Fig. 2From: Combining radiomics with thyroid imaging reporting and data system to predict lateral cervical lymph node metastases in medullary thyroid cancerWorkflow of a radiomics approach for prediction of lateral cervical lymph node metastases (LCLNM) in medullary thyroid carcinoma from preoperative thyroid ultrasound. A Preprocessing of thyroid ultrasound images included manual segmentation, cropping and normalization. B Radiomics features were extracted from preprocessed thyroid ultrasound images. C A two-step method of Student’s t-test and least absolute shrinkage and selection operator (LASSO) regression were used to select features for modeling. D A regularized logistic regression model was developed and validated in 5-fold cross-validation to predict the LCLNM statusBack to article page