Smartwatch-derived acoustic markers for deficits in cognitively relevant everyday functioning

Abstract

Detecting subtle deficits in everyday functioning due to cognitive impairments is important for early detection of neurodegenerative diseases, particularly Alzheimer’s disease. However, current standards for the assessment of everyday functioning are based on qualitative, subjective ratings. Speech has been shown to be good objective markers for cognitive impairments, but the association with cognitive-relevant everyday functioning remains uninvestigated. In this study, we demonstrate the feasibility of smartwatch-based application using acoustic features as objective markers for detecting deficits in everyday functioning. To this end, we collected voice data during performing cognitive tasks and daily conversation, as possible application scenarios, from 54 older adults along with a measure of everyday functioning. Machine-learning model using acoustic features can detect individuals with deficits in everyday functioning with up to 77.8% accuracy, which was higher than that of 63.0% using a standard neuropsychological test score. We also identified common acoustic features discriminating for deficits in everyday functioning robustly across both types of voice data during the cognitive tasks and daily conversation. Our results suggest that common acoustic features extracted from different types of voice data can be used as markers for deficits in everyday functioning.