl.28 · Cited by 601
Clock Drawing: A Neuropsychological Analysis
Freedman, M., Leach, L. R., Kaplan, E., Winocur, G., Shulman, K. I., & Delis, D. C. (1994) · Cited by 590
The Natural History of Alzheimer's Disease: Description of Study Cohort and Accuracy of Diagnosis
Becker, J. T. (1994) · Archives of Neurology · doi:10.1001/archneur.1994.00540180063015 · Cited by 576
Boston Diagnostic Aphasia Examination
(2012) · SpringerReference · doi:10.1007/springerreference_183639 · Cited by 527
A model and an hypothesis for language structure
Yngve, V. H. (1960) · Cited by 522
The GPCOG: A New Screening Test for Dementia Designed for General Practice
Brodaty, H., Pond, D., Kemp, N. M., Luscombe, G., Harding, L., Berman, K., & Huppert, F. A. (2002) · Journal of the American Geriatrics Society · doi:10.1046/j.1532-5415.2002.50122.x · Cited by 493 · cited 1×
“The most commonly used batteries are the Mini-Mental State Examination (MMSE) [27], the Montreal Cognitive Assessment (MoCA) [28], the Hierarchical Dementia ScaleRevised (HDS-R) [29], the Clinical Dementia Rating (CDR) [30], the Clock Drawing Test (CDT) [31], the Alzheimer’s disease Assessment Scale, Cognitive part (A…”
MethodAutomatic speech analysis for the assessment of patients with predementia and Alzheimer's disease
König, A., Satt, A., Sorin, A., Hoory, R., Toledo‐Ronen, O., Derreumaux, A., Manera, V., Verhey, F., Aalten, P., Robert, P. H., et al. (2015) · Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring · doi:10.1016/j.dadm.2014.11.012 · Cited by 380 · cited 3×
“, [61, 102]), which summarize the rates of false alarms and false negatives, are reported in less than half of the reviewed studies.”
BackgroundDeveloping and adopting safe and effective digital biomarkers to improve patient outcomes
Coravos, A., Khozin, S., & Mandl, K. D. (2019) · npj Digital Medicine · doi:10.1038/s41746-019-0090-4 · Cited by 361 · cited 1×
“In addition, novel AI/machine learning digital biomarkers [124] could be used in combination with established biomarkers to target populations at risk of later dementia onset, as has already been proposed [101].”
BackgroundSpoken Language Derived Measures for Detecting Mild Cognitive Impairment
Roark, B., Mitchell, M., Hosom, J.-P., Hollingshead, K., & Kaye, J. (2011) · IEEE Transactions on Audio, Speech, and Language Processing · doi:10.1109/tasl.2011.2112351 · Cited by 355 · cited 5×
“[78] and Khodabakhsh and Demiroǧlu [84].”
BackgroundInfluentialRiley, K. P., Snowdon, D. A., Desrosiers, M. F., & Markesbery, W. R. (2005) · Neurobiology of Aging · doi:10.1016/j.neurobiolaging.2004.06.019 · Cited by 303 · cited 1×
“High early idea density seems to be a lower risk predictor for developing AD later in life, whereas lower idea density appears associated with brain atrophy [106].”
BackgroundGlobal Deterioration Scale (GDS).
Reisberg, ., Ferris, S. H., de Leon, M. J., & Crook, T. (1988) · PubMed · Cited by 296
Toth, L., Hoffmann, I., Gosztolya, G., Vincze, V., Szatloczki, G., Banreti, Z., Pakaski, M., & Kalman, J. (2018) · Current Alzheimer Research · doi:10.2174/1567205014666171121114930 · Cited by 290 · cited 8×
“While this is based on word accuracy, some of the customized ASR systems are phone-based [63, 99] and seem to work better with speech generated by participants with AD.”
InfluentialAlzheimer’s Dementia Recognition Through Spontaneous Speech: The ADReSS Challenge
Luz, S., Haider, F., Fuente, S. D. L., Fromm, D., & MacWhinney, B. (2020) · Interspeech 2020 · doi:10.21437/interspeech.2020-2571 · Cited by 274
Self-citationShould Health Care Demand Interpretable Artificial Intelligence or Accept “Black Box” Medicine?
Wang, F., Kaushal, R., & Khullar, D. (2020) · Annals of Internal Medicine · doi:10.7326/m19-2548 · Cited by 244 · cited 1×
“While the accepted opinion is that the clinicians’ ability to be able to interpret an AI model is essential for the adoption of AI technologies in medicine, the issue is still the subject of lively debate, with influential machine learning researchers like Geoff Hinton arguing that “clinicians and regulators should no…”
BackgroundBucks, R. S., Singh, S., Cuerden, J. M., & Wilcock, G. K. (2000) · Aphasiology · doi:10.1080/026870300401603 · Cited by 233 · cited 1×
“, [4, 5]) to more recent work on computational speech technology (e.”
