Table 3 Descriptive data of the performance metrics (accuracy, precision, recall and specificity) quantified in this study.

From: Binary decisions of artificial intelligence to classify third molar development around the legal age thresholds of 14, 16 and 18 years

Age threshold

Sex

Accuracy

Precision

Recall

Specificity

14 years

Females

0.863

0.943

0.780

0.929

Males

0.872

0.900

0.849

0.842

16 years

Females

0.828

0.850

0.831

0.802

Males

0.883

0.917

0.831

0.896

18 years

Females

0.829

0.836

0.860

0.768

Males

0.939

0.962

0.901

0.927

  1. Total training sample = 9.680; total validation sample = 1.960.
Back to article page