Linguistic Features Identify Alzheimer’s Disease in Narrative Speech
The DOI matches its Crossref record; 97 of the 98 references it cites were found in a registry or an index.
Abstract
Background: Although memory impairment is the main symptom of Alzheimer’s disease (AD), language impairment can be an important marker. Relatively few studies of language in AD quantify the impairments in connected speech using computational techniques. Objective: We aim to demonstrate state-of-the-art accuracy in automatically identifying Alzheimer’s disease from short narrative samples elicited with a picture description task, and to uncover the salient linguistic factors with a statistical factor analysis. Methods: Data are derived from the DementiaBank corpus, from which 167 patients diagnosed with “possible” or “probable” AD provide 240 narrative samples, and 97 controls provide an additional 233. We compute a number of linguistic variables from the transcripts, and acoustic variables from the associated audio files, and use these variables to train a machine learning classifier to distinguish between participants with AD and healthy controls. To examine the degree of heterogeneity of linguistic impairments in AD, we follow an exploratory factor analysis on these measures of speech and language with an oblique promax rotation, and provide interpretation for the resulting factors. Results: We obtain state-of-the-art classification accuracies of over 81% in distinguishing individuals with AD from those without based on short samples of their language on a picture description task. Four clear factors emerge: semantic impairment, acoustic abnormality, syntactic impairment, and information impairment. Conclusion: Modern machine learning and linguistic analysis will be increasingly useful in assessment and clustering of suspected AD.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“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 [14]
In the citing paper
“For example, a large number of lexical, syntactic, and acoustic features was used to classify patients with the amnestic AD presentation versus HC in the DementiaBank corpus with 82% accuracy.”
From Automatic classification of AD pathology in FTD phenotypes using natural speech (2024) · cites this paper as [14]
In the citing paper
“Acoustic, syntactic, and semantic features, including grammatical complexity and information content, can correlate strongly with cognitive decline [11,32].”
From A Conversational Robot for Older Adults with Alzheimer's Disease (2020) · cites this paper as [11]
In the citing paper
“Our linguistic analysis for cognitive monitoring has followed the work of Fraser et al. [11] and Roark et al. [32] and is restricted to examining features that can be automatically extracted from transcriptions of conversations.”
From A Conversational Robot for Older Adults with Alzheimer's Disease (2020) · cites this paper as Fraser et al. [11]
In the citing paper
“Accuracy of 92% was obtained in distinguishing people suffering from AD and healthy control group using top 25 ranked features [134].”
From Recent Advancements in Pathogenesis, Diagnostics and Treatment of Alzheimer’s Disease (2020) · cites this paper as [134]
7 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
A fast learning algorithm for deep belief nets
Hinton, G. E., Osindero, S., & Teh, Y.-W. (2006) · Neural Computation · doi:10.1162/neco.2006.18.7.1527 · Cited by 16,336
Costello, A. B., & Osborne, J. (2005) · University of Massachusetts Amherst · doi:10.7275/jyj1-4868 · Cited by 10,297 · cited 1×
“To fairly examine the degree of heterogeneity of linguistic impairments in patients with AD, we chose promax, an oblique rotation which allows factors to be correlated with each other [71].”
Evaluating the use of exploratory factor analysis in psychological research.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999) · Psychological Methods · doi:10.1037/1082-989x.4.3.272 · Cited by 9,587
Brysbaert, M., & New, B. (2009) · Behavior Research Methods · doi:10.3758/brm.41.4.977 · Cited by 2,821 · cited 1×
“Specifically, we use the SUBTL frequency norms [53] and the combined Bristol and Gilhooly-Logie norms [54, 55] for familiarity, imageability, and age-ofacquisition.”
The CHILDES Project: Tools for Analyzing Talk
Pye, C., & MacWhinney, B. (1994) · Language · doi:10.2307/416745 · Cited by 2,301
Cognition and anatomy in three variants of primary progressive aphasia
Gorno‐Tempini, M. L., Dronkers, N. F., Rankin, K. P., Ogar, J. M., Phengrasamy, L., Rosen, H. J., Johnson, J. K., Weiner, M. W., & Miller, B. L. (2004) · Annals of Neurology · doi:10.1002/ana.10825 · Cited by 1,625 · cited 2×
“All of the high loadings reflect characteristics of 629 semantically impoverished language, similar to that 630 seen in semantic dementia, a degenerative disorder 631 specifically affecting the temporal lobe [72].”
Using uh and um in spontaneous speaking
Clark, H. (2002) · Cognition · doi:10.1016/s0010-0277(02)00017-3 · Cited by 1,235 · cited 1×
“As described in the previous section , we remove the filled pauses uh, um, er, and ah, but we keep track of the frequency of occurrence of each type, as previous work has suggested that they may serve different purposes [49].”
Automatic analysis of syntactic complexity in second language writing
Lu, X. (2010) · International Journal of Corpus Linguistics · doi:10.1075/ijcl.15.4.02lu · Cited by 1,041
Age-of-acquisition, imagery, concreteness, familiarity, and ambiguity measures for 1,944 words
Gilhooly, K. J., & Logie, R. H. (1980) · Behavior Research Methods & Instrumentation · doi:10.3758/bf03201693 · Cited by 868 · cited 1×
“Specifically, we use the SUBTL frequency norms [53] and the combined Bristol and Gilhooly-Logie norms [54, 55] for familiarity, imageability, and age-ofacquisition.”