BackgroundThe Lawton Instrumental Activities of Daily Living (IADL) Scale.
Graf, C. L. (2008) · PubMed · Cited by 228
Katz Index of Independence in Activities of Daily Living (ADL).
Wallace, M. E., & Shelkey, M. (2007) · PubMed · Cited by 225
Gosztolya, G., Vincze, V., Tóth, L., Pákáski, M., Kálmán, J., & Hoffmann, I. (2019) · Computer Speech & Language · doi:10.1016/j.csl.2018.07.007 · Cited by 196 · cited 10×
“While this is based on word accuracy, some of the customized ASR systems are phone-based [63, 99] and seem to work better with speech generated by participants with AD.”
BackgroundMethodInfluentialHaider, F., de la Fuente, S., & Luz, S. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2019.2955022 · Cited by 190 · cited 10×
“, [89]), and especially if a standardized available feature set is used [11].”
BackgroundMethodInfluentialSelf-citationDifferential sensitivity to pitch distance, particularly in speech
’t Hart, J. (1981) · The Journal of the Acoustical Society of America · doi:10.1121/1.385592 · Cited by 178 · cited 1×
“[89] report F0 and functionals in semitones, because research argues that using semitones to express F0 reduces gender differences [109], which is corroborated by the choice of semitones in the standardized eGeMAPS [11].”
MethodFrazier, L. (1985) · Natural Language Parsing · doi:10.1017/cbo9780511597855.005 · Cited by 169
Predicting probable Alzheimer's disease using linguistic deficits and biomarkers
Orimaye, S. O., Wong, J. S.-M., Golden, K. J., Wong, C. P., & Soyiri, I. N. (2017) · BMC Bioinformatics · doi:10.1186/s12859-016-1456-0 · Cited by 158
Speech in Alzheimer's Disease: Can Temporal and Acoustic Parameters Discriminate Dementia?
Meilán, J. J. G., Martínez-Sánchez, F., Carro, J., López, D. E., Millian-Morell, L., & Arana, J. M. (2014) · Dementia and Geriatric Cognitive Disorders · doi:10.1159/000356726 · Cited by 152 · cited 6×
“Another group of studies (14) use solely voice recordings [50, 57, 60, 61, 64, 65, 71, 79, 80, 82, 84, 89, 97, 100].”
BackgroundInfluential¿Es hora de jubilar al Mini-Mental?
Carnero-Pardo, C. (2014) · Neurología · doi:10.1016/j.nrl.2013.07.003 · Cited by 138
López-de-Ipiña, K., Alonso, J. B., Solé-Casals, J., Barroso, N., Henriquez, P., Faundez-Zanuy, M., Travieso, C. M., Ecay-Torres, M., Martínez-Lage, P., & Eguiraun, H. (2015) · Cognitive Computation · doi:10.1007/s12559-013-9229-9 · Cited by 132
Left-corner parsing and psychological plausibility
Resnik, P. (1992) · Proceedings of the 14th conference on Computational linguistics - · doi:10.3115/992066.992098 · Cited by 128 · cited 1×
“Both indices have been associated with working memory [107] and showed a declining pattern in the longitudinal analysis of the written work by Iris Murdoch, a novelist who was diagnosed with AD [108].”
BackgroundDetecting Dementia Through Interactive Computer Avatars
Tanaka, H., Adachi, H., Ukita, N., Ikeda, M., Kazui, H., Kudo, T., & Nakamura, S. (2017) · IEEE Journal of Translational Engineering in Health and Medicine · doi:10.1109/jtehm.2017.2752152 · Cited by 127 · cited 13×
“In fact, 5 report samples with less than 30 participants [68, 79, 83, 89, 94].”
BackgroundInfluentialWallin, A., Nordlund, A., Jonsson, M. A. G., Lind, K., Edman, Å., Göthlin, M., Stålhammar, J., Eckerström, M., Kern, S., Börjesson‐Hanson, A., et al. (2015) · Journal of Cerebral Blood Flow & Metabolism · doi:10.1038/jcbfm.2015.147 · Cited by 123 · cited 1×
“Gothenburgh MCI database: This includes comprehensive assessments of young elderly participants during their Memory Clinic appointments and senior citizens that were recruited as their healthy counterparts [54].”
MethodKonig, A., Satt, A., Sorin, A., Hoory, R., Derreumaux, A., David, R., & Robert, P. H. (2018) · Current Alzheimer Research · doi:10.2174/1567205014666170829111942 · Cited by 119
Should the Mini-Mental State Examination be retired?
Pardo, C. C. (2014) · Neurología (English Edition) · doi:10.1016/j.nrleng.2013.07.005 · Cited by 111 · cited 2×
“However, despite being widely used in clinical and epidemiological investigations, MMSE has been criticized for having ceiling effects, especially when used to assess pre-clinical AD [72].”
