Reconnecting Brain Networks After Stroke: A Scoping Review of Conventional, Neuromodulatory, and Feedback-Driven...

Stroke leads to lasting disability by disrupting the connectivity of functional brain networks. Although several rehabilitation methods are promising, our full understanding of how these strategies restore network function is still limited. Here, we map how non-invasive brain stimulation (NIBS), brain-computer interface (BCI)/neurofeedback, virtual reality (VR), and robot-assisted therapy restore connectivity within the sensorimotor network (SMN), default mode network (DMN), and salience network, and we contextualize these effects within the known temporal evolution of post-stroke motor network reorganization.

This scoping review adhered to PRISMA guidelines and searched PubMed, Cochrane, and Medline from January 2015 to January 2025 for clinical trials focused on stroke rehabilitation with functional connectivity outcomes. Included studies used conventional therapy, neuromodulation, or feedback-based interventions.

Twenty-three studies fulfilled the inclusion criteria, covering interventions like robotic training, transcranial stimulation (tDCS/TMS), brain-computer interfaces, virtual reality, and cognitive training. Motor impairments were linked to disrupted interhemispheric sensorimotor connectivity, while cognitive issues reflected changes in frontoparietal and default mode networks. Combining neuromodulation with feedback-based methods showed better network recovery than standard therapy alone, with clinical improvements closely associated with connectivity alterations.

Effective stroke rehabilitation depends on targeting specific disrupted networks through various modalities. Robotic interventions focus on restoring structural motor pathways, feedback-enhanced methods improve temporal synchronization, and cognitive training aims to enhance higher-order network integration. Future research should work toward standardizing connectivity assessment protocols and conducting multicenter trials. This will help develop evidence-based, network-focused rehabilitation guidelines that effectively translate mechanistic insights into personalized clinical treatments.