Abstract
Despite the emerging role of artificial intelligence (AI) in glioma grading, its clinical adoption remains in its early stages. This meta-analysis aims to assess the role of AI in differentiating glioma grades using magnetic resonance imaging (MRI). Twenty-five studies matched the inclusion criteria and were included after systematic searches through "PubMed" electronic database. The quality of the included studies was assessed utilizing Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). A bivariate random-effects model was employed to estimate the pooled effect of the sensitivity and specificity, followed by an estimation of the summary receiver operating characteristic (SROC) curve. The overall results suggest relatively high sensitivity and specificity among the assessed AI methods for discriminating glioma grades. Convolutional Neural Networks (CNN) demonstrated the highest diagnostic accuracy, with a sensitivity of 93% (95% CI: 88%-97%) and specificity of 92% (95% CI: 90%-94%). This meta-analysis highlights the potential role of AI models based on MRI in supporting clinicians in glioma grading.