基于超声特征联合临床资料的预测模型对甲状腺癌淋巴结转移的评估价值OA
The evaluation value of a prediction model based on ultrasonic features combined with clinical data for lymph node metastasis of thyroid cancer
目的 探究超声特征联合临床资料的预测模型对甲状腺癌淋巴结转移的评估价值.方法 本研究选取2021年10月～2024年3月于佛山市第二人民医院诊断为甲状腺癌的200例患者作为研究对象.根据是否发生淋巴转移将患者分为转移组(n=101)和非转移组(n=99).应用Spearman相关分析超声特征参数和血清指标之间的相关性.采用Logistic回归分析甲状腺癌淋巴结转移的独立影响因素,应用受试者工作特征(ROC)曲线和曲线下面积(AUC)评估超声特征联合血清指标对甲状腺癌淋巴结转移的诊断效能.结果 两组患者的结节病灶数目、结节直径、钙化纵横比、血流参数[包括峰值收缩期血流速度(PSV)、搏动指数(PI)、阻力指数(RI)],内部血流情况以及结节与被膜接触面积,差异均有统计学意义(P<0.05).游离三碘甲状腺原氨酸(FT3)和PSV、PI、RI之间存在负相关关系(P<0.05),而游离甲状腺素(FT4)、促甲状腺激素(TSH)、甲状腺球蛋白(TG)与PSV、PI、RI存在正相关关系(P<0.05).Logistic回归分析结果表明,结节直径、PSV、PI、RI、结节与被膜接触面积、TSH、TG均为甲状腺癌淋巴结转移的独立影响因素(P<0.05).ROC曲线分析结果显示,超声特征参数与血清指标联合诊断时AUC、准确度、敏感度、特异度分别为0.887、86.28％、85.31％、86.64％.二者联合诊断甲状腺癌淋巴结转移的价值高于单一指标诊断(P<0.05).结论 超声特征联合临床资料的预测模型应用于甲状腺癌淋巴结转移患者的诊断效能较高,可根据超声特征参数以及血清指标对患者的病情予以判断.
OBJECTIVE To explore the evaluation value of a prediction model combining ultrasonic features and clinical data for lymph node metastasis of thyroid cancer.METHODS In this study,200 patients diagnosed with thyroid cancer in our hospital from October 2021 to March 2024 were selected as the research subjects.The patients were divided into a metastasis group(n=101)and a non-metastasis group(n=99)according to whether lymph node metastasis occurred.Spearman correlation analysis was used to investigate the correlation between ultrasound characteristic parameters and serum indicators.Logistic regression analysis was employed to identify the independent influencing factors for lymph node metastasis of thyroid cancer.The receiver operating characteristic(ROC)curve and the area under the curve(AUC)were utilized to evaluate the diagnostic efficacy of the combination of ultrasound characteristics and serum indicators for lymph node metastasis of thyroid cancer.RESULTS The number of lesions,nodule diameter,aspect ratio of calcified foci,blood flow parameters[peak systolic velocity(PSV),pulsatility index(PI),and resistance index(RI)],internal vascularity patterns,and contact area between nodules and the capsule differed significantly between the two patient groups(P<0.05).There is a negative correlation between FT3 and PSV,PI,and RI(P<0.05),while FT4,TSH,and TG have positive correlations with PSV,PI,and RI(P<0.05).The results of Logistic regression analysis indicate that nodule diameter,PSV,PI,RI,the contact area between the nodule and the capsule,TSH,and TG are all independent influencing factors for lymph node metastasis of thyroid cancer(P<0.05).The results of ROC curve analysis show that the combined diagnosis of ultrasound characteristic parameters and serum indicators has an AUC,accuracy,sensitivity,and specificity of 0.887,86.28％,85.31％,and 86.64％,respectively.The value of combined diagnosis of thyroid cancer lymph node metastasis by these two indicators is higher than that of single indicator diagnosis(P<0.05).CONCLUSION The prediction model combining ultrasonic features and clinical data has a relatively high accuracy when applied to the diagnosis of patients with lymph node metastasis of thyroid cancer.The patient's condition can be judged based on the ultrasonic characteristic parameters and serum indicators.
卢咏峰;朱娟芬;白伟波
佛山市第二人民医院 超声医学科,广东 佛山 528000佛山市第二人民医院 超声医学科,广东 佛山 528000佛山市第二人民医院 甲状腺疝血管小儿外科,广东 佛山 528000
甲状腺癌超声特征预测模型淋巴结转移评估价值
Thyroid Neoplasmsultrasonic featuresprediction modellymph node metastasisevaluation value
《中国耳鼻咽喉头颈外科》 2026 (2)
72-76,5
佛山市自筹经费类科技创新项目(2320001007323)
评论