Search

Abstract Background Physical activity (PA) interventions have been shown to yield positive effects on cognitive functions. However, it is unclear which type of PA intervention is the most effective in children and adolescents with Neurodevelopmental Disorders (NDDs). This study aimed to compare the effectiveness of different types of PA interventions on cognitive functions in children and adolescents with NDDs, with additional analyses examining intervention effects across specific NDD types including attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD). Methods In this systematic review and network meta-analysis, seven databases (Web of Science, PubMed, Medline, APA PsycINFO, Embase, CINAHL, and SPORTDiscus) for randomized controlled trials from database inception to September 2023 were searched. Randomized controlled trials comparing the effectiveness of PA intervention with any non-pharmacological treatment or control group on cognitive functions in children and adolescents diagnosed with NDDs aged 5–17 years were included. Frequentist network meta-analyses were performed based on standardized mean differences (SMD) using random effects models to examine post-intervention differences in cognitive functions, including attention, memory, and executive functions. Intervention dropout was assessed as a measure of treatment acceptability. Results Thirty-one randomized controlled trials (n = 1,403, mean age 10.0 ± 1.9 years) with 66 arms were included in the network. Mind-body exercise (MBE; SMD = 1.91 for attention; 0.92 for executive functions), exergaming (SMD = 1.58 for attention; 0.97 for memory; 0.94 for executive functions), and multi-component physical activity (MPA; SMD = 0.79 for executive functions) were associated with moderate to substantial cognitive improvements compared with usual care, whereas the effectiveness of aerobic exercise (AE) was non-significant. Exergaming (SMD = 0.78, 95%CI 0.12 to 1.45) and MPA (SMD = 0.64, 95%CI 0.11 to 1.18) were more effective than AE for executive functions. When analyzing specific NDD types, exergaming lost its superiority over usual care for attention and memory in ADHD, nor for executive functions in ASD. Instead, MPA demonstrated significant benefits across these domains and populations. The certainty of evidence for these comparisons was very low to low. No significant differences in acceptability were observed among MBE, exergaming, and MPA. Conclusions The findings in this study suggest that MBE, exergaming, and MPA were effective interventions for improving domain-specific cognitive functions in children and adolescents with NDDs. AE demonstrated non-significant effectiveness for all outcomes. MBE emerges as particularly advantageous for attention. MPA yielded consistent improvements in memory and executive functions across NDD types. Further high-quality randomized controlled trials of direct comparisons are needed to confirm and expand on the findings from this NMA. Trial registration PROSPERO CRD42023409606.

Abstract Background The application of artificial intelligence (AI) in academic writing has raised concerns regarding accuracy, ethics, and scientific rigour. Some AI content detectors may not accurately identify AI-generated texts, especially those that have undergone paraphrasing. Therefore, there is a pressing need for efficacious approaches or guidelines to govern AI usage in specific disciplines. Objective Our study aims to compare the accuracy of mainstream AI content detectors and human reviewers in detecting AI-generated rehabilitation-related articles with or without paraphrasing. Study design This cross-sectional study purposively chose 50 rehabilitation-related articles from four peer-reviewed journals, and then fabricated another 50 articles using ChatGPT. Specifically, ChatGPT was used to generate the introduction, discussion, and conclusion sections based on the original titles, methods, and results. Wordtune was then used to rephrase the ChatGPT-generated articles. Six common AI content detectors (Originality.ai, Turnitin, ZeroGPT, GPTZero, Content at Scale, and GPT-2 Output Detector) were employed to identify AI content for the original, ChatGPT-generated and AI-rephrased articles. Four human reviewers (two student reviewers and two professorial reviewers) were recruited to differentiate between the original articles and AI-rephrased articles, which were expected to be more difficult to detect. They were instructed to give reasons for their judgements. Results Originality.ai correctly detected 100% of ChatGPT-generated and AI-rephrased texts. ZeroGPT accurately detected 96% of ChatGPT-generated and 88% of AI-rephrased articles. The areas under the receiver operating characteristic curve (AUROC) of ZeroGPT were 0.98 for identifying human-written and AI articles. Turnitin showed a 0% misclassification rate for human-written articles, although it only identified 30% of AI-rephrased articles. Professorial reviewers accurately discriminated at least 96% of AI-rephrased articles, but they misclassified 12% of human-written articles as AI-generated. On average, students only identified 76% of AI-rephrased articles. Reviewers identified AI-rephrased articles based on ‘incoherent content’ (34.36%), followed by ‘grammatical errors’ (20.26%), and ‘insufficient evidence’ (16.15%). Conclusions and relevance This study directly compared the accuracy of advanced AI detectors and human reviewers in detecting AI-generated medical writing after paraphrasing. Our findings demonstrate that specific detectors and experienced reviewers can accurately identify articles generated by Large Language Models, even after paraphrasing. The rationale employed by our reviewers in their assessments can inform future evaluation strategies for monitoring AI usage in medical education or publications. AI content detectors may be incorporated as an additional screening tool in the peer-review process of academic journals.