Figure 1From: A molecular computational model improves the preoperative diagnosis of thyroid nodules Expression mean of 49 malignant (red) and 38 benign (green) samples for each marker (A). ROC analysis for KIT, CDH1, LSM7, C21orf4, DDI2 separately. Among the markers, KIT resulted to be the most powerful in discriminating benign from malignant thyroid tumors (AUC = 0.9) (B). ROC analysis for KIT, CDH1, LSM7, C21orf4, DDI2, and BRAF status in combination (AUC = 0.88) (C).Back to article page