■基于不同机器学习(ML)模型的比较,我们开发了该模型,该模型整合了传统的手工制作(HC)特征和多参数MRI中基于ResNet50网络的深度迁移学习(DTL)特征,以预测鼻窦鳞状细胞癌(SNSCC)的Ki-67状态.
■回顾性分析了两百三十一名SNSCC患者[训练队列(n=185),测试队列(n=46)]。病理分级,临床,分析MRI特征,选择独立预测因子。从脂肪饱和T2加权成像中提取HC和DTL影像组学特征,对比增强T1加权成像,和表观扩散系数图。然后,将HC和DTL特征融合以形成基于深度学习的影像组学(DLR)特征。在特征选择和影像组学签名(RS)构建之后,我们比较了RS-HC的预测能力,RS-DTL,和RS-DLR。
■没有发现基于病理的独立预测因子,临床,和MRI特征。选择功能后,保留了42个HC和10个DTL影像组学特征。支持向量机(SVM)LightGBM,ExtraTrees(ET)是RS-HC的最佳分类器,RS-DTL,和RS-DLR。在训练组中,RS-DLR的预测能力明显优于RS-DTL和RS-HC(p<0.050);在测试集中,RS-DLR的曲线下面积(AUC)(AUC=0.817)也最高,但是DLR-RS和HC-RS之间的性能没有显着差异。
■HC和DLR模型对SNSCC患者的Ki-67表达均显示良好的预测功效。尤其是,RS-DLR模型代表了提高预测能力的机会。
UNASSIGNED: Based on comparison of different machine learning (ML) models, we developed the model that integrates traditional hand-crafted (HC) features and ResNet50 network-based deep transfer learning (DTL) features from multiparametric MRI to predict Ki-67 status in sinonasal squamous cell carcinoma (SNSCC).
UNASSIGNED: Two hundred thirty-one SNSCC patients were retrospectively reviewed [training cohort (n = 185), test cohort (n = 46)]. Pathological grade, clinical, and MRI characteristics were analyzed to choose the independent predictor. HC and DTL
radiomics features were extracted from fat-saturated T2-weighted imaging, contrast-enhanced T1-weighted imaging, and apparent diffusion coefficient map. Then, HC and DTL features were fused to formulate the deep learning-based
radiomics (DLR) features. After feature selection and
radiomics signature (RS) building, we compared the predictive ability of RS-HC, RS-DTL, and RS-DLR.
UNASSIGNED: No independent predictors were found based on pathological, clinical, and MRI characteristics. After feature selection, 42 HC and 10 DTL
radiomics features were retained. The support vector machine (SVM), LightGBM, and ExtraTrees (ET) were the best classifier for RS-HC, RS-DTL, and RS-DLR. In the training cohort, the predictive ability of RS-DLR was significantly better than those of RS-DTL and RS-HC (p< 0.050); in the test set, the area under curve (AUC) of RS-DLR (AUC = 0.817) was also the highest, but there was no significant difference of the performance between DLR-RS and HC-RS.
UNASSIGNED: Both the HC and DLR model showed favorable predictive efficacy for Ki-67 expression in patients with SNSCC. Especially, the RS-DLR model represented an opportunity to advance the prediction ability.