A portable and efficient dementia screening tool using eye tracking machine learning and virtual reality.

Area of Science:

• Neuroscience
• Artificial Intelligence
• Medical Technology

Background:

• Dementia, particularly Alzheimer's Disease, poses a global health challenge.
• Current diagnostic methods for preclinical dementia and mild cognitive impairment (MCI) are often invasive, costly, or unsuitable for large-scale screening.
• Eye-tracking technology offers a non-invasive method to detect subtle cognitive changes.

Purpose of the Study:

• To introduce and evaluate the VR Eye-tracking Cognitive Assessment (VECA) tool for non-invasive, efficient, and large-scale dementia screening.
• To assess VECA's ability to predict cognitive impairment using eye-tracking data, machine learning, and virtual reality.
• To compare VECA's performance against traditional cognitive assessments.

Main Methods:

• Developed and deployed the VR Eye-tracking Cognitive Assessment (VECA) tool, integrating eye-tracking, machine learning, and virtual reality.
• Collected eye-tracking data from 201 participants at Shenzhen Baoan Chronic Hospital.
• Utilized a support vector regression model to predict Montreal Cognitive Assessment (MoCA) scores and classify cognitive impairment.

Main Results:

• The support vector regression model achieved a high correlation (0.9) with MoCA scores, outperforming baseline models.
• Established optimal cut-off scores for identifying cognitive impairment with 88.5% sensitivity and 83% specificity.
• Demonstrated VECA's efficacy across participants with varying educational backgrounds.

Conclusions:

• VECA shows significant potential as a portable and efficient tool for early dementia screening.
• The integration of eye-tracking, machine learning, and VR offers a promising approach for cognitive health assessments.
• VECA provides a viable, non-invasive alternative for large-scale cognitive impairment detection.