Naturalistic Driving Detects Mild Cognitive Impairment
Researchers at Arizona State University and collaborators in a 2026 JMIR Med Inform paper develop deep learning models that use naturalistic in-vehicle GPS, accelerometer, and gyroscope data to detect mild cognitive impairment (MCI). In a study of 22 participants, full-trip models achieved 78% accuracy and 77% AUC, outperforming turn-only inputs; the authors propose a frequency-based risk score for interpretable screening. Findings support scalable, noninvasive early cognitive monitoring.
Scoring Rationale
Solid peer-reviewed application using naturalistic driving data, demonstrating practical accuracy, but small sample size and limited cohort constrain generalizability.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