BackgroundFinding a balance: The Carolinas Conversation Collection
Pope, C., & Davis, B. H. (2011) · Corpus Linguistics and Linguistic Theory · doi:10.1515/cllt.2011.007 · Cited by 110 · cited 1×
“It is collected and distributed by the Medical University of South Carolina [55].”
BackgroundThomas, C., Keselj, V., Cercone, N., Rockwood, K., & Asp, E. (2006) · IEEE International Conference Mechatronics and Automation, 2005 · doi:10.1109/icma.2005.1626789 · Cited by 110 · cited 9×
“From our point of view, the most desirable data type are conversational dialogues (5 studies), where interactive speech is prompted with the least possible constraints [10, 66, 79, 80, 98].”
BackgroundMethodInfluentialA study of voice activity detection techniques for NIST speaker recognition evaluations
Mak, M., & Yu, H.-B. (2013) · Computer Speech & Language · doi:10.1016/j.csl.2013.07.003 · Cited by 109
Bertola, L., Mota, N. B., Copelli, M., Rivero, T., Diniz, B. S., Romano-Silva, M. A., Ribeiro, S., & Malloy-Diniz, L. F. (2014) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2014.00185 · Cited by 107 · cited 9×
“In some studies, the research question targets a specific aspect of language, such as syntactical complexity [59], or a particular way of representing it, such as speech graph attributes [57].”
BackgroundMethodInfluentialPredicting MCI Status From Multimodal Language Data Using Cascaded Classifiers
Fraser, K. C., Lundholm Fors, K., Eckerström, M., Öhman, F., & Kokkinakis, D. (2019) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2019.00205 · Cited by 97 · cited 10×
“[7], mentions that data are available upon request to authors.”
BackgroundMethodInfluentialA Review of Automated Speech and Language Features for Assessment of Cognitive and Thought Disorders
Voleti, R., Liss, J. M., & Berisha, V. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2019.2952087 · Cited by 96
WATSON, C. M. (1999) · Aphasiology · doi:10.1080/026870399402181 · Cited by 95 · cited 1×
“, [4, 5]) to more recent work on computational speech technology (e.”
BackgroundEmotional Prosody Perception and Production in Dementia of the Alzheimer’s Type
Horley, K., Reid, A., & Burnham, D. (2010) · Journal of Speech, Language, and Hearing Research · doi:10.1044/1092-4388(2010/09-0030 · Cited by 91
The Syndrome of Functional Memory Disorder: Definition, Etiology, and Natural Course
Schmidtke, K., Pohlmann, S., & Metternich, B. (2008) · The American Journal of Geriatric Psychiatry · doi:10.1097/jgp.0b013e318187ddf9 · Cited by 91
Dementia detection using automatic analysis of conversations
Mirheidari, B., Blackburn, D., Walker, T., Reuber, M., & Christensen, H. (2019) · Computer Speech & Language · doi:10.1016/j.csl.2018.07.006 · Cited by 82 · cited 12×
“[8, 12, 86], who compare participants with an impairment caused by neurodegenerative disease (ND group, including AD) to an impairment caused by functional memory disoders (FMD); and Weiner and Schultz [68] and Weiner et al.”
BackgroundInfluentialClark, D. G., McLaughlin, P. M., Woo, E., Hwang, K., Hurtz, S., Ramirez, L., Eastman, J., Dukes, R., Kapur, P., DeRamus, T. P., et al. (2016) · Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring · doi:10.1016/j.dadm.2016.02.001 · Cited by 80 · cited 11×
“Within-subject comparisons have also been attempted, comparing participants who remained in a certain cognitive status to those who changed [67, 74].”
BackgroundMethodInfluentialEvaluation of speech-based protocol for detection of early-stage dementia
Satt, A., Sorin, A. S., Ronen, O., Barkan, O., Kompatsiaris, I. Y., Kokonozi, A., & Tsolaki, M. (2013) · Interspeech (USB) · doi:10.21437/interspeech.2013-32 · Cited by 78 · cited 5×
“, [61, 102]), which summarize the rates of false alarms and false negatives, are reported in less than half of the reviewed studies.”
BackgroundInfluentialYancheva, M., Fraser, K., & Rudzicz, F. (2015) · Proceedings of SLPAT 2015: 6th Workshop on Speech and Language Processing for Assistive Technologies · doi:10.18653/v1/w15-5123 · Cited by 76 · cited 1×
“[114], who extracted a more comprehensive feature set, including lexicosyntactic, acoustic, and semantic measures, and used them to predict MMSE scores.”
