Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.
Who will be affected?
Readers: No access to all 28 journals. We recommend accessing our articles via PubMed Central
Authors: No access to the submission form or your user account.
Reviewers: No access to your user account. Please download manuscripts you are reviewing for offline reading before Wednesday, July 01, 2020 at 7:00 PM.
Editors: No access to your user account to assign reviewers or make decisions.
Copyeditors: No access to user account. Please download manuscripts you are copyediting before Wednesday, July 01, 2020 at 7:00 PM.
Pérez-Toro PA, Ferrante F, Pérez G, Tee BL, de Leon J, Nöth E, Schuster M, Maier A, Slachevsky A, Gorno-Tempini ML, Ibáñez A, Orozco-Arroyave JR, García A
Automated Speech Markers of Alzheimer Dementia: Test of Cross-Linguistic Generalizability
Automated speech markers of Alzheimer’s dementia: A test of cross-linguistic generalizability
Paula Andrea Pérez-Toro;
Franco Ferrante;
Gonzalo Pérez;
Boon Lead Tee;
Jessica de Leon;
Elmar Nöth;
Maria Schuster;
Andreas Maier;
Andrea Slachevsky;
Maria Luisa Gorno-Tempini;
Agustín Ibáñez;
Juan Rafael Orozco-Arroyave;
Adolfo García
ABSTRACT
Background:
Automated speech and language analysis can reveal early markers of Alzheimer’s disease (AD). However, most studies target uninterpretable features in Anglophone samples only, overlooking clinical utility and cross-linguistic validity.
Objective:
Here we aimed to tackle both gaps by examining whether interpretable speech and language markers of AD generalize between two typologically different languages.
Methods:
We analyzed semi-spontaneous recordings from 211 participants, including English- and Spanish-speaking patients and controls. We automatically extracted speech timing features (indicating semantic memory retrieval effort) and vocabulary selection features (capturing semantic memory navigation patterns). Classifiers trained on English-speaking participants were tested in a within-language setting (involving English-speaking patients and controls) and in a between-language setting (involving Spanish-speaking patients and controls).
Results:
AD detection was maximal upon combining timing and vocabulary features in the within-language setting (AUC = .88) and when considering timing features in the between-language setting (AUC = .75). Timing features also predicted cognitive outcomes in both settings.
Conclusions:
Automated proxies of semantic memory processes can yield interpretable AD markers within and across languages.
Citation
Please cite as:
Pérez-Toro PA, Ferrante F, Pérez G, Tee BL, de Leon J, Nöth E, Schuster M, Maier A, Slachevsky A, Gorno-Tempini ML, Ibáñez A, Orozco-Arroyave JR, García A
Automated Speech Markers of Alzheimer Dementia: Test of Cross-Linguistic Generalizability