Early detection of Alzheimer's Disease (AD) is essential for improving patient outcomes, and analyzing driving performance may serve as an effective early detection tool.
This study uses driving simulator data to classify AD patients and controls, finding that machine learning algorithms, particularly the random forest classifier, can accurately differentiate between the two groups.
Key driving features that indicate AD include Pothole Avoidance, Road Signs Recalled, Inattention Measurements, Reaction Time, and Detection Times, all of which correlate with cognitive functions commonly impaired in AD patients.