目的 探究基于常规超声的弹性相对指数（elasticity contrast index，ECI）联合S-Detect对甲状腺乳头状癌（PTC）颈部淋巴结转移（CLNM）的预测价值。 方法 回顾性分析2023年11月至2024年8月在石河子大学第一附属医院进行诊治的135例甲状腺乳头状癌患者并记录所有患者的临床基线资料、常规超声、超声弹性成像及S-Detect分析结果，筛选出PTC 发生 CLNM 的独立预测因子并构建预测模型。绘制受试者工作特征曲线并比较预测模型的曲线下面积，选取最佳预测模型构建风险概率列线图，分析列线图模型的预测效能及临床适用性。 结果 年龄、结节最大径、边界、包膜侵犯、S-Detect横切面识别的形状结果与ECI弹性分级为预测PTC发生CLNM 的独立预测因子（均P < 0.05），6个指标联合建立预测模型的AUC为0.890（95%CI： 0.835 ~ 0.945），校准曲线结果表明一致性良好，临床决策曲线分析表明阈值概率为2% ~ 91.5%时，列线图模型可获得较好的净收益。 结论 年龄、结节最大径、边界、包膜侵犯、S-Detect横切面识别的形状结果与ECI弹性分级为预测PTC发生CLNM的独立预测因子，以上指标联合建立的列线图模型可有效预测 CLNM 的发生概率。

Objective To evaluate the predictive value of dual-modality ultrasound， incorporating conventional ultrasound and ultrasound elastography， in combination with S-Detect for cervical lymph node metastasis （CLNM） in patients with papillary thyroid carcinoma （PTC）. Methods A retrospective analysis was conducted on the clinical data of 135 patients diagnosed with PTC who received treatment at the First Affiliated Hospital of Shihezi University between November 2023 and August 2024. For all patients， clinical baseline characteristics， conventional ultrasound findings， ultrasound elastography results， and S-Detect analysis data were collected. Independent predictors of CLNM in PTC were identified， and predictive models were developed. Receiver operating characteristic （ROC） curves were generated to compare the area under the curve （AUC） of the models. The most effective predictive model was selected to construct a risk probability nomogram， and the predictive performance and clinical applicability of this nomogram were subsequently evaluated. Results Age， maximum nodule diameter， boundary characteristics， capsular invasion， transverse-sectional morphological findings assessed by S-Detect， and ECI-based elasticity grading were identified as independent predictors of CLNM in PTC （all P < 0.05）. The AUC of the predictive model constructed using these six variables was 0.890 （95%CI： 0.835 ~ 0.945）. The calibration curve demonstrated strong agreement between predicted and observed outcomes， and decision curve analysis indicated that the nomogram provided a favorable net clinical benefit within a threshold probability range of 2% to 91.5%. Conclusions Age， maximum nodule diameter， boundary characteristics， capsular invasion， sonographic features assessed by S-Detect in the transverse plane， and ECI-based elasticity grading are independent predictors of CLNM in PTC. A nomogram model incorporating these parameters demonstrates effective performance in predicting the likelihood of CLNM.