基于瘤体及瘤周多参数MRI对乳腺病变良恶性诊断列线图预测模型的构建与评价  

Construction and evaluation of nomogram prediction model for benign and malignant breast lesions based on tumor and peritumoral multi-parameter MRI

作　　者：张春福[1] 彭波[1] 黄崎[1] 张雪峰[1] 才春红[1] 海洋[1] 张巍巍[1] ZHANG Chunfu;PENG Bo;HUANG Qi;ZHANG Xuefeng;CAI Chunhong;HAI Yang;ZHANG Weiwei(Department of Radiology,Daqing Oilfield General Hospital,Daqing 163001,China)

机构地区：[1]大庆油田总医院放射科,黑龙江大庆163001

出　　处：《陕西医学杂志》2024年第1期72-76,共5页Shaanxi Medical Journal

基　　金：黑龙江省科研基金资助项目(2021-KYYWFMY-0049)。

摘　　要：目的:建立基于瘤体及瘤周多参数MRI的乳腺病变良恶性鉴别诊断的列线图模型,并验证其预测效能。方法:纳入经病理学检查明确乳腺病变性质的100例患者作为研究对象,所有患者均行核磁共振(MRI)检查和病理检查,根据病理检查结果分为乳腺良性病变组(n=62)和乳腺恶性病变组(n=38)。收集患者临床资料、瘤体各参数、瘤周各参数以及乳腺病变良恶性情况。多因素Logistic回归分析筛选乳腺恶性病变的危险因素并构建列线图预测模型,采用受试者工作特征(ROC)曲线和Hosmer-Lemeshow拟合优度检验验证模型的预测效能及拟合优度;内部验证采用Bootstrap。结果:乳腺恶性病变组病灶直径、平均扩散峰度(MK)、MDp/t、瘤周与瘤体MKp/n高于乳腺良性病变组(均P<0.05);乳腺恶性病变组表观扩散系数(ADC)值、平均扩散率(MD)、非对称磁化转移率(MTRasym)、MKp/t、MDp/n低于乳腺良性病变组(均P<0.05)。多因素分析结果显示,病灶直径、MK、MDp/t、MKp/n升高,ADC值、MD、MTRasym、MKp/t、MDp/n降低是乳腺恶性病变的独立影响因素(均P<0.05)。基于上述独立影响因素构建乳腺恶性病变的列线图预测模型,曲线下面积(AUC)为0.827。Hosmer-Lemeshow拟合优度检验显示P值为0.004。采用Bootstrap法,生成的校准曲线拟合良好。结论:瘤体及瘤周多参数MRI对乳腺病变良恶性鉴别诊断具有重要预测价值,基于乳腺恶性病变的独立影响因素构建的列线图预测效果良好,能直观预测乳腺发生恶性病变的概率。Objective:To establish a nomogram model for differential diagnosis of benign and malignant breast lesions based on multi-parameter MRI of tumor and peritumor,and to verify its predictive efficacy.Methods:A total of 100 patients with breast lesions confirmed by pathological examination were included in the study.All patients underwent magnetic resonance imaging(MRI)examination and pathological examination.According to the results of pathological examination,they were divided into benign breast lesions group(n=62)and malignant breast lesions group(n=38).The clinical data,tumor parameters,peritumoral parameters and benign and malignant breast lesions were collected.Multivariate Logistic regression analysis was used to screen the risk factors of breast malignant lesions and construct a nomogram prediction model.The receiver operating characteristic(ROC)curve and Hosmer-Lemeshow goodness of fit test were used to verify the prediction efficiency and goodness of fit of the model.Bootstrap was used for internal validation.Results:The lesion diameter,mean kurtosis(MK),MDp/t,MKp/n of peritumoral and tumor in malignant breast lesions group were higher than those in benign breast lesions group(all P<0.05).The apparent diffusion coefficient(ADC)value,mean diffusivity(MD),asymmetric magnetization transfer rate(MTRasym),MKp/t and MDp/n in the malignant breast lesions group were lower than those in the benign breast lesions group(all P<0.05).Multivariate analysis showed that the increase of lesion diameter,MK,MDp/t and MKp/n,and the decrease of ADC value,MD,MTRasym,MKp/t and MDp/n were independent influencing factors for breast malignant lesions(all P<0.05).Based on the above independent influencing factors,a nomogram prediction model of breast malignant lesions was constructed,and the area under the curve(AUC)was 0.827.The Hosmer-Lemeshow goodness-of-fit test showed a P value of 0.004.Using the Bootstrap method,the generated calibration curve fits well.Conclusion:Tumor and peritumoral multi-parameter MRI has important predictiv

关 键 词：乳腺病变 良恶性 鉴别诊断 瘤体参数 瘤周参数 核磁共振 列线图 

分 类 号：R737.9[医药卫生—肿瘤]