Verbal fluency performance in dementia of the Alzheimer’s type: a meta-analysis
Henry, J. D., Crawford, J. R., & Phillips, L. H. (2004) · Neuropsychologia · doi:10.1016/j.neuropsychologia.2004.02.001 · Cited by 847
Research works printed but not linked: 3
- Kempler, 1995, Dementia and Communication: Research and Clinical Implications, 98
- Brunét, 1978, Le Vocabulaire de Jean Giraudoux: Structure et Evolution
- Honoré, 1979, Some simple measures of richness of vocabulary, Assoc Lit Linguist Comput Bull, 7, 172
Cited by
Items 101–200 of 800
Ostrand, R., & Gunstad, J. (2021) · Journal of Geriatric Psychiatry and Neurology · doi:10.1177/0891988720933358 · Cited by 49
“This is in contrast to most existing work that only predicted outcomes for already-impaired participants (e.g., a study by Kavé & Dassa47) and/or merely conducted a binary by-group classification to discriminate patients from controls, rather than predicting the degree of impairment (e.g., see studies by Fraser et al…”
BackgroundInfluentialPauses for Detection of Alzheimer’s Disease
Yuan, J., Cai, X., Bian, Y., Ye, Z., & Church, K. (2021) · Frontiers in Computer Science · doi:10.3389/fcomp.2020.624488 · Cited by 49
“Fraser et al. (2016) used 370 acoustic and linguistic features to train logistic regression models for classifying AD and normal speech.”
MethodMeilán, J. J. G., Martínez‐Sánchez, F., Martínez‐Nicolás, I., Llorente, T. E., & Carro, J. (2020) · Behavioural Neurology · doi:10.1155/2020/4683573 · Cited by 64
“[5] have found a close correlation between the severity of dementia and different linguistic metrics; nevertheless, the results in studies with patients in the mild-tomoderate stages of AD [6, 7] cannot be generalized to earlier stages of the disease when compensatory abilities may result in more subtle impairments in…”
BackgroundResultGeng, M., Xie, X., Ye, Z., Wang, T., Li, G., Hu, S., Liu, X., & Meng, H. (2022) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2022.3195113 · Cited by 48
“wider context, speech and language impairments are also commonly found among older adults experiencing natural ageing and neurocognitive disorders, for example, Alzheimer’s disease [22], [23].”
BackgroundA finer-grained linguistic profile of Alzheimer's disease and Mild Cognitive Impairment
Chapin, K., Clarke, N., Garrard, P., & Hinzen, W. (2022) · Journal of Neurolinguistics · doi:10.1016/j.jneuroling.2022.101069 · Cited by 57
Longitudinal 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
“Consequently, machine learning approaches have often employed lexical and sometimes syntactic structure as categorisation features [4].”
BackgroundAgbavor, F., & Liang, H. (2022) · Brain Sciences · doi:10.3390/brainsci13010028 · Cited by 47
“[13 – 15].”
Personalized Adversarial Data Augmentation for Dysarthric and Elderly Speech Recognition
Jin, Z., Geng, M., Deng, J., Wang, T., Hu, S., Li, G., & Liu, X. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3323888 · Cited by 62
“In a wider context, speech and language impairments are also commonly found among older adults experiencing natural ageing and neurocognitive disorders, for example, Alzheimer’s disease [21], [22].”
BackgroundAn analysis of eye-movements during reading for the detection of mild cognitive impairment
Fraser, K., Fors, K. L., Kokkinakis, D., & Nordlund, A. (2017) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d17-1107 · Cited by 57
“…through automated analysis of speech and language production (Thomas et al., 2005; Pakhomov et al., 2010; Guinn and Habash, 2012; Meil´an et al., 2014; Jarrold et al., 2014; Fraser et al., 2016; Rentoumi et al., 2014; Garrard et al., 2014; Prud’hommeaux and Roark, 2015; Yancheva et al., 2015).”
MethodSelf-citationAlzheimer’s Dementia Recognition From Spontaneous Speech Using Disfluency and Interactional Features
Nasreen, S., Rohanian, M., Hough, J., & Purver, M. (2021) · Frontiers in Computer Science · doi:10.3389/fcomp.2021.640669 · Cited by 56
Speech pause distribution as an early marker for Alzheimer’s disease
Pastoriza-Domínguez, P., Torre, I. G., Diéguez-Vide, F., Gómez‐Ruiz, I., Geladó, S., López, J. B., Ávila, A., Matías‐Guiu, J. A., Pytel, V., & Hernández-Fernändez, A. (2021) · Speech Communication · doi:10.1016/j.specom.2021.11.009 · Cited by 62
Kobayashi, M., Kosugi, A., Takagi, H., Nemoto, M., Nemoto, K., Arai, T., & Yamada, Y. (2019) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-29390-1_4 · Cited by 45
“Among them, numerous studies investigated the difference of information content in description tasks [54][55] and found that individuals with MCI and AD tend to produce descriptions with lower information content than healthy controls in both verbal and written picture descriptions tasks [56][57].”
