Design and Assessment of a Robust and Generalizable ANN-Based Classifier for the Prediction of Premature Birth by means of Multichannel Electrohysterographic Records

Figure 5

Box and whisker plots of the AUCs obtained on the test datasets by the different classifiers in the 30 iterations on each of the 10 SMOTE-balanced datasets (a). value associated with each comparison between AUC distributions of the different classifiers when ANOVA analysis is performed (b) (bold figures indicate statistically significant difference, ).

(a)

(b)