影像组学联合剪切波弹性成像预测甲状腺乳头状癌颈部中央区淋巴结转移的价值  

The value of imaging omics combined with shear wave elastography in predicting cervical lymph node metastasis in papillary thyroid carcinoma

作　　者：陈华彬[1] 黄晓文 曾福强[1] 邹斌[1] 林勇平[1] 张路生 李程[1] 陈文娟[1] CHEN Huabin;HUANG Xiaowen;ZENG Fuqiang;ZOU Bin;LIN Yongping;ZHANG Lusheng;LI Cheng;CHEN Wenjuan(Department of Ultrasound,Zhongshan Hospital of Traditional Chinese Medicine,Zhongshan 528400,China)

机构地区：[1]广东省中山市中医院超声科,广东中山528400

出　　处：《医学影像学杂志》2023年第11期1964-1967,共4页Journal of Medical Imaging

基　　金：广东省中山市社会公益与基础研究(医疗卫生)项目(编号:2022B1136)。

摘　　要：目的探讨影像组学联合剪切波弹性成像(SWE)预测甲状腺乳头状癌(PTC)颈部中央区淋巴结转移的价值。方法选取178例PTC患者为模型组,比较发生与未发生中央区淋巴结转移患者SWE参数差异,使用GE影像组学软件分析超声表现特征,建立预测PTC颈部中央区淋巴结转移的模型。另选取109例PTC患者为验证组,验证基于超声影像组学特征和SWE参数建立的预测模型的评估效能。结果模型组颈部中央区淋巴结转移发生率43.26%,验证组发生率39.45%。超声影像组学经LASSO回归最终筛选7项与PTC颈部中央区淋巴结转移的特征。以影像组学评分和Emax联合预测PTC颈部中央区淋巴结转移的AUC为0.853。校准曲线显示,模型预测与病理结果的吻合度良好(P=0.811);决策曲线显示,模型的有效阈值概率区间为0~0.36,净获益值最大约0.082。结论基于影像组学建模联合SWE参数预测PTC颈部中央区淋巴结转移的效能良好,且基于预测结果开展淋巴结清扫能够获得更为理想的临床效果。Objective To explore the value of imaging omics combined with shear wave elastography(SWE)in predicting cervical lymph node metastasis in papillary thyroid carcinoma(PTC).Methods 178 PTC patients were selected and included in the model group.The differences in SWE parameters between patients with and without central lymph node metastasis were compared.We analyzed imagomics features using GE imagomics software and established a model for predicting lymph node metastasis in the central cervical region of PTC.Another 109 PTC patients in the same period were selected as the validation group to verify the evaluation effectiveness of the prediction model based on ultrasound imaging omics features and SWE parameters.Results The incidence of lymph node metastasis in the central region of the neck in the model group was 43.26%;The incidence of validation group was 39.45%.Through LASSO regression,seven features related to lymph node metastasis in the central region of the neck of PTC were finally screened for ultrasound imaging omics.The AUC for predicting lymph node metastasis in the central cervical region of PTC using a combination of imaging omics score and Emax was 0.853.The calibration curve showed a good agreement between the model prediction and pathological results(P=0.811);The decision curve showed that the effective country probability interval of the model was between 0 and 0.36,with a maximum net profit value of approximately 0.082.Conclusion It is effective to predict lymph node metastasis in the central cervical region of PTC based on imaging modeling and SWE parameters,and more ideal clinical benefits can be obtained by carrying out lymph node dissection based on the predicted results.

关 键 词：甲状腺乳头状癌 淋巴结转移 剪切波弹性成像 影像组学 

分 类 号：R736.1[医药卫生—肿瘤] R445[医药卫生—临床医学]