MethodSpeech-Based Detection of Alzheimer’s Disease in Conversational German
Weiner, J., Herff, C., & Schultz, T. (2016) · Interspeech 2016 · doi:10.21437/interspeech.2016-100 · Cited by 72 · cited 11×
“Within-subject comparisons have also been attempted, comparing participants who remained in a certain cognitive status to those who changed [67, 74].”
BackgroundMethodResultInfluentialUnderstanding spoken language through TalkBank
MacWhinney, B. (2019) · Behavior Research Methods · doi:10.3758/s13428-018-1174-9 · Cited by 66 · cited 3×
“From the remaining 12 studies, nine use data from DementiaBank (Pitt Corpus or Mandarin Lu) and do report data origin and availability.”
MethodInfluentialPakhomov, S., Chacón, D. A., Wicklund, M., & Gundel, J. K. (2010) · Behavior Research Methods · doi:10.3758/s13428-010-0037-9 · Cited by 63
Detecting Signs of Dementia Using Word Vector Representations
Mirheidari, B., Blackburn, D., Walker, T., Venneri, A., Reuber, M., & Christensen, H. (2018) · Interspeech 2018 · doi:10.21437/interspeech.2018-1764 · Cited by 63 · cited 13×
“There are only 6 studies where not only the dataset, but also each experimental group contains 100 or more participants/speech samples [6, 9, 11, 85, 92, 95].”
BackgroundMethodInfluentialPakhomov, S. V. S., & Hemmy, L. S. (2014) · Cortex · doi:10.1016/j.cortex.2013.05.009 · Cited by 61
Toward the Automation of Diagnostic Conversation Analysis in Patients with Memory Complaints
Mirheidari, B., Blackburn, D., Harkness, K., Walker, T., Venneri, A., Reuber, M., & Christensen, H. (2017) · Journal of Alzheimer’s Disease · doi:10.3233/jad-160507 · Cited by 54 · cited 8×
“[8, 12, 86], who compare participants with an impairment caused by neurodegenerative disease (ND group, including AD) to an impairment caused by functional memory disoders (FMD); and Weiner and Schultz [68] and Weiner et al.”
BackgroundMethodInfluentialThe heterogeneity of picture-supported narratives in Alzheimer’s disease
Duong, A., Giroux, F., Tardif, A., & Ska, B. (2005) · Brain and Language · doi:10.1016/j.bandl.2004.10.007 · Cited by 53 · cited 5×
“[69], who attempt to describe AD and HC discourse patterns through cluster analysis.”
BackgroundInfluentialTwo-stage feature selection of voice parameters for early Alzheimer’s disease prediction
Mirzaei, S., El Yacoubi, M., Garcia-Salicetti, S., Boudy, J., Kahindo, C., Cristancho-Lacroix, V., Kerhervé, H., & Rigaud, A.-S. (2018) · IRBM · doi:10.1016/j.irbm.2018.10.016 · Cited by 52
dos Santos, L., Júnior, E. A. C., Oliveira, O. N., Amancio, D. R., Mansur, L. L., & Aluísio, S. M. (2017) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p17-1118 · Cited by 52 · cited 7×
“Apart from English, only Portuguese and Chinese have been researched exclusively with text features [57, 82, 90].”
BackgroundInfluentialMultilingual word embeddings for the assessment of narrative speech in mild cognitive impairment
Fraser, K. C., Lundholm Fors, K., & Kokkinakis, D. (2019) · Computer Speech & Language · doi:10.1016/j.csl.2018.07.005 · Cited by 51 · cited 8×
“Exceptionally, two of these studies work with written rather than spoken language [87, 88].”
BackgroundMethodInfluentialHigher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease
Nasrolahzadeh, M., Mohammadpoory, Z., & Haddadnia, J. (2018) · Cognitive Neurodynamics · doi:10.1007/s11571-018-9499-8 · Cited by 51 · cited 11×
“Out of 51 reviewed papers, only seven did attempt three-way [50, 62–64] or four-way [12, 65, 66] classification.”
BackgroundMethodResultInfluentialLongitudinal Monitoring and Detection of Alzheimer's Type Dementia from Spontaneous Speech Data
Luz, S. (2017) · 2017 IEEE 30th International Symposium on Computer-Based Medical Systems (CBMS) · doi:10.1109/cbms.2017.41 · Cited by 49 · cited 6×
“The same applies to Luz et al. [10], who made available their identifiers for the CCC dataset.”
BackgroundResultInfluentialSelf-citationSpeech Processing for Early Alzheimer Disease Diagnosis: Machine Learning Based Approach
Ben Ammar, R., & Ben Ayed, Y. (2018) · 2018 IEEE/ACS 15th International Conference on Computer Systems and Applications (AICCSA) · doi:10.1109/aiccsa.2018.8612831 · Cited by 43 · cited 8×
“It is reported in all papers but five, of which two are not very recent [66, 93], another two do not report CV but report using a hold-out set [82, 91], and only one reports using neither CV nor a hold-out set procedure [95].”
