术前MRI影像组学模型预测保乳手术治疗乳腺癌的可行性

Preoperative MRI radiomics model for predicting the feasibility oftreating breast cancer with breast conserving surgery

目的建立术前MRI影像组学模型,观察以之预测保乳手术(BCS)治疗乳腺癌的可行性。方法纳入90例接受BCS治疗的乳腺癌患者,其中61例BCS成功(成功组)、29例失败(失败组);按2∶1将其分为训练集(n=60)和测试集(n=30),并于各集内划分亚组,比较组间及亚组间临床及MRI资料,提取并筛选MRI影像组学特征,构建模型,评估其预测BCS可行性的价值。结果训练集2亚组间接受新辅助治疗情况、患侧乳房体积及有无子灶差异均有统计学意义(P均<0.05);测试集2亚组间肿瘤体积、患侧乳房体积及有无子灶差异均有统计学意义(P均<0.05)。最终以5个MRI组学特征及有无子灶建立影像组学模型,其预测训练集和测试集BCS可行性的曲线下面积分别为0.961和0.855。结论术前MRI影像组学模型可用于预测以BCS治疗乳腺癌的可行性。

Objective To construct a radiomics model based on preoperative MRI,and to explore its value for predicting the feasibility of treating breast cancer with breast conserving surgery(BCS).Methods Totally 90 patients with breast cancer who underwent BCS treatment were retrospectively enrolled,including 61 cases with successful(success group)and 29 cases with failed BCS(fail group).Then the patients were divided into training set(n=60)and test set(n=30)at the ratio of 2∶1 and further divided into subgroups within the sets.Clinical and MRI data were compared between groups and subgroups.MRI radiomics features were extracted and screened to construct the model,and its value for predicting the feasibility of BCS was analyzed.Results There were significant differences of receiving neoadjuvant therapy,the volume of affected breast and the presence or absence of multifocality between subgroups in training set(all P<0.05),also of tumor volume,the volume of affected breast and the presence or absence of multifocality between subgroups in test set(all P<0.05).Finally 5 MRI radiomics features and presence or absence of multifocality were selected to construct the radiomics model,and the area under the curve of the model for predicting the feasibility of BCS in training set and test set was 0.961 and 0.855,respectively.Conclusion Preoperative MRI radiomics model was helpful for predicting the feasibility of treating breast cancer with BCS.