MétaCan
Menu
← Back to cohort
Record W7119467929 · doi:10.1002/alz70856_106013

Digital Voice as an Alternative Screening Tool to the Montreal Cognitive Assessment

2025· article· en· W7119467929 on OpenAlexaboutno aff
Hamzah Anan, Amjad Ajam, Abdulrazzaq Qattea, Xavier Serrano, Edward Searls, Kristi Ho, Zexu Li, Alexa Burk, Margaret Low, Owen Tan, Chenglin Lyu, Eric G. Steinberg, Jesse Mez, Michael L Alosco, Katherine A. Gifford, Vijaya B. Kolachalama, Honghuang Lin, R. Au, Huitong Ding

Bibliographic record

VenueAlzheimer s & Dementia · 2025
Typearticle
Languageen
FieldMedicine
TopicDementia and Cognitive Impairment Research
Canadian institutionsnot available
Fundersnot available
KeywordsMontreal Cognitive AssessmentCognitionAssociation (psychology)Range (aeronautics)Feature (linguistics)DemographicsCognitive impairment

Abstract

fetched live from OpenAlex

Abstract Background The Montreal Cognitive Assessment (MoCA) is a commonly used screening tool for cognitive impairment. Despite translation into multiple languages to facilitate broader use globally, there are inherent education and cultural biases that result in variations in cognitive screening accuracies. Acoustic voice features are emerging as a more education, language and culturally agnostic indicator of cognitive status, but as a surrogate to the MoCA has not been adequately explored. This pilot study aimed to examine the association between spectral acoustic features extracted by two functionals and MoCA total scores. Method We included 80 participants from the Boston University Alzheimer's Disease Research Center, whose responses to a picture‐description task were digitally recorded and administered the MoCA. Using openSMILE, each recording was divided into 20‐ms frames, using a sliding window that advanced by 10 ms for every segment. For each segment, we extracted 26 RASTA‐style filtered auditory spectrum (bands 1‐26) low‐level descriptors (LLD). Two functionals were applied to the 26 LLD to summarize information across each recording: the mean (mean value of LLD for all segments in each recording) and “upleveltime75” (percentage of time the signal exceeds 75% of the feature range above the minimum). Linear regression models were used to assess the association of these acoustic features with MoCA total scores, adjusting for age, sex, and education. Result Participant demographics included age: mean 68.7, SD 9.27 years; 81.25% college graduate or higher; 60.0% women. Among these participants, 29 were cognitively impaired. The mean MoCA score was 26.49 (SD = 2.54). Among the 26 upleveltime75 functional‐based spectral features, 5 showed significant negative associations with MoCA. The strongest effect was observed for band 25 (beta = –0.77, SE = 0.26, p = 0.0038). In contrast, no significant associations were observed for the acoustic features generated by the mean functional. Conclusion These results suggest that digital voice contains cognitive‐related signals and show promise as a potential globally appropriate cognitive screening tool given the ease in which it can be collected and the availability of automated open‐source tools for analysis. Further exploration analyzing voice recordings across different languages and cultural/education strata are warranted.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.006
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.005
Threshold uncertainty score0.016

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.006
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0020.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0000.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0050.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.029
GPT teacher head0.360
Teacher spread0.330 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueAlzheimer s & Dementia→Same topicDementia and Cognitive Impairment Research→French-language works237,207→