BackgroundMethodInfluentialSpeech and language alterations in dementia syndromes: Characteristics and treatment
Ross, G. W., Cummings, J. L., & Benson, D. F. (1990) · Aphasiology · doi:10.1080/02687039008249087 · Cited by 40 · cited 1×
“Although memory loss is often considered the signature symptom of AD, language impairment may also appear in its early stages [3].”
BackgroundSpeech Processing Approach for Diagnosing Dementia in an Early Stage
Sadeghian, R., Schaffer, J. D., & Zahorian, S. A. (2017) · Interspeech 2017 · doi:10.21437/interspeech.2017-1712 · Cited by 40 · cited 10×
“A few studies (6) work with audio files and associated ASR transcriptions [12, 44, 62, 63, 77, 99].”
BackgroundMethodInfluentialThe natural history of Alzheimer’s disease
Corey‐Bloom, J., & Rafii, M. S. (2017) · doi:10.1201/9781315381572-56 · Cited by 39
Rentoumi, V., Paliouras, G., Danasi, E., Arfani, D., Fragkopoulou, K., Varlokosta, S., & Papadatos, S. (2017) · 2017 8th IEEE International Conference on Cognitive Infocommunications (CogInfoCom) · doi:10.1109/coginfocom.2017.8268212 · Cited by 39 · cited 6×
“Some studies choose random guess or naive estimations (ZeroR) [10, 11, 66, 74, 88] as their baseline performance.”
BackgroundMethodInfluentialEvaluación y diagnóstico del deterioro cognitivo leve
Peña, M. M., Carrasco, P. M., Luque, M. L., & García, A. I. R. (2012) · Revista de Logopedia Foniatría y Audiología · doi:10.1016/j.rlfa.2012.03.002 · Cited by 38 · cited 1×
“Most of these tests have been translated into different languages, such as the Spanish version of the MMSE (MEC) [35], which is used in a few reviewed papers.”
MethodTelephone-based Dementia Screening I: Automated Semantic Verbal Fluency Assessment
Tröger, J., Linz, N., König, A., Robert, P. H., & Alexandersson, J. (2018) · International Conference on Pervasive Computing Technologies for Healthcare · doi:10.1145/3240925.3240943 · Cited by 38 · cited 9×
“Another four studies take into account feasibility for clinical screening within their system design, for example, collecting data directly with a computerized decision support system [60], through humanrobot interaction [94], a computer-supported screening tool [77], or simulating telephonebased data [96].”
BackgroundResultInfluentialDetecting Alzheimer’s Disease from Continuous Speech Using Language Models
Guo, Z., Ling, Z., & Li, Y. (2019) · Journal of Alzheimer’s Disease · doi:10.3233/jad-190452 · Cited by 38 · cited 5×
“Similar patterns apply to other studies where the number of participants and samples are reported [92, 98].”
BackgroundResultInfluentialDetecting and Predicting Alzheimer's Disease Severity in Longitudinal Acoustic Data
Al-Hameed, S., Benaissa, M., & Christensen, H. (2017) · International Conference on Bioinformatics Research and Applications (ICBRA) · doi:10.1145/3175587.3175589 · Cited by 35 · cited 1×
“[112], who extracted 811 acoustic features to build a regression model able to predict MMSE scores with an average mean absolute error of 3.”
MethodPredicting Dementia Screening and Staging Scores from Semantic Verbal Fluency Performance
Linz, N., Tröger, J., Alexandersson, J., Wolters, M. K., König, A., & Robert, P. H. (2017) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw.2017.100 · Cited by 34 · cited 1×
“[111], for instance, trained a regression model that used the SVF to predict MMSE scores and obtained a mean absolute error of 2.”
MethodA Method for Analysis of Patient Speech in Dialogue for Dementia Detection
Luz, S. F., de la Fuente, S., & Albert, P. (2018) · Edinburgh Research Explorer (University of Edinburgh) · doi:10.48550/arxiv.1811.09919 · Cited by 34 · cited 13×
“Only two studies meet all five criteria with a “yes” [10, 84].”
BackgroundMethodInfluentialSelf-citationMirheidari, B., Blackburn, D., OrMalley, R., Walker, T., Venneri, A., Reuber, M., & Christensen, H. (2019) · ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2019.8682423 · Cited by 34 · cited 12×
“[8, 12, 86], who compare participants with an impairment caused by neurodegenerative disease (ND group, including AD) to an impairment caused by functional memory disoders (FMD); and Weiner and Schultz [68] and Weiner et al.”
