Supplementary Data from A Combined Nomogram Model to Preoperatively Predict Histologic Grade in Pancreatic Neuroendocrine Tumors

Figure S1. The recruitment pathway in this study. Table S1. Clinical characteristics of patients with pNETs in the training and validation set. Figure S2. Schematic of the undecimated three-dimensional wavelet transform applied to each CT image. Table S2. Radiomics features extracted in this study. Figure S3. The construction procedure of radiomics signature model. Figure S4. The correlation matrix for the selected radiomics features and clinical characteristics. Figure S5. The correlation matrix for the selected radiomics features and Ki-67 Index/rate of nuclear mitosis. Figure S7. Scatter plots for selected radiomics features and rate of nuclear mitosis. Figure S8. Scatter plots for radiomics signature/nomogram and the Ki-67 Index/rate of nuclear mitosis. Table S3. Predictive performance of the radiomics signature incorporating maximum diameter and tumor clinical stage. Table S4. Multivariable Regression Results for Radiomics Signature.