Walking path images from real-time location data predict degree of cognitive impairment

Tamim Faruk, Leia C. Shum, Andrea Iaboni, Shehroz S. Khan

Abstract: Highlights•First study to classify degree of cognitive impairment using spatial walking path images•Images are made from real-time location data split into windows of variable length.•Our method significantly outperforms prior methods that rely on domain knowledge.•We apply GradCAM to produce an objective metric that distinguishes between classes.