目的:回顾性分析外侧象限乳腺癌患者乳腺原发肿瘤及腋窝淋巴结的超声特征,并构建列线图模型,为临床评估外侧象限乳腺癌患者腋窝淋巴结转移提供影像学依据.方法:回顾性分析无锡市锡山人民医院经病理证实的127例外侧象限乳腺癌患者腋窝淋巴结及乳腺原发肿瘤的超声影像学特征.伴腋窝淋巴结转移者分入阳性组(54例),不伴腋窝淋巴结转移者分入阴性组(73例).采用单变量和多变量Logistic回归分析,筛选淋巴结转移的危险因素.使用R语言将数据集随机分成训练集和验证集,基于训练集构建列线图预测模型,预测腋窝淋巴结转移风险,并在验证集中验证.受试者工作特征(receiver operating characteristic,ROC)曲线用于评估诊断性能,校正曲线和Hosmer-Lemeshow检验用于评估预测值与实际列线图预测值的一致性.结果:肿瘤针状边缘(OR=4.16,95％CI:1.25～13.79)和淋巴门结构不清晰(OR=19.20,95％CI:1.98～186.36)是外侧象限乳腺癌患者发生腋窝淋巴结转移的独立危险因素.据此构建预测外侧象限乳腺癌腋窝淋巴结转移的列线图模型.ROC曲线显示,训练集的曲线下面积(area under curve,AUC)为0.74(0.62～0.86),验证集AUC为0.73(0.62～0.84).训练集和验证集的Hosmer-Lemeshow检验分别为P=0.570和P=0.552.结论:超声有助于术前外侧象限乳腺癌患者腋窝淋巴结转移情况的评估;基于Logistic回归构建的列线图预测模型具有良好的安全性、可靠性和实用性.
Development a nomogram predictive model for axillary lymph node metastasis in lateral quadrant breast cancer based on B-ultrasound features
Objective:To retrospectively analyze the ultrasound features of primary breast tumors and axillary lymph nodes in patients with lateral quadrant breast cancer,and to construct a nomogram model to provide imaging evidence for better clinical assessment of axillary lymph node metastasis in these patients.Methods:We retrospectively analyzed ultrasonographic features of theaxillary lymph nodes and primary breast tumors in 127 patients with lateral quadrant breast cancer,confirmed by pathology at Xishan People's Hospital of Wuxi City.Patients with axillary lymph node metastasis were categorized into the positive group(54 cases),while those without axillary lymph node metastasis were classified into the negative group(73 cases).Univariate and multivariate logistic regression analyses were performed to identify therisk factors for lymph node metastasis.The dataset was randomly divided into a training set and a validation set using the R language.A nomogram prediction model was constructed based on the training set to predict the risk of axillary lymph node metastasis and validated in the validation set.Diagnostic performance was evaluated using receiver operating characteristic(ROC)curves,while calibration curves and the Hosmer-Lemeshow test were used to assess the consistency between thepredicted and actual values of the nomogram.Results:Tumor spiculated margin(OR=4.16,95％CI:1.25-13.79)and unclear lymphatic gate structure(OR=19.20,95％CI:1.98-186.36)were identified as independent risk factors of axillary lymph node metastasis in patients with lateral quadrant breast cancer.Furthermore,a nomogram model was developed to predict axillary lymph node metastasis in lateral quadrant breast cancer cases.The ROC curves howed that the area under the curve(AUC)for the training set was 0.74(95％CI:0.62-0.86)and the AUC for the validation set was 0.73(95％CI:0.62-0.84).Hosmer-Lemeshow test results indicated no significant deviation from goodness-of-fit for both the training set and validation set with P-values of 0.570 and 0.552,respectively.Conclusion:Ultrasound plays a valuable role in the assessment of axillary lymph node metastasis in patients with lateral quadrant breast cancer.The nomogram prediction model based on logistic regression demonstrates good safety,reliability,and practicality for clinical use.
ultrasoundlateral quadrant breast canceraxillary lymph node metastasisnomogram modellogistic regression
万方数据
维普
