Abstract： Objective:To investigate the value of radiomics analysis of iodine maps derived from IQon spectral CT for preoperative diagnosing cervical lymph nodes metastasis in patients with papillary thyroid cancer(PTC).Methods:A retrospec-tive analysis was conducted on 89 patients with PTC who underwent preoperative neck spectral CT enhanced scans at the First Hospital of Shanxi Medical University from October 2019 to October 2022.A total of 267 lymph nodes were assessed,they were divided into a training set(108 non-metastatic lymph nodes and 78 metastatic lymph nodes)and a validation set(50 non-metastatic lymph nodes and 31 metastatic lymph nodes)in a 7∶3 ratio.Manual delineation of three-dimensional regions of interest for the lymph nodes was performed on iodine density maps in the arterial and venous phases using ITK-SNAP soft-ware.Radiomic features of each lymph node were extracted using FAE software.The least absolute shrinkage and selection operator regression(LASSO)method was used for radiomic feature selection.Logistic regression was used to establish diagnostic models based on CT image feature of the cervical lymph nodes,radiomics signature,and the combined.The diagnostic perfor-mance of the three models was evaluated using ROC curves and the area under the curve(AUC).Results:Model 1 consisted of lymph node short axis,degree of enhancement,calcification,and cystic changes.LASSO regression selected nine radiomic features and established radiomics signature(Model 2).Model 3 consisted of cervical lymph nodes degree of enhancement,cal-cification,cystic changes and radiomics signature.In both the training and validation sets,Model 3 showed the optimal diag-nostic performance(AUC=0.94,0.91).Conclusion:Radiomics analysis of IQon spectral CT-derived iodine maps can realize ef-fective diagnosis of cervical lymph nodes metastasis in patients with PTC.