BackgroundMethodInfluentialBEA – A multifunctional Hungarian spoken language database
Gósy, M. (2013) · Repository of the Academy's Library (Library of the Hungarian Academy of Sciences) · Cited by 33
Reynolds, D. A. (2009) · Springer eBooks · doi:10.1007/978-0-387-73003-5_197 · Cited by 33
de Ipiña, K. L., Alonso, J. B., Solé‐Casals, J., Travieso, C. M., Ezeiza, A., Faúndez-Zanuy, M., Calvo, P. M., & Beitia, B. (2014) · Neurocomputing · doi:10.1016/j.neucom.2014.05.083 · Cited by 30
An Avatar-Based System for Identifying Individuals Likely to Develop Dementia
Mirheidari, B., Blackburn, D., Harkness, K., Walker, T., Venneri, A., Reuber, M., & Christensen, H. (2017) · Interspeech 2017 · doi:10.21437/interspeech.2017-690 · Cited by 30
Graph-Based Word Alignment for Clinical Language Evaluation
Prud’hommeaux, E. T., & Brian Roark, A. (2015) · Computational Linguistics · doi:10.1162/coli_a_00232 · Cited by 29 · cited 7×
“Others calculate their baseline by training a classifier with all the generated features, that is, before attempting to reduce the feature set with either selection or extraction methods [83, 95, 99], with cognitive scores only [7, 44, 70, 71] or by training a classifier with demographic scores only [63].”
BackgroundMethodInfluentialKato, S., Endo, H., Homma, A., Sakuma, T., & Watanabe, K. (2013) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2013.6610873 · Cited by 29 · cited 8×
“For example, the HC group is labelled as CON (control) [91], NC (normal cognition) [57, 64, 88, 99], CH (cognitively healthy) [82], and CN (cognitively normal) [67].”
BackgroundMethodInfluentialAutomatic Prosodic Analysis to Identify Mild Dementia
Gonzalez-Moreira, E., Torres-Boza, D., Hernández-Díaz, H. A. K., Ferrer, C. A., Garcia-Zamora, M. A., Espinoza-Cuadros, F., & Gómez, L. A. H. (2015) · BioMed Research International · doi:10.1155/2015/916356 · Cited by 26 · cited 10×
“, [89]), and especially if a standardized available feature set is used [11].”
BackgroundMethodInfluentialde la Fuente Garcia, S., Ritchie, C. W., & Luz, S. (2019) · BMJ Open · doi:10.1136/bmjopen-2018-026254 · Cited by 25
Self-citationAnalysis of Speech-Based Measures for Detecting and Monitoring Alzheimer’s Disease
Khodabakhsh, A., & Demiroğlu, C. (2014) · Methods in molecular biology · doi:10.1007/978-1-4939-1985-7_11 · Cited by 22 · cited 4×
“The most comprehensive studies include spectral, ASR-related, prosodic temporal, voice quality features [8, 49, 60, 79, 84, 92, 100], as well as features derived from the Higuchi Fractal Dimension [80] or from higher order spectral analysis [65].”
BackgroundInfluentialKristina, L. F., Kathleen, F., & Dimitrios, K. (2018) · Studies in Health Technology and Informatics · doi:10.3233/978-1-61499-852-5-705 · Cited by 22 · cited 6×
“The remaining studies work on disease progression by including SCI participants [59], predicting within-subject change [68] or discriminating between HC, MCI, and AD stages (e.”
BackgroundMethodInfluentialLearning multiview embeddings for assessing dementia
Pou-Prom, C., & Rudzicz, F. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1304 · Cited by 20 · cited 2×
“A few other works used the Pitt Corpus for similar purposes, such as Al-Hameed et al. [112], who extracted 811 acoustic features to build a regression model able to predict MMSE scores with an average mean absolute error of 3.1; or Pou-Prom and Rudzicz [113], who used a multiview embedding to capture different levels…”
BackgroundInfluentialAutomated speech-based screening for alzheimer's disease in a care service scenario
Tröger, J., Linz, N., Alexandersson, J., König, A., & Robert, P. (2017) · Proceedings of the 11th EAI International Conference on Pervasive Computing Technologies for Healthcare · doi:10.1145/3154862.3154915 · Cited by 19 · cited 8×
“Another essential novelty is related to longitudinal aspects of data [68, 77, 97].”
BackgroundInfluentialDanso, S., Terrera, G. M., Luz, S. F., & Ritchie, C. W. (2019) · Journal of Global Health · doi:10.7189/jogh.09.020322 · Cited by 19
Self-citationSpeech as a means of monitoring cognitive function of elderly speakers
D’Arcy, S., Rapčan, V., Pénard, N., Morris, M. E., Robertson, I. H., & Reilly, R. B. (2008) · Interspeech (USB) · doi:10.21437/interspeech.2008-582 · Cited by 18 · cited 6×
“It is reported in all papers but five, of which two are not very recent [66, 93], another two do not report CV but report using a hold-out set [82, 91], and only one reports using neither CV nor a hold-out set procedure [95].”
