⚠ Evidence note: Exercise has the strongest human evidence base in this vault, but protocol-level specifics are far less settled than the headline benefit.
Dual-Task Gait Cognitive Screening
Current understanding
To be populated as evidence accumulates.
Evidence log
- 2026-05-30 — Development of Machine Learning-Driven Dual-Task Gait Test Model for Cognitive Impairment Screening.: Community-dwelling adults ≥60 cross-sectional study built an ML model on dual-task gait parameters (serial subtraction, animal naming) reporting AUC 0.883 for dementia and 0.895 for cognitive impairment, with gait speed and stride variability as top predictors — initial accuracy evidence for scalable AI-aided dual-task gait as a primary-care cognitive screen; cross-sectional, no external validation reported. [human-observational] (novelty: 3)
Open questions
Related
Contributing findings
Development of Machine Learning-Driven Dual-Task Gait Test Model for Cognitive Impairment Screening.
30-may-2026
novelty 3
per-area 3
strengthens