Skip to main content

Table 4 Classification table of discriminant analysis. Predictive power of KIT, TC1, miR-222, and miR-146b for discriminating malignant from benign FNA. This procedure is designed to develop a set of discriminating functions which can help predict malignant vs. benign status based on the values of other quantitative variables; 51 cases were used to develop a model to discriminate among the two levels of malignant vs. benign; four predictor variables were entered. Amongst the 51 observations used to fit the model, 47 % or 92.16 % were correctly classified

From: The combination of four molecular markers improves thyroid cancer cytologic diagnosis and patient management

Actual mal_ben Group size Predicted
   Benign Malignant
Benign 13 9 4
(69.23 %) (30.77 %)
Malignant 38 0 38
(0.00 %) (100.00 %)
  1. Classification variable: Malignant vs Benign
  2. Independent variables: KIT, TC1, miR-222, miR-146