基于扩散峰度成像序列平均扩散峰度图的影像组学鉴别宫颈鳞癌与腺癌的应用

Application of Radiomics Methods Based on DKI Sequence MK Map for Differentiating Squamous Cell Carcinoma from Cervix Adenocarcinoma

目的探讨基于扩散峰度成像(DKI)序列平均扩散峰度(MK)图的影像组学方法鉴别宫颈鳞状细胞癌(CSCC)和宫颈腺癌(CA)的价值。资料与方法回顾性收集接受MRI检查(含DKI序列)并经手术病理证实的宫颈癌63例,分为训练组43例和验证组20例,构建多元Logistic回归预测模型,采用受试者工作特征(ROC)曲线评价该模型在训练组和验证组的诊断效能。结果提取得到386个影像组学特征,最终筛选出7个与宫颈癌病理分型相关度最高的组学特征。构建的模型在训练组鉴别不同病理类型宫颈癌的准确度、ROC曲线下面积、敏感度、特异度分别为74.4%、0.867、85.7%、69.0%,在验证组分别为80.0%、0.846、85.7%、76.9%。结论基于DKI序列MK图的影像组学方法可有效鉴别不同病理类型的宫颈癌,有助于临床决策的制定。

Purpose To explore the application value of radiomics based on mean kurtosis map of diffusion kurtosis imaging(DKI)sequence in differentiating cervix squamous cell carcinoma(CSCC)from cervix adenocarcinoma(CA).Materials and Methods A total of 63 patients with cervical cancer(CC)who underwent MRI(included DKI sequence)and confirmed by surgical pathology were retrospectively collected.All patients were divided into training group(n=43)and test group(n=20),building a multiple logical regression model,and ROC curves were drawn to evaluate the diagnostic effectiveness of the model and verify its effectiveness in the test group.Results A total of 386 radiomics features were extracted,and 7 radiomics features related to CC pathological classification were finally obtained by dimension reduction.The accuracy,area under curve(AUC),sensitivity and specificity of the constructed model for distinguishing different pathological types of CC were 74.4%,0.867,85.7%and 69.0%in the training group;and 80.0%,0.846,85.7%and 76.9%in the test group,respectively.Conclusion Radiomics based on mean kurtosis map of DKI sequence could effectively identify different pathological types of CC,contributing to clinical decision-making.