Diagnostic utility of speech-based biomarkers in mild cognitive impairment: a systematic review and meta-analysis
Age and Ageing · 29.08.2025 · 54 (10)
DOI: 10.1093/ageing/afaf316 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Age and Ageing, online erschienen am 28.10.2025. Verfasst von Zahra Jafari und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Communication Sciences and Disorders, Dalhousie University, Halifax, Nova Scotia, Canada. Hauptthemen laut MEDLINE: Cognitive Dysfunction, Speech. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 91 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Among various tools developed for mild cognitive impairment (MCI) detection, analysing speech features is a non-invasive and cost-effective approach that shows promise for early detection. This review aimed to systematically synthesise and analyse current evidence on the diagnostic utility of speech-based biomarkers for identifying MCI.
Methods
A systematic review and meta-analysis were conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed, Scopus, Ovid Medline and PsycINFO databases were searched up to April 2025 without restrictions on language, article status or year.
Results
Of 4432 identified records, 54 peer-reviewed articles met the inclusion criteria. Fixed-effects meta-analyses showed pooled estimates of 80.0% ‘accuracy’ [95% confidence intervals (CI): 70.0%–89.0%, P < .001, n = 21], 78.0% ‘area under the curve’ (95% CI: 70.0%–86.0%, P < .001, n = 21), 80.0% ‘sensitivity’ (95% CI: 71.0%–90.0%, P < .001, n = 22), and 77.0% ‘specificity’ (95% CI: 65.0%–89.0%, P < .001, n = 15) in differentiating MCI from cognitively unimpaired (CU) individuals. Egger’s regression tests indicated no publication bias (P ≥ .299), and the I2 statistic revealed no heterogeneity across studies (I2 = 0.00%, P = 1.00). Four studies also included a subjective cognitive decline group, reporting significant differences in certain speech features compared to CU.
Conclusions
Speech analysis demonstrates moderate classification performance, with balanced sensitivity and specificity, in distinguishing MCI from CU, suggesting its potential as an accurate and cost-effective diagnostic tool for MCI detection. Further research is needed to address variations in study methodologies, refine speech analysis protocols and validate findings in diverse populations to enhance generalisability.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- School of Communication Sciences and Disorders, Dalhousie University, Halifax, Nova Scotia, Canada.
- Geriatric Medicine, Dalhousie University, Halifax, Nova Scotia, Canada.
- Otolaryngology - Head & Neck Surgery, Dalhousie University, Halifax, Nova Scotia, Canada.
- Psychology and Neuroscience, Dalhousie University, Halifax, Nova Scotia, Canada.
ORCID-Profile: Zahra Jafari, Melissa K Andrew
Zitieren
APA
Jafari, Z., Andrew, M. K., & Rockwood, K. J. (2025). Diagnostic utility of speech-based biomarkers in mild cognitive impairment: a systematic review and meta-analysis. Age and Ageing, 54(10). https://doi.org/10.1093/ageing/afaf316
Vancouver
Jafari Z, Andrew MK, Rockwood KJ. Diagnostic utility of speech-based biomarkers in mild cognitive impairment: a systematic review and meta-analysis. Age and Ageing. 2025;54(10). doi:10.1093/ageing/afaf316
BibTeX
@article{Jafari2025,
author = {Jafari, Zahra and Andrew, Melissa K and Rockwood, Kenneth J},
title = {Diagnostic utility of speech-based biomarkers in mild cognitive impairment: a systematic review and meta-analysis},
journal = {Age and Ageing},
year = {2025},
volume = {54},
number = {10},
doi = {10.1093/ageing/afaf316}
}
EUVIMED