Cunningham, K. T., & Haley, K. L. (2020) · Journal of Speech Language and Hearing Research · doi:10.1044/2019_jslhr-19-00226 · Cited by 58
Comparing Acoustic-Based Approaches for Alzheimer’s Disease Detection
Balagopalan, A., & Novikova, J. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-759 · Cited by 56
Prediction of venous thromboembolism using semantic and sentiment analyses of clinical narratives
Sabra, S., Mahmood Malik, K., & Alobaidi, M. (2018) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2017.12.026 · Cited by 47
Automated Recognition of Alzheimer’s Dementia Using Bag-of-Deep-Features and Model Ensembling
Syed, Z. S., Syed, M. S. S., Lech, M., & Pirogova, E. (2021) · IEEE Access · doi:10.1109/access.2021.3090321 · Cited by 64
“For this, we computed a variety of textual features, inspired by the work of Fraser et al. [54] as part of our investigation into determining efficacy of handcrafted textual features derived from speech transcripts.”
MethodSpoken language biomarkers for detecting cognitive impairment
Alhanai, T., Au, R., & Glass, J. R. (2017) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru.2017.8268965 · Cited by 54
“Speech data used in experiments varied from being highly structured memory and reading tests [29, 28, 30], to organic, open-ended, conversation-like interactions (e.”
BackgroundSpeech 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
Leveraging Data Science for Global Health
Celi, L. A., Majumder, M. S., Ordóñez, P., Osorio, J. S., Paik, K., & Somai, M. M. (2020) · Springer eBooks · doi:10.1007/978-3-030-47994-7 · Cited by 55
Multiscale System for Alzheimer’s Dementia Recognition Through Spontaneous Speech
Edwards, E., Dognin, C., Bollepalli, B., & Singh, M. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2781 · Cited by 59
“The work of [15] extended this approach to classifi-cation, and obtained state-of-the-art results on DemantiaBank-fused linguistic and acoustic features extracted into a logistic regression classifier.”
BackgroundEvaluating voice-assistant commands for dementia detection
Liang, X., Batsis, J. A., Zhu, Y., Driesse, T. M., Roth, R. M., Kotz, D., & MacWhinney, B. (2022) · Computer Speech & Language · doi:10.1016/j.csl.2021.101297 · Cited by 43
“Participants were given credit for mentioning a given topic (Croisile et al., 1996; Bschor et al., 2001; Ahmed et al., 2013a,b; Fraser et al., 2016, 2019).”
Backgrounddos 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
“We automatically removed filled pauses such as uh , um , er , and ah (e.g. uh it seems to be summer out ), short false starts (e.g. just t the ones ), and repetition (e.g. mother’s fin-ished certain of the the dishes ), as in (Fraser et al., 2015).”
BackgroundMethodInfluentialVijverberg, E. G. B., Axelsen, T., Bihlet, A. R., Henriksen, K., Weber, F., Fuchs, K., Harrison, J., Kühn‐Wache, K., Alexandersen, P., Prins, N. D., et al. (2021) · Alzheimer s Research & Therapy · doi:10.1186/s13195-021-00882-9 · Cited by 46
“Digital speech biomarkers have, so far, only been used in an experimental diagnostic setting [26, 38, 39].”
BackgroundGuo, Y., Li, C., Roan, C., Pakhomov, S., & Cohen, T. (2021) · Frontiers in Computer Science · doi:10.3389/fcomp.2021.642517 · Cited by 48
“The majority of this prior work has involved the application of supervised machine learning methods (see e.g., Orimaye et al., 2014, 2017, 2018; Fraser et al., 2016; Yancheva and Rudzicz, 2016; Karlekar et al., 2018; Cohen and Pakhomov, 2020) to classify groups of transcripts, specific transcripts…”
Noninvasive automatic detection of Alzheimer's disease from spontaneous speech: a review
Qi, X., Zhou, Q., Dong, J., & Bao, W. (2023) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2023.1224723 · Cited by 59
“, 2014), lexical diversity (Fraser et al., 2016a), global vectors (GLoVe) (Pennington et al.”
BackgroundStark, B. C., & Fukuyama, J. (2020) · Language Cognition and Neuroscience · doi:10.1080/23273798.2020.1862258 · Cited by 48
“…the most commonly used elicitation method for adult clinical populations (acquired aphasia; primary progressive aphasia; other dementias) is the single picture description task, a type of expositional elicitation method (Bryant et al., 2016; Faroqi-Shah et al., 2020; K. C. Fraser et al., 2015).”
MethodZolnoori, M., Zolnour, A., & Topaz, M. (2023) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2023.102624 · Cited by 60
Formulaic Language in People with Probable Alzheimer’s Disease: A Frequency-Based Approach
Zimmerer, V. C., Wibrow, M., & Varley, R. A. (2016) · Journal of Alzheimer's Disease · doi:10.3233/jad-160099 · Cited by 43
“Our research program sees considerable potential in automatic, or semi-automated language analysis in diagnosis and classification, and such analyses, using very different approaches, are being implemented in dementias [27, 69], schizophrenia [70], and language acquisition [71].”
A Prototype for the Voice Analysis Diagnosis of Alzheimer’s Disease
Martínez‐Sánchez, F., Meilán, J. J. G., Carro, J., & Ivanova, O. (2018) · Journal of Alzheimer s Disease · doi:10.3233/jad-180037 · Cited by 50
“These alterations have been observed both in narrative tasks involving the description of images, in which semantic alterations are symptoms of the disease [10, 11], as well as in reading tasks [12].”
