首页|基于MRI影像组学鉴别诊断良恶性椎体压缩骨折的Meta分析

基于MRI影像组学鉴别诊断良恶性椎体压缩骨折的Meta分析

Differential Diagnosis Based on MRI Imaging Radiomics Meta-analysis of Benign and Malignant Vertebral Compression Fractures

扫码查看
目的 运用Meta分析评价基于MRI影像组学鉴别诊断良恶性椎体压缩性骨折(VCF)的准确性,推测其临床价值.方法 在中英文数据库中全面检索发表于2023年12月以前关于运用影像组学技术鉴别诊断良恶性VCF的研究,根据文献纳入排除标准进行筛选并进行质量评估及数据提取.利用Stata软件对提取数据进行Meta分析,计算合并敏感度、特异度等数据,绘制综合受试者工作特征曲线(SROC),对异质性来源进行回归分析和亚组分析,并采用Deek's法检验发表偏倚.结果 7项符合标准的研究被纳入分析,合并敏感度和特异度分别为0.92(95%CI:0.86~0.95)和 0.95(95%CI:0.90~0.98),绘制 SROC 曲线,其曲线下面积(AUC)为 0.98(95%CI:0.96~0.99).是否使用交叉验证、是否纳入临床特征及成像序列均是异质性来源.结论 影像组学模型可有效鉴别诊断良恶性VCF,具有良好临床应用前景.
Objective Meta-analysis was performed to evaluate the accuracy of MRI-based differential diagnosis of be-nign and malignant vertebral compression fracture(VCF)and to estimate its clinical value.Methods A comprehensive search was conducted in Chinese and English databases for studies published before December 2023 on the differential diag-nosis of benign and malignant VCFs using imaging histology,and the data were screened according to the criteria for ranking the literature,and quality assessment and data extraction were performed.Meta-analysis was performed on the extracted data using Stata software,combined sensitivity and specificity data were calculated,a composite subject work characteristic curve(SROC)was plotted,regression and subgroup analyses were performed on the sources of heterogeneity,and publication bias was examined using Deek's method.Results Eight studies that met the criteria were included in the analysis,with a com-bined sensitivity and specificity of 0.92(95%CI=0.86-0.95)and 0.95(95%CI=0.90-0.98),respectively,and plotting the SROC curve revealed an area under the curve(AUC)=0.98(95%CI=0.96-0.99).The use of cross-validation,the inclusion of clinical features and imaging sequences were all sources of heterogeneity.Conclusion Imaging histological models can effectively differentiate and diagnose benign and malignant VCF with good clinical application.

RadiomicsCompression fractureMagnetic resonance imagingMeta-analysis

苏超、田彧潇、侯晓宙、夏迪、张清

展开 >

100102 北京,中国中医科学院望京医院

100029 北京中医药大学

200355 济南,山东中医药大学

影像组学 压缩性骨折 磁共振影像 荟萃分析

中国中医科学院科技创新工程项目

C12021A02003

2024

临床放射学杂志
黄石市医学科技情报所

临床放射学杂志

CSTPCD北大核心
影响因子:0.872
ISSN:1001-9324
年,卷(期):2024.43(10)