背景:研究表明,深度学习影像组学(DLR)可以帮助区分胶质母细胞瘤(GBM)和孤立性脑转移瘤(SBM),但是整合人口统计学-MRI和DLR特征是否可以更准确地区分GBM和SBM仍不确定.
目的:构建并验证整合人口统计学-MRI和DLR特征的人口统计学-MRI深度学习影像组学列线图(DDLRN),以在术前区分GBM和SBM。
方法:回顾性。
方法:两百三十五例GBM(N=115)或SBM(N=120)患者,随机分为训练队列(90GBM和98SBM)和验证队列(25GBM和22SBM)。
■轴向T2加权快速自旋回波序列(T2WI),T2加权流体衰减反演恢复序列(T2-FLAIR),以及使用1.5T和3.0T扫描仪的对比度增强的T1加权自旋回波序列(CE-T1WI)。
结果:人口统计学-MRI特征具有七个影像学特征(\“池标志,\"\"不规则环标志,\"\"常规戒指标志,肿瘤内血管征,“瘤周水肿面积与肿瘤增强面积之比,T2-FLAIR与CE-T1WI上的病变面积之比,和肿瘤位置)和人口统计学因素(年龄和性别)。基于多参数MRI,影像组学和深度学习(DL)模型,DLR签名,和DDLRN进行了开发和验证。
方法:Mann-WhitneyU检验,皮尔逊测试,最小绝对收缩和选择运算符,并将支持向量机算法应用于影像组学和DL模型的特征选择和构建。
结果:DDLRN在训练和验证队列中显示出区别GBM和SBM的最佳表现,曲线下面积(AUC)为0.999和0.947,分别。此外,DLR签名(AUC=0.938)优于影像组学和DL模型,人口统计学-MRI特征(AUC=0.775)与验证队列中的T2-FLAIR影像组学和DL模型(AUC分别为0.762和0.749)相当.
结论:DDLRN整合了人口统计学-MRI和DLR特征,在区分GBM和SBM方面表现优异。
方法:3技术效果:阶段2。
BACKGROUND: Studies have shown that deep-learning radiomics (DLR) could help differentiate glioblastoma (GBM) from solitary brain metastasis (SBM), but whether integrating demographic-MRI and DLR features can more accurately distinguish GBM from SBM remains uncertain.
OBJECTIVE: To construct and validate a demographic-MRI deep-learning radiomics nomogram (DDLRN) integrating demographic-MRI and DLR signatures to differentiate GBM from SBM preoperatively.
METHODS: Retrospective.
METHODS: Two hundred and thirty-five patients with GBM (N = 115) or SBM (N = 120), randomly divided into a training cohort (90 GBM and 98 SBM) and a validation cohort (25 GBM and 22 SBM).
UNASSIGNED: Axial T2-weighted fast spin-echo sequence (T2WI), T2-weighted fluid-attenuated inversion recovery sequence (T2-FLAIR), and contrast-enhanced T1-weighted spin-echo sequence (CE-T1WI) using 1.5-T and 3.0-T scanners.
RESULTS: The demographic-MRI signature was constructed with seven imaging features (\"pool sign,\" \"irregular ring sign,\" \"regular ring sign,\" \"intratumoral vessel sign,\" the ratio of the area of peritumoral edema to the enhanced tumor, the ratio of the lesion area on T2-FLAIR to CE-T1WI, and the tumor location) and demographic factors (age and sex). Based on multiparametric MRI, radiomics and deep-learning (DL) models, DLR signature, and DDLRN were developed and validated.
METHODS: The Mann-Whitney U test, Pearson test, least absolute shrinkage and selection operator, and support vector machine algorithm were applied for feature selection and construction of radiomics and DL models.
RESULTS: DDLRN showed the best performance in differentiating GBM from SBM with area under the curves (AUCs) of 0.999 and 0.947 in the training and validation cohorts, respectively. Additionally, the DLR signature (AUC = 0.938) outperformed the radiomics and DL models, and the demographic-MRI signature (AUC = 0.775) was comparable to the T2-FLAIR radiomics and DL models in the validation cohort (AUC = 0.762 and 0.749, respectively).
CONCLUSIONS: DDLRN integrating demographic-MRI and DLR signatures showed excellent performance in differentiating GBM from SBM.
METHODS: 3 TECHNICAL EFFICACY: Stage 2.