Farahani Somayeh, Hejazi Marjaneh, Moradizeyveh Sahar, Di Ieva Antonio, Fatemizadeh Emad, Liu Sidong
Department of Medical Physics and Biomedical Engineering, School of Medicine, Tehran University of Medical Sciences, Tehran 14618-84513, Iran.
Centre for Health Informatics, Australian Institute of Health Innovation, Macquarie University, Sydney, NSW 2109, Australia.
Diagnostics (Basel). 2025 Mar 21;15(7):797. doi: 10.3390/diagnostics15070797.
Integrating deep learning (DL) into radiomics offers a noninvasive approach to predicting molecular markers in gliomas, a crucial step toward personalized medicine. This study aimed to assess the diagnostic accuracy of DL models in predicting various glioma molecular markers using MRI. Following PRISMA guidelines, we systematically searched PubMed, Scopus, Ovid, and Web of Science until 27 February 2024 for studies employing DL algorithms to predict gliomas' molecular markers from MRI sequences. The publications were assessed for the risk of bias, applicability concerns, and quality using the QUADAS-2 tool and the radiomics quality score (RQS). A bivariate random-effects model estimated pooled sensitivity and specificity, accounting for inter-study heterogeneity. Of 728 articles, 43 were qualified for qualitative analysis, and 30 were included in the meta-analysis. In the validation cohorts, MGMT methylation had a pooled sensitivity of 0.74 (95% CI: 0.66-0.80) and a pooled specificity of 0.75 (95% CI: 0.65-0.82), both with significant heterogeneity ( = 0.00, I = 80.90-84.50%). ATRX and TERT mutations had a pooled sensitivity of 0.79 (95% CI: 0.67-0.87) and 0.81 (95% CI: 0.72-0.87) and a pooled specificity of 0.85 (95% CI: 0.78-0.91) and 0.70 (95% CI: 0.61-0.77), respectively. Meta-regression analyses revealed that significant heterogeneity was influenced by data sources, MRI sequences, feature extraction methods, and validation techniques. While the DL models show promising prediction accuracy for glioma molecular markers, variability in the study settings complicates clinical translation. To bridge this gap, future efforts should focus on harmonizing multi-center MRI datasets, incorporating external validation, and promoting open-source studies and data sharing.
将深度学习（DL）与放射组学相结合，为预测神经胶质瘤中的分子标志物提供了一种非侵入性方法，这是迈向个性化医疗的关键一步。本研究旨在评估DL模型使用MRI预测各种神经胶质瘤分子标志物的诊断准确性。按照PRISMA指南，我们系统检索了PubMed、Scopus、Ovid和Web of Science，直至2024年2月27日，以查找采用DL算法从MRI序列预测神经胶质瘤分子标志物的研究。使用QUADAS-2工具和放射组学质量评分（RQS）对出版物的偏倚风险、适用性问题和质量进行评估。双变量随机效应模型估计合并敏感性和特异性，并考虑研究间的异质性。在728篇文章中，43篇符合定性分析要求，30篇纳入荟萃分析。在验证队列中，MGMT甲基化的合并敏感性为0.74（95%CI：0.66-0.80），合并特异性为0.75（95%CI：0.65-0.82），两者均存在显著异质性（ = 0.00，I = 80.90-84.50%）。ATRX和TERT突变的合并敏感性分别为0.79（95%CI：0.67-0.87）和0.81（95%CI：0.72-0.87），合并特异性分别为0.85（95%CI：0.78-0.91）和0.70（95%CI：0.61-0.77）。荟萃回归分析显示，显著异质性受数据来源、MRI序列、特征提取方法和验证技术的影响。虽然DL模型对神经胶质瘤分子标志物显示出有前景的预测准确性，但研究设置的变异性使临床转化变得复杂。为弥合这一差距，未来的努力应集中在协调多中心MRI数据集、纳入外部验证以及促进开源研究和数据共享上。
