[A study of cognitive impairment quantitative assessment method based on gait characteristics]

Area of Science:

• Neurology
• Biomedical Engineering
• Gerontology

Context:

• Alzheimer's disease (AD) is a leading cause of dementia in the elderly.
• Mild cognitive impairment (MCI) can precede dementia, and early intervention may slow progression.
• Emerging research links cognitive function with motor and gait abnormalities.

Purpose:

• To investigate the relationship between gait parameters and cognitive function in individuals with and without MCI.
• To develop a machine learning model for quantitatively evaluating cognitive levels using gait features.
• To identify specific gait markers predictive of cognitive status.

Summary:

• A study involving 193 participants (137 with MCI, 56 healthy controls) analyzed gait parameters during single and dual tasks using wearable devices.
• A machine learning model, employing recursive feature elimination (RFE), was built using gait cycle, kinematics, and spatio-temporal parameters to predict MoCA scores.
• Temporal and spatial parameters related to toe-off and heel strike were identified as significant predictors of cognitive level.

Impact:

• Gait analysis, particularly temporal and spatial parameters of toe-off and heel strike, shows clinical significance for evaluating cognitive function.
• These findings suggest potential for developing non-invasive tools for early AD detection and intervention.
• This research highlights the clinical application value of gait features in preventing or delaying AD onset.