ASHA journals
Browse
S1_AJSLP-22-00403He.pdf (721.21 kB)

Detect probably Alzheimer's disease dementia across subsections & language domains (He et al., 2023)

Download (721.21 kB)
online resource
posted on 2023-07-24, 15:53 authored by Rui He, Kayla Chapin, Jalal Al-Tamimi, Núria Bel, Marta Marquié, Maitee Rosende-Roca, Vanesa Pytel, Juan Pablo Tartari, Montse Alegret, Angela Sanabria, Agustín Ruiz, Mercè Boada, Sergi Valero, Wolfram Hinzen

Background: Decline in language has emerged as a new potential biomarker for the early detection of Alzheimer’s disease (AD). It remains unclear how sensitive language measures are across different tasks, language domains, and languages, and to what extent changes can be reliably detected in early stages such as subjective cognitive decline (SCD) and mild cognitive impairment (MCI).

Method: Using a scene construction task for speech elicitation in a new Spanish/Catalan speaking cohort (N = 119), we automatically extracted features across seven domains, three acoustic (spectral, cepstral, and voice quality), one prosodic, and three from text (morpholexical, semantic, and syntactic). They were forwarded to a random forest classifier to evaluate the discriminability of participants with probable AD dementia, amnestic and nonamnestic MCI, SCD, and cognitively healthy controls. Repeated-measures analyses of variance and paired-samples Wilcoxon signed-ranks test were used to assess whether and how performance differs significantly across groups and linguistic domains.

Results: The performance scores of the machine learning classifier were generally satisfactorily high, with the highest scores over .9. Model performance was significantly different for linguistic domains (p < .001), and speech versus text (p = .043), with speech features outperforming textual features, and voice quality performing best. High diagnostic classification accuracies were seen even within both cognitively healthy (controls vs. SCD) and MCI (amnestic and nonamnestic) groups.

Conclusion: Speech-based machine learning is powerful in detecting cognitive decline and probable AD dementia across a range of different feature domains, though important differences exist between these domains as well.

Supplemental Material S1. (A) The recruitment procedure, diagnostic criteria, neuropsychological battery, impact of the epidemic, and criteria for inclusion and exclusion. (B) Speech transcription instruction. (C) Feature list and definitions. (D) Detailed group-wise classifier performance from the random forest. (F) Post-hoc analysis of the RMANOVA tests. (F) Results and statistical comparison from Gradient Boosting.

He, R., Chapin, K., Al-Tamimi, J., Bel, N., Marquié, M., Rosende-Roca, M., Pytel, V., Tartari, J. P., Alegret, M., Sanabria, A., Ruiz, A., Boada, M., Valero, S., & Hinzen, W. (2023). Automated classification of cognitive decline and probable Alzheimer’s dementia across multiple speech and language domains. American Journal of Speech-Language Pathology, 32(5), 2075–2086. https://doi.org/10.1044/2023_AJSLP-22-00403

Funding

This research was supported by the Generalitat de Catalunya (Grant 2017SGR1265 awarded to W.H.), the Ministerio de Ciencia, Innovación y Universidades and the Agencia Estatal de Investigación (Grant PID2019-105241GB-I00/AEI/10.13039/501100011033 awarded to W.H.), the China Scholarship Council (Grant 202108390062 awarded to R.H.), and partially by the French Investissements d’Avenir - Labex Empirical Foundations of Linguistics program (ANR-10- LABX-0083 to the IdEx Université Paris Cité - ANR-18-IDEX-0001 awarded to J.A.T.). This study was also funded by Grifols, S.A.; Life Molecular Imaging; Laboratorios Echevarne; Araclon Biotech; and Ace Alzheimer Center Barcelona, and was supported by the Spanish Ministry of Health from Instituto de Salud Carlos III (Madrid; FISS PI10/00945) and by the Agència d’Avaluació de Tecnologia i Recerca Mèdiques. It was also funded by Departament de Salut de la Generalitat de Catalunya (Health Department of the Catalan Government; 390) Ace Alzheimer Center Barcelona is part of the Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (Spain) and is one of the participating centers of the Dementia Genetics Spanish Consortium. This study has also been funded by Instituto de Salud Carlos III (ISCIII) Acción Estratégica en Salud, integrated in the Spanish National Programa Estatal para Impulsar la Investigación Científico-Técnico y su Transferencia, Subprograma Estatal de Generación de Conocimiento Plan and financed by ISCIII-Subdirección General de Evaluación and the Fondo Europeo de Desarrollo Regional (FEDER-Una manera de hacer Europa) Grant PI19/00335 awarded to M.M.

History