Enhancing road safety: In-vehicle sensor analysis of cognitive impairment in older drivers
DOI: 10.1007/s11042-024-19833-1
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study addresses the challenge of detecting mild cognitive impairment (MCI) in older drivers through unobtrusive, real-world monitoring. As the global population ages, distinguishing between normal age-related decline and pathological cognitive deficits is critical for road safety. Traditional clinical evaluations are limited in scope and frequency, often failing to capture subtle changes in daily functioning. The authors propose that driving performance, a complex activity requiring significant cognitive resources, serves as a reliable indicator of cognitive status. The research aims to develop low-cost, programmable in-vehicle sensing hardware capable of collecting high-precision telematics data to identify early warning signs of MCI in naturalistic driving conditions. The methodology involved a longitudinal study conducted at Florida Atlantic University and the University of Central Florida, involving 236 participants aged 65 and older with valid driver’s licenses. Participants were categorized based on clinical assessments for MCI. The researchers installed custom-built telematics units (TMUs) in the vehicles of over 150 drivers. These units, based on Raspberry Pi hardware, integrated GPS sensors, inertial measurement units (IMUs), and On-Board Diagnostics (OBD) connectors to collect data from the vehicle’s Controller Area Network (CAN bus). Over a period of three years, the system recorded 7,794 data points across 19 driver behavior indexes, including trip frequency, duration, speed, harsh acceleration, braking, and turning events. Data preprocessing involved quantile normalization to handle outliers. The analysis compared driving patterns between drivers with and without MCI and employed Random Forest machine learning models to predict MCI status using various subsets of demographic and driving variables. The results revealed that drivers with MCI exhibited smoother and safer driving patterns compared to those without impairment, suggesting they are cognizant of their condition and self-regulate to avoid erratic behaviors. Statistical analysis showed that participants predominantly drove during daylight hours, with half of all trips occurring in the afternoon. In terms of predictive modeling, the Random Forest algorithm achieved an accuracy of 0.86 and an AUC of 0.95 when using all available variables. Feature importance analysis identified the number of night trips, total number of trips, and education level as the most influential factors in distinguishing MCI status. Models relying solely on age or driving kinematics performed less effectively than those incorporating trip frequency and demographic data. The significance of this work lies in its demonstration that unobtrusive in-vehicle sensing combined with machine learning can effectively identify early signs of cognitive decline. By leveraging real-world driving data, this approach offers a scalable solution for monitoring older drivers’ fitness-to-drive, potentially enabling timely interventions such as adaptive driving assistance or alerts. The findings support the development of intelligent systems that can respond to changing cognitive states in real-time, thereby enhancing road safety and prolonging the independence of older adults. The study highlights the potential of passive monitoring technologies to bridge the gap between clinical diagnosis and daily functional assessment.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via author_sweep_intake on 2026-05-27.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-27 |
| archive | success | canonical_url | — | — | 35 | 2026-06-09 |
| extract | success | pdftotext | — | — | 2 | 2026-06-09 |
| clean | success | clean | — | — | 1 | 2026-06-09 |
| chunk | success | chunk | — | — | 1 | 2026-06-09 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-09 |
| enrich | failed | — | — | — | 4 | 2026-07-02 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-09 |
| tag | success | vector_similarity | — | — | 8 | 2026-06-11 |
| verify | success | — | — | — | 1 | 2026-06-09 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-09; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- mci dementia driving
- cognitive impairment
- telematics crash prediction
- exposure measurement
- older drivers
- cognitive capacity variation
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: validation psychometrics, tool software
- Theoretical Contribution: computational model