Identification of digital voice biomarkers for cognitive health
Lin, H., Karjadi, C., Ang, T. F. A., Prajakta, J., McManus, C., Alhanai, T. W., Glass, J., & Au, R. (2020) · Exploration of Medicine · doi:10.37349/emed.2020.00028 · Cited by 39
“Studies examining the ability to distinguish individuals with cognitive impairment from typical controls based on voice and language parameters alone have reported accuracies as high as 90% [17-19].”
BackgroundImmersive Virtual Reality-Based Methods for Assessing Executive Functioning: Systematic Review
Kirkham, R., Kooijman, L., Albertella, L., Myles, D., Yücel, M., & Rotaru, K. (2024) · JMIR Serious Games · doi:10.2196/50282 · Cited by 40
“Eye tracking during reading aids Alzheimer disease identification [96], and linguistic attributes (eg, formation and fluency of sentences, syntax, and grammar) distinguish patients with Alzheimer disease from those with MCI [97].”
BackgroundIdentifying functional cognitive disorder: a proposed diagnostic risk model
McWhirter, L., Ritchie, C., Stone, J., & Carson, A. (2022) · CNS Spectrums · doi:10.1017/s1092852921000845 · Cited by 40
García‐Gutiérrez, F., Marquié, M., Muñoz, N., Alegret, M., Cano, A. C., de Rojas, I., García‐González, P., Olivé, C., Puerta, R., Orellana, A., et al. (2023) · Frontiers in Neuroscience · doi:10.3389/fnins.2023.1221401 · Cited by 40
“Fraser et al. (2016) identified several voice abnormalities in speech related to ADD.”
BackgroundZimmerer, V. C., Hardy, C. J. D., Eastman, J., Dutta, S., Varnet, L., Bond, R. L., Russell, L. L., Rohrer, J. D., Warren, J. D., & Varley, R. A. (2020) · Cortex · doi:10.1016/j.cortex.2020.08.027 · Cited by 46
“…which in turn may support early identification of decline in language function and allow sensitive tracking of behavior change (Ash et al., 2013; Boschi et al., 2017; Fraser et al., 2014; Fraser, Meltzer, & Rudzicz, 2015; Nevler et al., 2017; Wilson et al., 2010; Zimmerer, Wibrow, & Varley, 2016).”
BackgroundRentoumi, 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
“A semantic impairment with increased use of repetitions and pronouns, and reduced use of lexical variety was detected, as well as a syntactic deficit in the production of auxiliary verbs, gerunds and participles [7].”
BackgroundIvanova, O., Martínez‐Nicolás, I., & Meilán, J. J. G. (2024) · International Journal of Language & Communication Disorders · doi:10.1111/1460-6984.12888 · Cited by 38
“Yet, many studies on AD also agree about significant impairments in language (cf. Fraser et al., 2016; Vincze et al., 2021).”
InfluentialSelf-Supervised ASR Models and Features for Dysarthric and Elderly Speech Recognition
Hu, S., Xie, X., Geng, M., Jin, Z., Deng, J., Li, G., Wang, Y., Cui, M., Wang, T., Meng, H. M. L., et al. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3422839 · Cited by 60
“Neurocognitive disorders, such as Alzheimer’s disease (AD), are often found among older adults and manifest themselves in speech and language impairments [15], [16].”
BackgroundAutomatic Hierarchical Attention Neural Network for Detecting AD
Pan, Y., Mirheidari, B., Reuber, M., Venneri, A., Blackburn, D. J., & Christensen, H. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-1799 · Cited by 47
“In [3] and [27], an accuracy of 81.92% and a F-score of 77.50% was achieved, compared with 84.02% for our system.”
BackgroundResultInfluentialThomas, J. A., Burkhardt, H. A., Chaudhry, S., Ngo, A. D., Sharma, S., Zhang, L., Au, R., & Hosseini Ghomi, R. (2020) · Journal of Alzheimer’s Disease · doi:10.3233/jad-190783 · Cited by 37
“This agrees with a report by Fraser et al. [56], which details that in factor analysis for pre-dictingADdiagnosis,onesignificantfactorwasmade upofseveralparametersdescribingacousticabnor-mality,includingskewnessandkurtosisofanumber of MFCCs.”
The INESC-ID Multi-Modal System for the ADReSS 2020 Challenge
Pompili, A., Rolland, T., & Abad, A. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2833 · Cited by 46
“[11] used more than 350 features to capture lexical, syntactic, grammatical, and semantic phenomena from the transcriptions of a picture description task.”
BackgroundMethodAl-Hameed, S., Benaissa, M., Christensen, H., Mirheidari, B., Blackburn, D., & Reuber, M. (2019) · PLOS ONE · doi:10.1371/journal.pone.0217388 · Cited by 37
“[59] and in our previous works in [14, 16]).”
ResultWang, C., Liu, S., Li, A., & Liu, J. (2023) · Journal of Medical Internet Research · doi:10.2196/51501 · Cited by 41
“Language analysis can be a valuable tool, but it is just one aspect of the overall assessment [33].”
Song, J., Cho, E., Lee, H., Lee, S., Kim, S., & Kim, J. (2025) · Biosensors · doi:10.3390/bios15020102 · Cited by 46
Idea density for predicting Alzheimer's disease from transcribed speech
Sirts, K., Piguet, O., & Johnson, M. (2017) · Conference on Computational Natural Language Learning · doi:10.18653/v1/k17-1033 · Cited by 46
“Note however that the BOW features are conceptually much simpler than the acoustic and lexicosyntac-tic features extracted by Yancheva and Rudzicz (2016) and Fraser et al. (2015).”
