Related Experiment Video
Updated: May 9, 2026
Measuring the Kinematics of Daily Living Movements with Motion Capture Systems in Virtual Reality
Published on: April 5, 2018
Sleep quality metrics combined with virtual reality motion parameters enhance early detection of mild cognitive
Ruirui Zhang1, Huaiqing Sun1, Yaxuan Di1
1Department of Neurology, The First Affiliated Hospital with Nanjing Medical University, Nanjing Medical University, Nanjing, China.
Objective:
Alzheimer's disease (AD) is a progressive neurodegenerative disorder marked by cognitive and motor deficits. With its global prevalence increasing rapidly and no effective treatment available, early identification of high-risk individuals is critical. This study investigated the relationship between motor parameters extracted from virtual reality (VR) tasks, combined with sleep-related measures, and cognitive impairment in patients with mild cognitive impairment (MCI). Our goal was to determine whether integrating VR-derived digital markers with sleep quality metrics could provide an objective and clinically applicable tool for early detection.
Methods:
66 participants were recruited, including 28 healthy controls (HC) and 38 patients with MCI. Cognitive status was assessed using the Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE). All participants performed two scenario-based VR tasks, during which task completion time, accuracy, and overall performance scores were recorded. Group differences were evaluated using independent-samples t-tests, and these behavioral features and sleep quality metrics were further incorporated into ROC analyze to assess predictive performance for distinguishing MCI from HC.
Results:
Compared with HC, patients with MCI reported significantly poorer sleep quality based on the Pittsburgh Sleep Quality Index (PSQI) and subdomains such as sleep latency and habitual sleep efficiency. In the VR tasks, MCI patients required more time and achieved lower accuracy than HC, consistent with MoCA and MMSE scores. Correlation analysis confirmed strong associations between VR performance metrics and cognitive test scores. Importantly, integrating VR-derived digital markers with sleep parameters yielded superior predictive accuracy for MCI (AUC = 0.863; sensitivity = 86.84%; specificity = 71.43%; p < 0.001) compared with single-modality models.
Conclusion:
VR-based cognitive and sensorimotor tasks, when combined with sleep quality assessments, offer a robust and noninvasive approach for the early identification of prodromal AD. This multimodal strategy holds promise for enhancing clinical decision-making and enabling timely interventions.
