TRB Pubsindex
Text Size:
x

« Prev Next »

Return to Results | New Search

Bookmark and Share

Title:

Can we infer dementia from driving behaviors?

Accession Number:

01764251

Record Type:

Component

Abstract:

Emerging evidence suggests that atypical changes in driving behaviors may be early signals of mild cognitive impairment (MCI) and dementia. So far, research linking driving behavior changes and the risk of MCI/dementia is limited to a few pilot studies with small sample sizes and short duration of follow-up. The proposed study aims to test the above stated hypothesis using GPS data (collected through the DataLogger device) from the Longitudinal Research on Aging Drivers (LongROAD) project. This study contributes to the existing evidence about the relationship between changes in driving behaviors, space and performance and the risk of MCI and dementia. The artificial neural network (ANN) based classification algorithm developed in this study demonstrates the feasibility of using naturalistic driving data for early detection of MCI and dementia. Using decision trees, the authors perform the impurity-based feature importance ranking of aggregate driving behavior variables. Age is the most important feature, followed by the total number of miles driven and the number of left turns made in months. Based on such ranking, the authors train two ANN based classifiers: one using age as the sole input feature, and the other using age and driving behavior features. The latter classifier produces much better performance measured by F1-score, indicating the potential link between driving behavior changes and the risk of MCI/dementia.

Supplemental Notes:

This paper was sponsored by TRB committee AED50 Standing Committee on Artificial Intelligence and Advanced Computing Applications.

Report/Paper Numbers:

TRBAM-21-03247

Language:

English

Corporate Authors:

Transportation Research Board

Authors:

Shi, Rongye
Di, Xuan
DiGuiseppi, Carolyn
Eby, David W
Hill, Linda L
Mielenz, Thelma J
Molnar, Lisa J
Strogatz, David
Goldberg, Terry E
Li, Guohua

Pagination:

17p

Publication Date:

2021

Conference:

Transportation Research Board 100th Annual Meeting

Location: Washington DC, United States
Date: 2021-1-5 to 2021-1-29
Sponsors: Transportation Research Board; Transportation Research Board

Media Type:

Web

Features:

Figures; References; Tables

Subject Areas:

Highways; Safety and Human Factors

Source Data:

Transportation Research Board Annual Meeting 2021 Paper #TRBAM-21-03247

Files:

TRIS, TRB, ATRI

Created Date:

Dec 23 2020 11:23AM