BackgroundMethodResultInfluentialDetecting 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
“[26],abaselinedetec-tionmodelisalsobuiltinthispaperusingfeatures atdifferentlinguisticlevels.”
InfluentialAutomatic speech analysis to early detect functional cognitive decline in elderly population
Ambrosini, E., Caielli, M., Milis, M., Loizou, C. P., Azzolino, D., Damanti, S., Bertagnoli, L., Cesari, M., Moccia, S., Cid, M., et al. (2019) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2019.8856768 · Cited by 45
“were able to distinguish between AD and healthy subjects with a classification accuracy of >81% [11].”
BackgroundVerbal fluency in normal aging and cognitive decline: Results of a longitudinal study
Frankenberg, C., Weiner, J., Knebel, M., Abulimiti, A., Toro, P., Herold, C. J., Schultz, T., & Schröder, J. (2021) · Computer Speech & Language · doi:10.1016/j.csl.2021.101195 · Cited by 44
Bose, A., Dash, N. S., Ahmed, S., Dutta, M., Dutt, A., Nandi, R., Cheng, Y., & Mello, T. M. D. (2021) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2021.707628 · Cited by 45
“…marker of language breakdown in AD is the over production of pronouns, such as he , she , they , it , rather than use of the specific name or nouns (March et al., 2006; Ahmed et al., 2013; Jarrold et al., 2014; Fraser et al., 2016 for English; Kavé and Levy, 2003; Kavé and Goral, 2016, for Hebrew).”
BackgroundResultInfluentialJang, H., Soroski, T., Rizzo, M., Barral, O., Harisinghani, A., Newton‐Mason, S. M., Granby, S., Vasco, T. D. C., Lewis, C., Tutt, P., et al. (2021) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2021.716670 · Cited by 52
“It has been used in prior work for AD prediction using language (Fraser et al., 2016; Karlekar et al., 2018; Kong et al., 2019).”
BackgroundMethodInfluentialYe, Z., Hu, S., Li, J., Xie, X., Geng, M., Yu, J., Xu, J., Xue, B., Liu, S., Liu, X., et al. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413634 · Cited by 51
Ivanova, O., Meilán, J. J. G., Martínez‐Sánchez, F., Martínez‐Nicolás, I., Llorente, T. E., & Carcavilla-González, N. (2021) · Computer Speech & Language · doi:10.1016/j.csl.2021.101341 · Cited by 60
Liu, Z., Guo, Z., Ling, Z., & Li, Y. (2021) · ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp39728.2021.9413566 · Cited by 34
“Fraser et al. [8] designed 370 features, including POS, semantic, acoustic and other linguistic features, which achieved an accuracy of 81% for two-class classification between AD and control (CTRL) subjects.”
BackgroundEvaluation of Linguistic Markers of Word-Finding Difficulty and Cognition in Parkinson's Disease
Smith, K. M., Ash, S., Xie, S. X., & Grossman, M. (2018) · Journal of Speech Language and Hearing Research · doi:10.1044/2018_jslhr-l-17-0304 · Cited by 43
Li, J., Yu, J., Ye, Z., Wong, K. H., Mak, M., Mak, B. K.-W., Liu, X., & Meng, H. M. L. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414147 · Cited by 35
Enhancing the classification of aphasia: a statistical analysis using connected speech
Fromm, D., Greenhouse, J. B., Pudil, M., Shi, Y., & MacWhinney, B. (2021) · Aphasiology · doi:10.1080/02687038.2021.1975636 · Cited by 41
“Machine learning approaches have begun to be used in other areas of discourse impairments such as Alzheimer’s disease (Fraser et al., 2016), mild cognitive impairment (Lundholm Fors et al.”
BackgroundAutomated text‐level semantic markers of Alzheimer's disease
Sanz, C., Carrillo, F., Slachevsky, A., Forno, G., Tempini, M. L. G., Villagra, R., Ibáñez, A. M., Tagliazucchi, E., & García, A. M. (2022) · Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring · doi:10.1002/dad2.12276 · Cited by 45
Detecting late-life depression in Alzheimer's disease through analysis of speech and language
Fraser, K., Rudzicz, F., & Hirst, G. (2016) · Workshop on Computational Linguistics and Clinical Psychology (CLPsych) · doi:10.18653/v1/w16-0301 · Cited by 45
“Orimaye et al. (2014) obtained F -measure scores up to 0.74 on transcripts from DementiaBank, combining participants with different etiologies rather than focusing on AD.”
BackgroundMethodInfluentialSelf-citationGiving Them a Voice: Challenges to Narrative Agency in People with Dementia
Villar, F., Serrat, R., & Bravo-Segal, S. (2019) · Geriatrics · doi:10.3390/geriatrics4010020 · Cited by 37
“, ‘to make’), errors in naming, or the use of pronouns with no clear referent [18,19].”
BackgroundScalable diagnostic screening of mild cognitive impairment using AI dialogue agent
Tang, F., Uchendu, I., Wang, F., Dodge, H. H., & Zhou, J. (2020) · Scientific Reports · doi:10.1038/s41598-020-61994-0 · Cited by 40
“Note that this is quite different from “speech markers” that involve auditory changes in pronunciations 10–12 which reflect early symptomatic changes in speech generation.”
