Skip to main content

Table 2 Diagnostic performance metrics of the CCH-NET model in test_1 and test_2

From: Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study

Type

Total sample size

Acc

Pre

Rc

F1-Score

Sen

Sc

Positive predictive rate

Negative predictive rate

Test_1

36

0.778

0.778

0.778

0.778

0.778

0.778

0.778

0.778

Test_2

48

0.727

0.692

0.750

0.720

0.750

0.667

0.692

0.727

Test_total

84

0.738

0.727

0.762

0.744

0.762

0.714

0.727

0.750