摘要
目的:探讨联影3.0 T MRI鉴别乳腺良恶性病变的最佳ADC阈值。方法:收集经病理证实的56例乳腺病变患者(56个病灶),其中乳腺癌38例(恶性组),乳腺良性病变18例(良性组)。2组术前均行联影3.0 T MRI检查,获取病灶最低ADC值;采用独立t检验及箱图分析良恶性病变的ADC值差异;采用ROC曲线及散点图分析鉴别良恶性病变的最佳ADC阈值。结果:38个乳腺癌的平均ADC值为(0.914±0.196)×10mm^(2)/s,18个乳腺良性病变为(1.480±0.312)×10mm^(2)/s,差异有统计学意义(P<0.05)。ROC曲线分析:AUC为0.954时鉴别乳腺良恶性病变具有较高的诊断效能。约登指数为0.806时对应的ADC阈值1.106×10mm^(2)/s,阳性似然比15.40,阴性似然比0.147,敏感度86.1%,特异度94.4%;约登指数为0.722时对应的ADC阈值1.253×10mm^(2)/s,阳性似然比4.25,阴性似然比0.072,敏感度94.4%,特异度77.8%。结论:ADC阈值为1.106×10mm^(2)/s可最大程度区分乳腺良恶性病变;ADC阈值为1.253×10mm^(2)/s可减少乳腺癌漏诊,推荐在临床中应用。
Objective:To explore the optimal ADC threshold in differentiating benign and malignant breast lesions with 3.0 T MRI.Methods:A total of 56 breast lesions confirmed by pathology were collected,including 38 breast cancers and 18 benign lesions.All patients examined with 3.0 T MRI before operation.The minimum ADC values of the lesions were obtained,independent t test and box plot were used to analyze the difference of ADC values between benign and malignant lesions,and ROC curve and scatter diagram were used to analyze the optimal ADC threshold to differentiate benign and malignant lesions.Results:The mean ADC values of 38 breast cancers was (0.914±0.196)×10mm^(2)/s and that of 18 benign lesions was(1.480±0.312)×10mm^(2)/s,the difference was statistically significant (P<0.05).ROC curve analysis showed that the AUC of0.954 had a better efficiency in the differential diagnosis of begin and malignant lesions.When Youden index was 0.806,ADC threshold was 1.106×10mm^(2)/s,positive likelihood ratio was 15.4,negative likelihood ratio was 0.147,the sensitivity was 86.1%,the specificity was 94.4%;and when Youden index was 0.722,ADC threshold was 1.253×10mm^(2)/s,positive likelihood ratio was4.25,negative likelihood ratio was 0.072,the sensitivity was 94.4%,the specificity was 77.8%.Conclusions:ADC threshold of1.106×10mm^(2)/s can distinguish benign and malignant breast lesions to the greatest extent,while ADC threshold of 1.253×10mm^(2)/s can reduce missed diagnosis of breast cancer.
作者
许凡勇
XU Fanyong(Department of Radiology,Songjiang Hospital Affiliated to Medicine School of Shanghai Jiao Tong University,Shanghai 201600,China)
出处
《中国中西医结合影像学杂志》
2022年第6期553-557,共5页
Chinese Imaging Journal of Integrated Traditional and Western Medicine
基金
上海市松江区科技攻关项目(20SJKJGG80)。
关键词
乳腺疾病
磁共振成像
表观扩散系数
阈值
诊断
鉴别
Breast disease
Magnetic resonance imaging
apparent diffusion coefficients
Threshold value
Diagnosis
difference