Digital Neurorehabilitation: Virtual Reality, Robotics and Neuroplasticity After Stroke

Abstract

Stroke is a leading cause of long-term neurological disability, frequently producing persistent motor impairment, loss of independence, and limitations in social participation. Virtual reality and robot-assisted rehabilitation have emerged as strategies for increasing treatment intensity, repetition, task specificity, feedback, and patient engagement. This study aimed to analyze their effectiveness in promoting neuroplasticity and improving motor and functional outcomes in adults after stroke. A structured integrative literature review based on the Scientific Method and PRISMA 2020 principles was conducted using PubMed/MEDLINE, the Cochrane Library, Scopus, Web of Science, IEEE Xplore, PEDro, and official journal databases. Twenty publications were selected, including systematic reviews, meta-analyses, randomized controlled trials, narrative reviews, and a clinical practice guideline. Ten publications were systematic reviews or meta-analyses; five presented predominantly favorable findings, four reported conditional or mixed results, and one found no clear additional benefit. Virtual reality was associated with selected improvements in upper-limb motor function, dexterity, functional independence, quality of life, cortical activation, and functional connectivity. Robot-assisted rehabilitation provided intensive and measurable movement practice and improved selected impairment-based outcomes, although its superiority over dose-matched conventional rehabilitation was inconsistent. Improvements in motor control and strength did not always translate into greater independence or spontaneous use of the affected limb in daily life. The findings indicate that digital neurorehabilitation is most effective when it provides active, repetitive, intensive, task-specific, progressively challenging, and feedback-rich training. Virtual reality and robotics should be integrated as complementary components of conventional rehabilitation rather than universal replacements. Further randomized trials with standardized protocols, equivalent treatment doses, clinically meaningful outcomes, direct neuroplasticity measurements, safety assessment, and long-term follow-up are required.