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.
Citation: 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
Learning outcomes:
- Machine learning based on automatically extracted language features detected cognitive decline from early stages of the AD continuum in a new Spanish-Catalan dataset.
- Different speech and language domains showed differential discrimination performance between groups, with features extracted directly from speech performing better than those from the text.
- Before the onset of objective cognitive impairment, speech and language from older adults with Subjective Cognitive Decline (SCD) showed speech and language differences from controls without SCD, indicating potential heterogeneity in these non-clinical groups.
Abstract
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.
Discussion: 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.