Effects of brain-computer interface training on upper limb function recovery in stroke patients

Abstract

Background: In recent years, with the development of medical technology and the increase of inter-disciplinary cooperation technology, new methods in the field of artificial intelligence medicine emerge in an endless stream. Brain-computer interface (BCI), as a frontier technology of multidisciplinary integration, has been widely used in various fields. Studies have shown that BCI-assisted training can improve upper limb function in stroke patients, but its effect is still controversial and lacks evidence-based evidence, which requires further exploration and confirmation. Therefore, the main purpose of this paper is to systematically evaluate the efficacy of different BCI-assisted training on upper limb function recovery in stroke patients, to provide a reference for the application of BCI-assisted technology in stroke rehabilitation. Methods: We will search PubMed, Web of Science, The Cochrane Library, Chinese National Knowledge Infrastructure Database, Wanfang Data, Weipu Electronics, and other databases (from the establishment to February 2021) for full text in Chinese and English. Randomized controlled trials were collected to examine the effect of BCI-assisted training on upper limb functional recovery in stroke patients. We will consider inclusion, select high-quality articles for data extraction and analysis, and summarize the intervention effect of BCI-assisted training on the upper limb function of stroke patients. Two reviewers will screen titles, abstracts, and full texts independently according to inclusion criteria; Data extraction and risk of bias assessment were performed in the included studies. We will use a hierarchy of recommended assessment, development, and assessment methods to assess the overall certainty of the evidence and report findings accordingly. Endnote X8 will be applied in selecting the study, Review Manager 5.3 will be applied in analyzing and synthesizing. Results: The results will provide evidence for judging whether BCI is effective and safe in improving upper limb function in patients with stroke. Conclusion: Our study will provide reliable evidence for the effect of BCI technology on the improvement of upper limb function in stroke patients. PROSPERO registration number: CRD42021250378.

Global Trends and Hotspots in Research on Rehabilitation Robots: A Bibliometric Analysis From 2010 to 2020

“…With human intelligence and machine intelligence, combining human–computer interaction as a representative of technological breakthroughs between man and machine combined with the increasingly close, with the help of human–computer interaction technology and method, and combining human intelligence and machine intelligence make the two complementary advantages, to work together, and in the aspect of medical rehabilitation which spawned a significant theoretical innovation and breakthrough technology method. The brain–computer interface will be widely applied in future ( 61 , 62 ). How to establish the ability of high-broadband data connection between human and machine is also an important trend in the current international rehabilitation robot research.…”

New Insights into Brain Imaging Methods for Rehabilitation of Brain Diseases

“…Scholars have commonly expressed their concern regarding how to enhance the effectiveness of limb function in patients with hemiplegic stroke. The main clinical treatments for post-stroke limb dysfunction include physical therapy (European and Rehabilitation Medicine Bodies, 2018), occupational therapy (Powell et al, 2016), transcutaneous electrical nerve stimulation (Mahmood et al, 2019), functional electrical stimulation (Eraifej et al, 2017), transcranial direct current stimulation (Van Hoornweder et al, 2021), repetitive transcranial magnetic stimulation (Starosta et al, 2022), motor imagery (Monteiro et al, 2021), virtual reality , constraint induced movement therapy (Rocha et al, 2021), brain-computer interface technology (Xue et al, 2021), and robotics (Bruni et al, 2018), etc. These intervention strategies have limitations due to the need for trained professionals, lack of precision, consensus evidence, and high implementation costs .…”

A sequential learning model with GNN for EEG-EMG-based stroke rehabilitation BCI

“…Movements are composed of multiple sub-actions or action units, and the ability to capture the brain's sequential motor intention can provide patients with more refined feedback for better recovery (Xue et al, 2021 ). However, due to impaired motor function, stroke patients may perform movements in a compensatory manner, resulting in movement patterns that differ from those of a healthy individual (Alankus and Kelleher, 2012 ).…”