[Retracted] Prediction of Changes in Tumor Regression during Radiotherapy for Nasopharyngeal Carcinoma by Using the Computed Tomography-Based Radiomics

Figure 3

Features with an image ICC greater than 0.75, the results of feature screening using the LASSO method. (a) shows the centralized convergence diagram of the feature, the abscissa was the Lambda value, and the ordinate was the feature coefficient; (b) shows the model selection and result of feature selection, the abscissa was the Lambda value, and the ordinate was the variance. According to the principle of the most concise model, 3 features were screened.)

(a)

(b)