Analysis of factors influencing post-pandemic sleep quality among hospital clinical staff in Haikou, China.

As a leading cause of mortality and long-term disability worldwide, stroke places an immense strain on both societal and healthcare resources (–). Upper extremity (UE) hemiparesis is a particularly common sequela, affecting roughly 80% of individuals post-stroke (, ). These motor deficits, characterized by reduced strength, coordination, and manual dexterity, often result in significant activity limitations that severely compromise the stroke survivor's independence and quality of life (). To maximize therapeutic efficacy, it is essential to perform precise, frequent, and objective evaluations of upper extremity function. The International Classification of Functioning, Disability and Health (ICF) framework suggests that such assessments should span body function, activity, and participation domains (). Systematic monitoring not only tracks recovery trajectories but also equips clinicians with essential data to forecast outcomes and customize intervention plans (, ). In the realm of clinical research and practice, the Action Research Arm Test (ARAT) is widely recognized as the reference standard for assessing upper extremity motor function due to its hierarchical structure, which comprehensively evaluates grasp, grip, pinch, and gross movements—domains that are critical for executing activities of daily living (ADLs) (, ). While it demonstrates robust psychometric qualities, such as high validity, reliability, and responsiveness (–), traditional administration is resource-heavy. It requires specific physical props and the constant supervision of a trained clinician for manual scoring. Such logistical barriers restrict assessment frequency in high-volume clinical environments and impede the adoption of tele-rehabilitation or remote monitoring strategies (, ). While clinically valuable, the conventional ARAT requires standardized physical props, dedicated clinical space, and manual administration. These requirements can make frequent serial assessments time-consuming and susceptible to inter-rater variability. Consequently, there is a need for valid and automated assessment alternatives. Beyond establishing statistical equivalence, the ARAT-VR aims to address these logistical limitations. By digitizing physical props and automating the scoring workflow, the VR system reduces equipment dependencies and may decrease both clinician burden and scoring subjectivity. Furthermore, optical tracking within VR platforms can capture kinematic metrics (e.g., movement trajectory and velocity) without external sensors, offering supplementary insights into movement quality and compensatory strategies.