Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Detecting Alzheimer's Disease from Continuous Speech Using Language Models.

Zhiqiang Guo¹, Zhenhua Ling¹, Yunxia Li²

¹National Engineering Laboratory for Speech and Language Information Processing, University of Science and Technology of China, Hefei, China.

Journal of Alzheimer'S Disease : JAD

|July 20, 2019

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Association of peripheral NLRP3 inflammasome activation with prevalent and incident chronic kidney disease in patients with obstructive sleep apnea: a prospective study.

Sleep & breathing = Schlaf & Atmung·2026

Same author

Atrophy of specific amygdala subregion precedes the conversion from mild cognitive impairment with neuropsychiatric symptoms to dementia.

International psychogeriatrics·2026

Same author

Clinical phenotypes and genetic mutation analysis of 45 neonatal-onset methylmalonic acidemia.

Frontiers in pediatrics·2026

Same author

Trends in the burden of low back pain attributable to occupational ergonomic factors among the global population aged 15-64 years (1990-2021): a study based on the GBD database.

Journal of health, population, and nutrition·2026

Same author

A Crown Ether-Derived Chiral Phosphoric Acid for Catalytic Asymmetric Synthesis of 3,4-Dihydro-2<i>H</i>-1,2,4-benzothiadiazine-1,1-dioxides.

The Journal of organic chemistry·2026

Same author

Discovery of High-Performance Extremophiles and Extremozymes Using Machine Learning and Structure-Based Clustering.

Environmental science & technology·2026

Same journal

Gut microbiota alterations in Alzheimer's disease and mild cognitive impairment: A systematic review and meta-analysis.

Journal of Alzheimer's disease : JAD·2026

Same journal

Stage-dependent relationship between sleep duration and cortical tau deposition in cognitively impaired individuals: A cross-sectional study.

Journal of Alzheimer's disease : JAD·2026

Same journal

Multilayer brain connectivity and long-term cognitive changes in individuals at risk for Alzheimer's disease.

Journal of Alzheimer's disease : JAD·2026

Same journal

The creation and verification of a detection model for mild cognitive impairment by employing eye-tracking and gait metrics.

Journal of Alzheimer's disease : JAD·2026

Same journal

Compartment-specific analysis reveals disrupted astrocytic calcium transients in anesthetized Alzheimer's disease mice.

Journal of Alzheimer's disease : JAD·2026

Same journal

<i>APOE</i> ε4-specific associations between agitation, cerebrospinal fluid biomarkers, and cognitive decline in cognitively impaired patients.

Journal of Alzheimer's disease : JAD·2026

See all related articles

Language models (LMs) improve Alzheimer's disease (AD) detection from speech by analyzing word choices. AD patients use simpler, less descriptive language compared to healthy individuals.

Area of Science:

Computational linguistics
Neuroscience
Speech analysis

Background:

Alzheimer's disease (AD) detection from speech is an active research area.
Existing studies often overlook language models (LMs) for linguistic feature extraction and lexical analysis.
Few studies investigate lexical-level differences between AD patients and healthy controls in speech.

Purpose of the Study:

Achieve state-of-the-art performance in automatic AD detection using N-gram LMs.
Utilize LMs to extract linguistic features for enhanced AD detection.
Identify and analyze lexical usage differences between AD patients and healthy individuals.

Main Methods:

Utilized the DementiaBank corpus (242 control, 256 AD samples).
Established baseline models using feature selection and five machine learning algorithms.

Keywords:

Geriatric assessment language machine learning statistical

Related Experiment Videos

Extracted perplexity features via LMs to develop enhanced detection models.

Investigated lexical differences using a proportion test on unigram probabilities.

Main Results:

Baseline model achieved 80.7% detection accuracy.
Accuracy increased to 85.4% with LM-derived perplexity features.
AD patients exhibited a tendency towards more general, less informative, and less precise word usage.

Conclusions:

LM-derived perplexity features enhance automatic AD detection from continuous speech.
Statistical N-gram LMs can effectively capture lexical-level differences in speech between AD patients and healthy controls.