Objective: To investigate the diagnostic value of radiomics features derived from iodine map in diagnosing lateral cervical lymph node metastasis (LLNM) in patients with papillary thyroid cancer (PTC). Methods: A retrospective analysis was performed on the clinical and imaging data of 200 patients with PTC who underwent preoperative dual-energy CT(DECT) at the First Affiliated Hospital with Nanjing Medical University between June 2022 and June 2024, all with pathological confirmation. A total of 777 lymph nodes were included, comprising 379 metastatic lymph nodes and 398 non-metastatic lymph nodes. Patients were randomly allocated into training and validation sets at a ratio of 7:3. Clinical and CT features of lymph nodes,including size, shape, edge, enhancement, calcification, cystic change, extranodal extension, were enrolled. Quantitative DECT parameters including iodine concentration (IC), normalized IC (NIC), effective atomic number (Zeff), normalized Zeff (NZeff) and slope of energy spectrum curve (λHU) in arterial and venous phase. Radiomics features were manually extracted from arterial and venous phase iodine maps by sketch Region of interest (ROI). Results: All clinical and DECT features showed significant difference between LLNM and Non-LLNM (P<0.05) except for extranodal extension. Compared with the Conventional imaging model (training set, AUC=0.803，95% CI 0.765 to 0.841; validation set, AUC=0.804,95% CI 0.748 to 0.861), the DECT model (training set, AUC=0.939,95% CI 0.918 to 0.960; validation set, AUC=0.933,95% CI 0.901 to 0.965), radiomics model (training set, AUC=0.977,95% CI 0.966 to 0.988; validation set, AUC=0.931,95% CI 0.900 to 0.962) and combined model (training set, AUC=0.977,95% CI 0.965 to 0.988; validation set, AUC=0.938,95% CI 0.908 to 0.968) showed excellent diagnostic performance in both training and validation set . In training set, radiomics model was superior to combined model (P<0.01), while no significant difference was observed between the radiomics and combined models overall(P=0.950). Conclusion: Radiomics features based on iodine mapping extraction have high diagnostic value in diagnosing LLNM in PTC patients.