BackgroundInfluentialEspinoza-Cuadros, F., Garcia-Zamora, M. A., Torres-Boza, D., Ferrer, C. A., Benavides, A. M., Gonzalez-Moreira, E., & Gómez, L. A. H. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-13623-3_23 · Cited by 18 · cited 7×
“However, there is a fair amount of papers using data in a variety of languages, including: Italian [91], Portuguese [57, 90], Chinese and Taiwanese [82], French [49, 69, 77, 96, 102], Hungarian [62, 63, 99], Spanish [83, 89, 100], Swedish [7, 59, 87], Japanese [64, 71, 94], Turkish [84], Persian [65], Greek [61, 88],…”
BackgroundMethodInfluentialBeltrami, D., Calzà, L., Gagliardi, G., Ghidoni, E., Marcello, N., Favretti, R. R., & Tamburini, F. (2016) · Language Resources and Evaluation Conference (LREC) · doi:10.63317/29dtk8secyzx · Cited by 18 · cited 10×
“, [91]), or taking part in a semi-standardized (e.”
BackgroundMethodInfluentialWeiner, J., Frankenberg, C., Telaar, D., Wendelstein, B., Schröder, J., & Schultz, T. (2016) · Language Resources and Evaluation Conference (LREC) · doi:10.63317/2ksqnso2i9ea · Cited by 18
Assessing Alzheimer’s disease from speech using the i-vector approach
Egas López, J. V., Tóth, L., Hoffmann, I., Kálmán, J., Pákáski, M., & Gosztolya, G. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-26061-3_30 · Cited by 18 · cited 9×
“MCI and AD are more homogeneous due to being diagnostic categories that need to meet certain clinical criteria to be assigned, although some studies do refer to AD as dementia [62, 95].”
BackgroundMethodInfluentialImplementation of the Standard I-vector System for the Kaldi Speech Recognition Toolkit
Madikeri, S. R., Dey, S., Motlíček, P., & Ferràs, M. (2016) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 17
Speech-Based Automated Cognitive Impairment Detection From Remotely-Collected Cognitive Test Audio
Yu, B., Williamson, J. R., Mundt, J. C., & Quatieri, T. F. (2018) · IEEE Access · doi:10.1109/access.2018.2856478 · Cited by 16 · cited 8×
“Another essential novelty is related to longitudinal aspects of data [68, 77, 97].”
BackgroundMethodInfluentialGuinn, C. I., Singer, B., & Habash, A. (2014) · IEEE Symposium on Computational Intelligence in Healthcare and e-health (CICARE) · doi:10.1109/cicare.2014.7007840 · Cited by 15 · cited 9×
“Similar patterns apply to other studies where the number of participants and samples are reported [92, 98].”
BackgroundMethodResultInfluentialALZUMERIC: A decision support system for diagnosis and monitoring of cognitive impairment
de Lizarduy, U. M., Salomón, P. C., Gómez‐Vilda, P., Ecay‐Torres, M., & de Ipiña, K. L. (2017) · Loquens · doi:10.3989/loquens.2017.037 · Cited by 15 · cited 11×
“Another modality is video, which is available in some datasets such as CCC [10, 98], AZTITXIKI [79], AZTIAHORE [60, 80], IVA [12, 85], or the one in Tanaka et al.”
BackgroundMethodInfluentialChien, Y.-W., Hong, S.-Y., Cheah, W.-T., Fu, L.-C., & Chang, Y.-L. (2018) · 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC) · doi:10.1109/smc.2018.00557 · Cited by 15 · cited 9×
“It is reported in all papers but five, of which two are not very recent [66, 93], another two do not report CV but report using a hold-out set [82, 91], and only one reports using neither CV nor a hold-out set procedure [95].”
BackgroundMethodInfluentialAlignment of spoken narratives for automated neuropsychological assessment
Prud’hommeaux, E. T., & Roark, B. (2011) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru.2011.6163979 · Cited by 14 · cited 4×
“Some examples are studies investigating the potential of topic model features [87], emotional features [79], fractal dimension features [80], higher order spectral features [65], or feature extracted automatically, as opposed to manually [73, 85, 96].”
BackgroundMethodInfluentialDetection of Intra-Personal Development of Cognitive Impairment From Conversational Speech.
Weiner, J., & Schultz, T. (2016) · ITG Symposium of Speech Communication · Cited by 14
Ekberg, K., & Reuber, M. (2016) · Communication & Medicine · doi:10.1558/cam.v12i1.26851 · Cited by 13
CAL: A Smart Home Environment for Monitoring Cognitive Decline
Fredericks, E. M., Bowers, K. M., Price, K. A., & Hariri, R. H. (2018) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2018.00155 · Cited by 12 · cited 1×
“In terms of aims and objectives, research suggests that embedded devices installed in the home to monitor patient health and safety may delay institutionalization [123], and therefore more emphasis should be placed on the feasibility of remote evaluations.”