BackgroundVoice Activated IoT Devices for Healthcare: Design Challenges and Emerging Applications
Spachos, P., Gregori, S., & Deen, M. J. (2022) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2022.3179680 · Cited by 46
“Recent studies include the identification of depression and post-traumatic stress disorder [38], [39], neurodegenerative diseases [40], [41], and metabolic and cardiovascular diseases [42], [43].”
BackgroundBiosensors toward behavior detection in diagnosis of alzheimer’s disease
Sun, X., Sun, X., Wang, Q., Wang, X., Feng, L., Yang, Y., Jing, Y., Yang, C., & Zhang, S. (2022) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2022.1031833 · Cited by 38
“Based on experiments, four linguistic features including semantic impairment, acoustic abnormality, syntactic impairment, and information impairment are proved to be useful in the recognition of AD, and the accuracy of this method reaches up to 81% (Fraser et al., 2015) (Figure 4III).”
MethodImproving Alzheimer's Disease Detection for Speech Based on Feature Purification Network
Liu, N., Yuan, Z., & Tang, Q. (2022) · Frontiers in Public Health · doi:10.3389/fpubh.2021.835960 · Cited by 34
Yamada, Y., Shinkawa, K., Nemoto, M., Nemoto, K., & Arai, T. (2023) · Computer Speech & Language · doi:10.1016/j.csl.2023.101514 · Cited by 38
Automatic speech analysis for detecting cognitive decline of older adults
Huang, L., Yang, H. J., Che, Y., & Yang, J. (2024) · Frontiers in Public Health · doi:10.3389/fpubh.2024.1417966 · Cited by 48
Enriching Neural Models with Targeted Features for Dementia Detection
Palo, F. D., & Parde, N. (2019) · Annual Meeting of the Association for Computational Linguistics: Student Research Workshop (ACL SRW) · doi:10.18653/v1/p19-2042 · Cited by 32
“Sentiment scores were obtained using NLTK’s sentiment library and psycholinguistic scores were obtained from an open source repository3 based on the work of Fraser et al. (2016).”
BackgroundMethodInfluentialParkinson’s Disease Detection from Speech Using Single Frequency Filtering Cepstral Coefficients
Kadiri, S. R., Kethireddy, R., & Alku, P. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-3197 · Cited by 36
“Finally, the studies belonging to the linguistic approach examine the use of vocabulary, phrase construction and repetition of words by Parkinsonian speakers [11, 12].”
BackgroundDEPAC: a Corpus for Depression and Anxiety Detection from Speech
Tasnim, M., Ehghaghi, M., Diep, B., & Novikova, J. (2022) · Workshop on Computational Linguistics and Clinical Psychology (CLPsych) · doi:10.18653/v1/2022.clpsych-1.1 · Cited by 56
Foltz, P. W., Chandler, C., Diaz-Asper, C. M., Cohen, A. S., Rodriguez, Z. B., Holmlund, T. B., & Elvevåg, B. (2022) · Schizophrenia Research · doi:10.1016/j.schres.2022.07.011 · Cited by 33
“For example, the development of features such as counts of words recalled may reveal slowing in executive function (e.g., Ghanavati et al., 2019), measures of temporal switching between clusters may be related to working memory capacity (Oh et al., 2019) and similarity between words may indicate…”
Ke, W., Crist, R. M., Clogston, J. D., Stern, S. T., Dobrovolskaia, M. A., Grodzinski, P. A., & Jensen, M. A. (2022) · Advanced Drug Delivery Reviews · doi:10.1016/j.addr.2022.114591 · Cited by 36
The Application of Artificial Intelligence in Alzheimer's Research
Zhao, Q., Xu, H., Li, J., Akhtar, F., & Qiao, L. (2023) · Tsinghua Science & Technology · doi:10.26599/tst.2023.9010037 · Cited by 42
Wang, N., Chen, M., & Subbalakshmi, K. P. (2020) · medRxiv · doi:10.1101/2020.06.24.20139592 · Cited by 38
Voleti, R., Woolridge, S. M., Liss, J., Milanovic, M., Bowie, C. R., & Berisha, V. (2019) · Interspeech (USB) · doi:10.21437/interspeech.2019-2960 · Cited by 34
Language-related features for early detection of Alzheimer Disease
Ammar, R. B., & BenAyed, Y. (2020) · Procedia Computer Science · doi:10.1016/j.procs.2020.09.071 · Cited by 36
Yamada, Y., Shinkawa, K., Kobayashi, M., Nishimura, M., Nemoto, M., Tsukada, E., Ota, M., Nemoto, K., & Arai, T. (2021) · Frontiers in Digital Health · doi:10.3389/fdgth.2021.653904 · Cited by 34
“Many studies suggested the MFCC is a good representation for detecting various types of diseases including depression, Parkinson’s disease, and AD (25, 57, 58).”
BackgroundMethodAustin, J., Hollingshead, K., & Kaye, J. A. (2017) · Journal of Medical Internet Research · doi:10.2196/jmir.7671 · Cited by 33
“Our results supported this hypothesis independent of years of formal education, which is consistent with previous work linking the decline in vocabulary to cognitive function [23].”
