Hu Yue, Cao Xin, Chen Hongyi, Geng Daoying, Lv Kun
Department of Radiology, Huashan Hospital, Fudan University, 12 Wulumuqi Middle Road, Shanghai, 200040, P.R. China.
Shanghai Engineering Research Center of Intelligent Imaging for Critical Brain Diseases, Shanghai, China.
Cancer Imaging. 2025 Aug 4;25(1):97. doi: 10.1186/s40644-025-00920-x.
Accurate preoperative grading of gliomas is critical for therapeutic planning and prognostic evaluation. We developed a noninvasive machine learning model leveraging whole-brain resting-state functional magnetic resonance imaging (rs-fMRI) biomarkers to discriminate high-grade (HGGs) and low-grade gliomas (LGGs) in the frontal lobe.
This retrospective study included 138 patients (78 LGGs, 60 HGGs) with left frontal gliomas. A total of 7134 features were extracted from the mean amplitude of low-frequency fluctuation (mALFF), mean fractional ALFF, mean percentage amplitude of fluctuation (mPerAF), mean regional homogeneity (mReHo) maps and resting-state functional connectivity (RSFC) matrix. Twelve predictive features were selected through Mann-Whitney U test, correlation analysis and least absolute shrinkage and selection operator method. The patients were stratified and randomized into the training and testing datasets with a 7:3 ratio. The logical regression, random forest, support vector machine (SVM) and adaptive boosting algorithms were used to establish models. The model performance was evaluated using area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity.
The selected 12 features included 7 RSFC features, 4 mPerAF features, and 1 mReHo feature. Based on these features, the model was established using the SVM had an optimal performance. The accuracy in the training and testing datasets was 0.957 and 0.727, respectively. The area under the receiver operating characteristic curves was 0.972 and 0.799, respectively.
Our whole-brain rs-fMRI radiomics approach provides an objective tool for preoperative glioma stratification. The biological interpretability of selected features reflects distinct neuroplasticity patterns between LGGs and HGGs, advancing understanding of glioma-network interactions.
准确的胶质瘤术前分级对于治疗方案规划和预后评估至关重要。我们开发了一种非侵入性机器学习模型，利用全脑静息态功能磁共振成像（rs-fMRI）生物标志物来区分额叶的高级别胶质瘤（HGGs）和低级别胶质瘤（LGGs）。
这项回顾性研究纳入了138例左侧额叶胶质瘤患者（78例LGGs，60例HGGs）。从低频波动平均振幅（mALFF）、平均分数ALFF、波动平均百分比振幅（mPerAF）、平均局部一致性（mReHo）图和静息态功能连接（RSFC）矩阵中提取了总共7134个特征。通过曼-惠特尼U检验、相关分析和最小绝对收缩和选择算子方法选择了12个预测特征。患者按7:3的比例分层随机分为训练集和测试集。使用逻辑回归、随机森林、支持向量机（SVM）和自适应增强算法建立模型。使用受试者操作特征曲线下面积、准确性、敏感性和特异性评估模型性能。
所选的12个特征包括7个RSFC特征、4个mPerAF特征和1个mReHo特征。基于这些特征，使用SVM建立的模型具有最佳性能。训练集和测试集的准确性分别为0.957和0.727。受试者操作特征曲线下面积分别为0.972和0.799。
我们的全脑rs-fMRI放射组学方法为术前胶质瘤分层提供了一种客观工具。所选特征的生物学可解释性反映了LGGs和HGGs之间不同的神经可塑性模式，推进了对胶质瘤-网络相互作用的理解。
