Advancing Mild Cognitive Impairment Detection: Integrating VR, MRI, and Neuropsychological Insights for Comprehensive Diagnosis

Abstract

Identifying mild cognitive impairment (MCI) early is key to averting its progression into dementia, as MCI marks the intermediate phase between normal cognitive aging and dementia. Traditionally, MCI diagnosis has leaned on biomarkers such as magnetic resonance imaging (MRI) and neuropsychological evaluations. Yet, these methods come with their own set of challenges, sparking a growing interest in the use of virtual reality (VR) as a novel diagnostic tool. VR's ability to harness behavioral data from daily interactions has shown promise in accurately identifying MCI, although its practical application in clinical settings remains a topic of debate. This study delves into the relationship between VR and traditional biomarkers and assesses the advantage of incorporating VR into the diagnostic process for MCI. We engaged 54 participants who underwent neuropsychological tests, participated in VR test, and received MRI scans. Analysis through machine learning revealed that neuropsychological measures alone provided an accuracy of 90.7%. Yet, when VR biomarkers were combined with these measures, accuracy surged to 94.4%. This investigation highlights VR's potential as an effective biomarker in the early detection of MCI and suggests that its integration with conventional biomarkers can not only boost diagnostic precision but also reduce the financial and time burdens associated with traditional methods.