Identifying and Avoiding Confusion in Dialogue with People with Alzheimer's Disease
Chinaei, H., Currie, L. C., Danks, A., Lin, H., Mehta, T., & Rudzicz, F. (2017) · Computational Linguistics · doi:10.1162/coli_a_00290 · Cited by 30
“Much of this work is based on prior work in which we used similar linguistic features to train machine learning classifiers to distinguish between participants with and without AD (Fraser, Meltzer, and Rudzicz 2016).”
BackgroundMethodResultInfluentialSelf-citationMultivariate word properties in fluency tasks reveal markers of Alzheimer's dementia
Ferrante, F. J., Migeot, J., Birba, A., Amoruso, L., Pérez, G. N., Hesse, E., Tagliazucchi, E., Estienne, C. F., Serrano, C. M., Slachevsky, A., et al. (2023) · Alzheimer s & Dementia · doi:10.1002/alz.13472 · Cited by 38
“1,16,57 .”
BackgroundFrom privacy to anti-discrimination in times of machine learning
Hagendorff, T. (2019) · Ethics and Information Technology · doi:10.1007/s10676-019-09510-5 · Cited by 32
“can be investigated and inductive conclusions can be drawn about certain personality traits (Fraser et al. 2016; Haque et al. 2018).”
BackgroundKishimoto, T., Takamiya, A., Liang, K., Funaki, K., Fujita, T., Kitazawa, M., Yoshimura, M., Tazawa, Y., Horigome, T., Eguchi, Y., et al. (2020) · Contemporary Clinical Trials Communications · doi:10.1016/j.conctc.2020.100649 · Cited by 33
“Methods for automatically detecting and evaluating the severity of neurocognitive disorders include one based on acoustic characteristics [19, 36–40], one using linguistic information [41–43], and some fusing both those approaches [44, 45].”
MethodAdvances in Multimodal Behavioral Analytics for Early Dementia Diagnosis: A Review
Guruge, C. J. P., Oviatt, S. L., Haghighi, P. D., & Pritchard, E. (2021) · INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION · doi:10.1145/3462244.3479933 · Cited by 42
“The above cognitive and personality changes associated with dementia are common and prominent ones that are well documented in the literature, and these deficits clearly interact with one another [33], [74], [89], [92], [95], [107], [113], [114], [136], [137], [138], [139], [140], [141], [142], [143], [144].”
BackgroundRobotti, C. A., Costantini, G., Saggio, G., Cesarini, V., Calastri, A., Maiorano, E., Piloni, D., Perrone, T., Sabatini, U., Ferretti, V. V., et al. (2021) · Journal of Voice · doi:10.1016/j.jvoice.2021.11.004 · Cited by 35
“Age, median (IQR), years 57 (39-67) 59 (48-69) 41 (29-54) < 0.”
Liu, Z., Paek, E. J., Yoon, S. O., Casenhiser, D. M., Zhou, W., & Zhao, X. (2022) · Journal of Alzheimer s Disease · doi:10.3233/jad-215137 · Cited by 33
“Prior research has shown that NLP and ML techniques using various features from speech transcripts and their associated audio files can evaluate AD and differentiate people with AD from healthy control group [35–42].”
InfluentialYamada, Y., Shinkawa, K., & Shimmei, K. (2020) · JMIR Mental Health · doi:10.2196/16790 · Cited by 34
“As for the features related to perseveration, we found no significant difference in our dataset, while previous work reported the statistical significance of these features [22].”
ResultAn 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
“[15, 16] have used the DementiaBank corpus (containing speech of patients with AD, vascular dementia, MCI and healthy controls describing the ‘Cookie Theft' picture) to predict changes in patients' Mini Mental State Examination (MMSE) scores over time.”
MethodWang, N., Cao, Y., Hao, S., Shao, Z., & Subbalakshmi, K. P. (2021) · Interspeech (USB) · doi:10.21437/interspeech.2021-2024 · Cited by 56
“…We used two tools to generate linguistic features: 1) Like [2], we converted transcripts into CHAT format, then ran EVAL and FREQ commands in CLAN [31] to generate a composite profile of 34 measures and Moving Average Type Token Ratio [32]; 2) we generated 22 Part-of-Speech tags using NLTK [15].”
BackgroundMethodGarcía, A. M., Johann, F., Echegoyen, R. E., Calcaterra, C., Riera, P., Belloli, L. M. L., & Carrillo, F. (2023) · Behavior Research Methods · doi:10.3758/s13428-023-02240-z · Cited by 37
Detecting 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
Vincze, V., Szatlóczki, G., Tóth, L., Gosztolya, G., Pákáski, M., Hoffmann, I., & Kálmán, J. (2020) · Clinical Linguistics & Phonetics · doi:10.1080/02699206.2020.1827043 · Cited by 38
“It is known that AD influences the temporal characteristics of speech (F. F. Martínez-Sánchez et al., 2018; Fraser et al., 2016; Hoffmann et al., 2010; De Ipiña et al., 2015; Meilán et al., 2018, 2012, 2020; Tóth et al., 2018) and these phonetic changes occur even in mild AD (mAD) (Forbes-McKay & Venneri, 2005; Kato e…”
BackgroundMethodSpeech and language characteristics differentiate Alzheimer's disease and dementia with Lewy bodies
Yamada, Y., Shinkawa, K., Nemoto, M., Ota, M., Nemoto, K., & Arai, T. (2022) · Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring · doi:10.1002/dad2.12364 · Cited by 45
Wang, H.-L., Tang, R., Ren, R., Dammer, E. B., Guo, Q., Peng, G., Cui, H., Zhang, Y.-M., Wang, J., Xie, X., et al. (2022) · BMC Medicine · doi:10.1186/s12916-022-02584-x · Cited by 34
“In order to be consistent with the China RSF center and the previous study [18, 19], the samples with “Possible AD” and “Probable AD” labels are merged to compose the AD group in our study.”
