Inertia sensor based wearables can help predict cognitive disorders

A new study published in Neurology suggests that wearable inertia sensor-based gait-based models may be a viable diagnostic indicator of cognitive disorders (CD) in older persons.

Potential indicators of cognitive problems include alterations in gait. Using gait speed and variability data from a wearable inertia sensor, Jeongbin Park and colleagues created a model for separating older adults with CD from those with normal cognition. They then compared the model's CD diagnostic performance to that of the model using the Mini-Mental State Examination (MMSE).

In order to measure the gait characteristics of community-dwelling older adults with normal gait, researchers recruited them from the Korean Longitudinal Study on Cognitive Ageing and Dementia. They walked on a 14-m-long walkway three times at comfortable paces while wearing a wearable inertia sensor placed at the center of their body mass. The dataset was divided at random into the validation (20%) and development (80%) datasets. Using logistic regression analysis on the development dataset, a model was created for categorizing CD, and it was then verified on the validation dataset. The model's diagnostic performance was compared to that of the MMSE in both datasets. Utilizing receiver operator characteristics research, we derived the optimal cutoff score for our model.