JMIR Publications

ABSTRACT

Background:

The prevalence of dementia is expected to rise with an aging population, necessitating accessible early detection methods. Serious games have emerged as potential cognitive screening tools. They provide not only an engaging platform for assessing cognitive function but also serve as valuable indicators of cognitive health through engagement levels observed during play.

Objective:

This study aims to examine the differences in engagement-related behaviours between older adults with and without dementia during serious gaming sessions. Further, it seeks to identify the key contributors that enhance the effectiveness of Machine Learning (ML) for dementia classification based on engagement-related behaviours.

Methods:

Over eight weeks, fifteen older adults (10 cognitively intact, 5 with Alzheimer’s) participated in a serious game intervention. Their engagement-behaviours were analyzed, with 1774 data points categorized into 47 behaviour codes, augmented by 54 additional features including personal characteristics and environmental factors. Codes underwent one-hot encoding and were processed using Random Forest classifiers to distinguish between participant groups.

Results:

The analysis revealed a 64% difference in engagement behaviours between the groups. Key variations were noted in torso movements, voice modulation, facial expressions, concentration levels, and age. Furthermore, the integration of engagement behaviours, environmental disturbances, technical issues, and personal features, into the ML models resulted in more robust and accurate dementia prediction.

Conclusions:

Engagement-related behaviours observed during serious gaming offer crucial markers for identifying dementia. ML models that incorporate these unique behavioural markers present a promising, non-invasive approach for early dementia screening in a variety of settings.