Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Feb 2, 2021
Date Accepted: Mar 15, 2021
Predicting Future Accident Risks of Older Drivers by Speech Data during Interacting with a Voice Assistant
ABSTRACT
Background:
With the rapid growth in the elderly population worldwide, driving accidents involving older adults have become an increasingly serious social problem. Cognitive impairment was reported as a risk factor relevance to car accidents and cognitive test scores were used as predictors. However, it remains unclear whether and how car accident risks can be predicted by using daily behavioral data.
Objective:
The objective of this study was to investigate whether speech data that can be collected in everyday life could be used for predicting accident risks of older drivers.
Methods:
In this study, we collected speech data from 60 older adults during interactions with a voice assistant in addition to cognitive assessments including neuropsychological tests, and followed up 1.5 years later by using a questionnaire about their driving experiences related to car accidents.
Results:
We found that older drivers with accident/near-accident experiences had statistically discernible changes in speech features implying cognitive impairments including reduced speech rate and increased response time. Moreover, the model using speech features could predict future accident/near-accident experiences with 81.7% accuracy, 6.7% higher than that using cognitive assessment data, and could achieve up to 88.3% accuracy when combining both data.
Conclusions:
Our study provides the first empirical results suggesting that speech data recorded during the interaction with voice assistants could help predict future accident risks of older drivers by capturing subtle changes in cognitive function.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.