BackgroundShinkawa, K., Kosugi, A., Nishimura, M., Nemoto, M., Nemoto, K., Takeuchi, T., Numata, Y., Watanabe, R., Tsukada, E., Ota, M., et al. (2019) · Studies in health technology and informatics · doi:10.3233/shti190240 · Cited by 12 · cited 11×
“Others calculate their baseline by training a classifier with all the generated features, that is, before attempting to reduce the feature set with either selection or extraction methods [83, 95, 99], with cognitive scores only [7, 44, 70, 71] or by training a classifier with demographic scores only [63].”
BackgroundMethodInfluentialAffective Speech for Alzheimer's Dementia Recognition.
Haider, F., García, S. D. L. F., Albert, P., & Luz, S. F. (2020) · Edinburgh Research Explorer (University of Edinburgh) · Cited by 11
Self-citationdos Santos, L. B., Corrêa, E. A., Oliveira, L. N., Amancio, D. R., Mansur, L. L., & Aluísio, S. M. (2017) · arXiv (Cornell University) · doi:10.60692/hbh6f-1v045 · Cited by 9
Rochford, I., Rapčan, V., D’Arcy, S., & Reilly, R. B. (2012) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2012.6346770 · Cited by 4 · cited 2×
“Another pre-processing novelty is the use of dynamic threshold for pause behavior [76], which could be essential for personalized screening.”
BackgroundLewis, Z., & Cuddy, R. (1995) · Medical Entomology and Zoology · Cited by 1
Research works printed but not linked: 4
- World Health Organization, 2013, Mental health action plan 2013-2020, WHO Library Cataloguing-in-Publication Data, 1
- Joanette, 1995, Évaluation neuropsychologique et profils cognitifs des démences de type Alzheimer: dissociations transversales et longitudinales, Neuropsychologie Clinique des Démences, 91
- Bayles, 1993, ABCD: Arizona battery for communication disorders of dementia
- Darley, 1975, Motor speech disorders
Cited by
Digital medicine and the curse of dimensionality
Berisha, V., Krantsevich, C., Hahn, P. R., Hahn, S., Dasarathy, G., Turaga, P., & Liss, J. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00521-5 · Cited by 344
Górriz, J. M., Álvarez-Illán, I., Álvarez-Marquina, A., Arco, J. E., Atzmueller, M., Ballarini, F., Barakova, E., Bologna, G., Bonomini, P., Castellanos-Dominguez, G., et al. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101945 · Cited by 214
Kale, M., Wankhede, N., Pawar, R., Ballal, S., Kumawat, R., Goswami, M., Khalid, M., Taksande, B., Upaganlawar, A., Umekar, M., et al. (2024) · Ageing Research Reviews · doi:10.1016/j.arr.2024.102497 · Cited by 183
A Comprehensive Review of AI Techniques for Addressing Algorithmic Bias in Job Hiring
Albaroudi, E., Mansouri, T., & Alameer, A. (2024) · AI · doi:10.3390/ai5010019 · Cited by 285
Detecting Cognitive Decline Using Speech Only: The ADReSSo Challenge
Luz, S. F., Haider, F., de la Fuente, S., Fromm, D., & MacWhinney, B. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-1220 · Cited by 218
Self-citation
Metadata
- Authors
- de la Fuente Garcia, S., Ritchie, C. W., & Luz, S.
- Affiliations
- Usher Institute, Edinburgh Medical School, The University of Edinburgh, Scotland, UK
- University of Edinburgh (GB)
- Venue
- Journal of Alzheimer's Disease
- Publisher
- SAGE Publications
- Published
- 2020-12-08
- Volume
- 78
- Issue
- 4
- Pages
- 1547-1574
- ISSN
- 1387-2877 (ISSN-L)
- 1875-8908
- PMC
- PMC7836050
- OpenAlex
- W3093156670
- Cited by
- 5
- References
- 127
- Citations by year
- 2025: 1 · 2023: 1 · 2022: 1 · 2021: 2
- FWCI
- 0.17
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Journal of Alzheimer s Disease (publishedVersion, cc-by-nc)
- PubMed Central (publishedVersion, cc-by-nc)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, other-oa)
- Topics
- Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Health Sciences › Health Professions › Speech and Hearing › Voice and Speech Disorders
- Keywords
- Alzheimer's disease, speech analysis
- MeSH terms
- Machine Learning, Alzheimer Disease, Artificial Intelligence, Humans, Natural Language Processing, Speech, Disease Progression, 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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.

![Fig.2 Screening and selection procedure, following guidelines provided by PRISMA [19].](https://figures.cochat.ai/paper-figures/58d5ef02d2464a468815fbe4a20aa32d/artificial-intelligence-speech-and-language-processing-approaches-to-monitoring-figure-2.jpg)