Background:Brain–computer interface (BCI)-mediated neurofeedback training (BCI-NFT) has emerged as a highly promising treatment in the field of neurorehabilitation. Many previous studies have demonstrated the efficacy of BCI techniques in clinical rehabilitation, but children are largely neglected in BCI research. Purpose:This systematic review aimed to synthesize existing studies from technical and clinical application perspectives to identify the current state of research on noninvasive brain–computer interface (NBCI) technology in children with two major neurodevelopmental disorders, autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD). Methods:Five relevant electronic databases were searched (PubMed, Web of Science, the Cochrane Library, Embase, and the Cumulative Index of Nursing and Allied Health Literature). The publication dates ranged from the inception of each database to June 2024. Randomized controlled trials (RCTs) investigating the use of NBCI technology in children with ASD or ADHD were included. Manual searches of the clinical trial registry platforms and the reference lists of reviews related to the study topic were also conducted. Two independent reviewers performed the literature screening, data extraction, and risk of bias assessment. Results:A total of 24 RCTs involving 1998 children with ASD or ADHD were included in this systematic review. With respect to input brain signals, functional magnetic resonance imaging (fMRI) (4.2%), electroencephalography (EEG) combined with fMRI (4.2%), and EEG combined with galvanic skin response (GSR) sensors (4.2%) were utilized in one study each. Seven studies employed EEG combined with electrooculogram (EOG) (29.1%), and the remaining fourteen studies used EEG alone (58.3%). Compared with those of the controls, significant improvements in both behavioral aspects and brain activity in patients were observed in eleven studies (45.8%). NBCI technology has a positive effect on both the behavioral and brain activity levels of children with ASD or ADHD, while it still faces challenges in the paediatric population, particularly in terms of signal processing and the unique cognitive and physiological developmental stages of children, which may complicate the application of these technologies in this population. Conclusion:It demonstrated that there has a high potential for NBCI application in the field of neurodevelopmental disorders. Future research should focus on developing advanced machine learning algorithms to improve neural signal decoding capabilities and on creating child-appropriate application paradigms to explore the long-term efficacy of these algorithms.
Read full abstract