Driving behaviors harbor early indicators of dementia

April 28, 2021 — Using naturalistic driving knowledge and machine studying strategies, researchers at Columbia University Mailman School of Public Health and Columbia’s Fu Foundation School of Engineering and Applied Science have developed extremely correct algorithms for detecting delicate cognitive impairment and dementia in older drivers. Naturalistic driving knowledge confer with knowledge captured by means of in-vehicle recording units or different applied sciences within the real-world setting. These knowledge might be processed to measure driving publicity, house and efficiency in nice element. The findings are revealed within the journal Geriatrics.

The researchers developed random forests fashions, a statistical method extensively utilized in AI for classifying illness standing, that carried out exceptionally properly. “Based on variables derived from the naturalistic driving data and basic demographic characteristics, such as age, sex, race/ethnicity and education level, we could predict mild cognitive impairment and dementia with 88 percent accuracy, “stated Sharon Di, affiliate professor of civil engineering and engineering mechanics at Columbia Engineering and the examine’s lead writer.

The investigators constructed 29 variables utilizing the naturalistic driving knowledge captured by in-vehicle recording units from 2977 contributors of the Longitudinal Research on Aging Drivers (LongROAD) challenge, a multisite cohort examine sponsored by the AAA Foundation for Traffic Safety. At the time of enrollment, the contributors have been lively drivers aged 65-79 years and had no vital cognitive impairment and degenerative medical situations. Data used on this examine spanned the time interval from August 2015 by means of March 2019.

Among the 2977 contributors whose vehicles have been instrumented with the in-vehicle recording units, 33 have been newly recognized with delicate cognitive impairment and 31 with dementia by April 2019. The researchers skilled a sequence of machine studying fashions for detecting delicate cognitive impairment/dementia and located that the mannequin based mostly on driving variables and demographic traits was 88 % correct, significantly better than fashions based mostly on demographic traits solely (29 %) and driving variables solely (66 %). Further evaluation revealed that age was most predictive of delicate cognitive impairment and dementia, adopted by the proportion of journeys traveled inside 15 miles of dwelling, race/ethnicity, minutes per journey chain (i.e., size of journeys beginning and ending at dwelling), minutes per journey, and variety of onerous braking occasions with deceleration charges ≥ 0.35 g.

“Driving is a complex task involving dynamic cognitive processes and requiring essential cognitive functions and perceptual motor skills. Our study indicates that naturalistic driving behaviors can be used as comprehensive and reliable markers for mild cognitive impairment and dementia,” stated Guohua Li, MD, DrPH, professor of epidemiology and anesthesiology at Columbia Mailman School of Public Health and Vagelos College of Physicians and Surgeons, and senior writer. “If validated, the algorithms developed in this study could provide a novel, unobtrusive screening tool for early detection and management of mild cognitive impairment and dementia in older drivers.”


Co-authors are Carolyn DiGuiseppi, Colorado School of Public Health; David W. Eby and Lisa Molnar, University of Michigan Transportation Research Institute; Linda Hill, University of California San Diego School of Public Health; Thelma J. Mielenz, Columbia Mailman School of Public Health; David Strogatz, Bassett Research Institute; Howard Andrews, Terry Goldberg, Barbara Lang, and Minjae Kim, Columbia Vagelos College of Physicians and Surgeons.

The examine was supported by the AAA Foundation for Traffic Safety.

Columbia University Mailman School of Public Health

Founded in 1922, the Columbia University Mailman School of Public Health pursues an agenda of analysis, schooling, and repair to handle the vital and complicated public well being points affecting New Yorkers, the nation and the world. The Columbia Mailman School is the seventh largest recipient of NIH grants amongst colleges of public well being. Its almost 300 multi-disciplinary college members work in additional than 100 nations around the globe, addressing such points as stopping infectious and power illnesses, environmental well being, maternal and youngster well being, well being coverage, local weather change and well being, and public well being preparedness. It is a pacesetter in public well being schooling with greater than 1,300 graduate college students from 55 nations pursuing quite a lot of grasp’s and doctoral diploma applications. The Columbia Mailman School can be dwelling to quite a few world-renowned analysis facilities, together with ICAP and the Center for Infection and Immunity. For extra info, please go to http://www.mailman.columbia.edu.

Columbia Engineering, based mostly in New York City, is among the prime engineering colleges within the U.S. and one of many oldest within the nation. Also referred to as The Fu Foundation School of Engineering and Applied Science, the School expands data and advances know-how by means of the pioneering analysis of its greater than 220 college, whereas educating undergraduate and graduate college students in a collaborative setting to turn out to be leaders knowledgeable by a agency basis in engineering. The School’s college are on the heart of the University’s cross-disciplinary analysis, contributing to the Data Science Institute, Earth Institute, Zuckerman Mind Brain Behavior Institute, Precision Medicine Initiative, and the Columbia Nano Initiative. Guided by its strategic imaginative and prescient, “Columbia Engineering for Humanity,” the School goals to translate concepts into improvements that foster a sustainable, wholesome, safe, related, and artistic humanity.

Disclaimer: AAAS and EurekAlert! aren’t chargeable for the accuracy of reports releases posted to EurekAlert! by contributing establishments or for using any info by means of the EurekAlert system.


Please enter your comment!
Please enter your name here