超声影像组学标签对乳腺癌腋窝淋巴结转移的预测价值

目的探讨基于常规超声的影像组学特征预测乳腺癌腋窝淋巴结转移的应用价值。 方法回顾性收集2020年1月至2020年10月于中山大学肿瘤防治中心就诊经手术病理确诊的265例乳腺癌患者的临床资料和术前超声图像，按超声检查时间顺序，将患者分为训练集（159例）和验证集（106例）。应用ImageJ软件手动勾画病灶区域，使用Pyradiomics从每个病灶区域中提取影像组学特征，采用多种方法逐步筛选特征，应用Logistic回归构建预测乳腺癌腋窝淋巴结转移的超声影像组学标签。在训练集和验证集上采用ROC曲线、校准曲线和决策曲线评估超声影像组学标签预测乳腺癌腋窝淋巴结转移的效能。 结果最终筛选出8个关键超声影像组学特征用于构建超声影像组学标签。该标签在训练集和验证集中预测乳腺癌腋窝淋巴结转移的ROC曲线下面积分别为0.805（95%CI：0.734~0.876）、0.793（95%CI：0.706~0.880）。在校准曲线中，该标签在训练集和验证集均表现出较好的校准度（P=0.592、0.593），决策曲线分析进一步表明了该标签具有一定的临床实用性。 结论基于超声的影像组学标签在预测乳腺癌腋窝淋巴结转移方面具有一定价值，可为治疗前乳腺癌的准确分期以及治疗方案的合理选择提供参考依据。

Ultrasound-based radiomics to predict axillary lymph node metastasis in breast cancer

ObjectiveTo explore the value of a radiomics model based on ultrasound imaging in predicting the axillary lymph node status of patients with breast cancer. MethodsA total of 265 patients with early-stage breast cancer were retrospectively analyzed, all of whom underwent preoperative breast ultrasound examination at Sun Yat-sen University Cancer Center from January 2020 to October 2020. According to the order of examination time, the patients were divided into a training group (n=159) and a validation group (n=106). ImageJ software was used to manually delineate the lesion area in the ultrasound image along the tumor boundary. Pyradiomics was used to extract 1130 features from each lesion area, and three statistical methods were used to screen the features. Finally, a logistic regression model was used to construct ultrasound imaging radiomics model. The receive operating characteristic (ROC) curve, calibration curve, and decision curve were used to evaluate the performance and value of the ultrasound imaging radiomics model in predicting axillary lymph node status. ResultsA total of eight key image features were selected to construct the ultrasound imaging radiomics model. The area under the ROC curve values of the model in the training group and the validation group were 0.805 (95% confidence interval [CI]: 0.734-0.876) and 0.793 (95% CI: 0.706-0.880), respectively. The calibration curve showed that the model had a good calibration in both the training and validation groups (P=0.592、0.593); besides, the decision curve analysis confirmed that the model had some clinical practicability. ConclusionUltrasound-based imaging radiomics model is of great value in predicting the axillary lymph node status of patients with breast cancer before surgery, which could guide clinicians in the accurate staging of breast cancer and selection of appropriate therapeutic regimen.