ResultKhan, Y. F., Kaushik, B., Rahmani, M. K. I., & Ahmed, M. E. (2022) · IEEE Access · doi:10.1109/access.2022.3223681 · Cited by 33
“Automatic ML classification techniques have been used to differentiate between healthy and dementia patients with a standard accuracy of 81% [12].”
MethodWilliams, E., Theys, C., & McAuliffe, M. J. (2023) · PLoS ONE · doi:10.1371/journal.pone.0288556 · Cited by 33
“Interestingly, findings from picture description tasks have not indicated that pwAD rely on words of lower AoA [9, 20].”
BackgroundResultInfluentialDetecting Dementia through Retrospective Analysis of Routine Blog Posts by Bloggers with Dementia
Masrani, V., Murray, G., Field, T. S., & Carenini, G. (2017) · doi:10.18653/v1/w17-2329 · Cited by 28
“Our methods are similar to Fraser et al. (2015), with the main difference being the dataset used and their inclusion of audio and test-specific features, which are not available in our case.”
BackgroundMethodInfluentialOrimaye, S. O., Wong, S.-M. J., & Wong, C. P. (2019) · PLoS ONE · doi:10.1371/journal.pone.0214103 · Cited by 28
Automated assessment of speech production and prediction of MCI in older adults
Sanborn, V. E., Ostrand, R., Ciesla, J. A., & Gunstad, J. (2020) · Applied Neuropsychology Adult · doi:10.1080/23279095.2020.1864733 · Cited by 31
“…may be even more sensitive to AD risk than are traditional neuropsychological tests of language function (Bayles et al., 1992; Bucks et al., 2000; Fraser et al., 2016; Giles et al., 1996; Kav (cid:1) e & Dassa, 2018; Konig et al., 2018; Lopez-de-Ipina et al., 2018; Meilan, 2014; Meilan et al.,…”
BackgroundArtificial intelligence as an aid to diagnosing dementia: an overview
Dashwood, M., Churchhouse, G., Young, M., & Kuruvilla, T. (2021) · Progress in Neurology and Psychiatry · doi:10.1002/pnp.721 · Cited by 29
Polsinelli, A. J., Moseley, S., Grilli, M. D., Glisky, E. L., & Mehl, M. R. (2020) · The Journals of Gerontology Series B · doi:10.1093/geronb/gbaa055 · Cited by 29
Anthes, E. (2020) · Nature · doi:10.1038/d41586-020-02732-4 · Cited by 26
Hollands, S., Blackburn, D. J., & Christensen, H. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-10433 · Cited by 40
“Linguistic features have been found to be highly successful for the detection of dementia [17, 18] and indeed most classifiers use linguistic information as an feature set for cognitive impairment detection, some systems exclusively.”
BackgroundMazzon, G., Ajčević, M., Cattaruzza, T., Menichelli, A., Guerriero, M., Capitanio, S., Pesavento, V., Dore, F., Sorbi, S., Manganotti, P., et al. (2019) · Current Alzheimer Research · doi:10.2174/1567205016666190506141733 · Cited by 31
Metadata
- Authors
- Fraser, K. C., Meltzer, J. A., & Rudzicz, F.
- Affiliations
- University of Toronto (CA)
- Rotman Research Institute (CA)
- University Health Network (CA)
- Corresponding author
- Rudzicz, Frank
- Venue
- Journal of Alzheimer's Disease
- Publisher
- SAGE Publications
- Published
- 2015-11-27
- Volume
- 49
- Issue
- 2
- Pages
- 407-422
- ISSN
- 1387-2877 (ISSN-L)
- 1875-8908
- OpenAlex
- W1853705225
- Semantic Scholar
- bdd4aee4ea351e615eae0e4ab317bf974f1f731b
- Cited by
- 797
- References
- 98
- Citations by year
- 2026: 92 · 2025: 87 · 2024: 81 · 2023: 93 · 2022: 77 · 2021: 104 · 2020: 115 · 2019: 59 · 2018: 41 · 2017: 33 · 2016: 14
- FWCI
- 23.78
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Journal of Alzheimer s Disease (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Carnegie Mellon University
- Alzheimer Society
- National Institutes of Health (5p01ag003705-04)
- Medical Center, University of Pittsburgh
- Natural Sciences and Engineering Research Council of Canada
- National Institute on Aging (AG-005133, P50AG005133, AG003705)
- National Institute on Deafness and Other Communication Disorders (R01DC008524)
- Topics
- Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
- Life Sciences › Neuroscience › Cognitive Neuroscience › Neurobiology of Language and Bilingualism
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- Alzheimer's disease, language impairment
- MeSH terms
- Machine Learning, Aged, Aged, 80 and over, Alzheimer Disease, Diagnosis, Computer-Assisted, Factor Analysis, Statistical, Female, Humans, Language Disorders, Linguistics, Male, Mental Status Schedule, Middle Aged, Photic Stimulation, Speech, Verbal Behavior, Logistic Models, Narration
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
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.