Lexical Speech Features of Spontaneous Speech in Older Persons With and Without Cognitive Impairment: Reliability Analysis

Connected speech
DOI: 10.2196/46483 Publication Date: 2023-10-11T12:38:31Z
ABSTRACT
Abstract Background Speech analysis data are promising digital biomarkers for the early detection of Alzheimer disease. However, despite its importance, very few studies in this area have examined whether older adults produce spontaneous speech with characteristics that sufficiently consistent to be used as proxy markers cognitive status. Objective This preliminary study seeks investigate consistency across lexical and without impairment. Methods A total 39 from a larger, ongoing (age: mean 81.1, SD 5.9 years) were included. Participants completed neuropsychological testing both picture description tasks expository elicit speech. T-scores ≤40 on ≥2 tests categorized having mild impairment (MCI). features computed automatically by using Python Natural Language Toolkit. Results Reliability indices based correlations similar persons MCI (with r ranging 0.49 0.65 within tasks). Intraindividual variability was generally preserved features. rate filler most cognitively intact group, group. Conclusions Our findings suggest calculated properties varying levels These encourage further investigation utility other monitoring status over time.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (40)
CITATIONS (2)