Skip to main content
. 2026 Mar 28;16:10811. doi: 10.1038/s41598-026-42492-1

Table 4.

Show the results of EfficientNetB0.

Fold Class Precision Recall F1-Score ROC-AUC
Fold(1) No Finding 0.71 0.75 0.73 0.87
Pneumonia 0.72 0.86 0.79 0.95
Pneumothorax 0.48 0.83 0.61 0.90
Effusion 0.69 0.82 0.75 0.88
Cardiomegaly 0.67 0.85 0.75 0.95
Fold(1)—validation accuracy 0.89 – – – –
Fold(2) No Finding 0.68 0.76 0.72 0.87
Pneumonia 0.70 0.89 0.78 0.95
Pneumothorax 0.52 0.79 0.63 0.89
Effusion 0.70 0.78 0.74 0.87
Cardiomegaly 0.59 0.89 0.71 0.95
Fold(2)—validation accuracy 0.88 – – – –
Fold(3) No Finding 0.68 0.76 0.72 0.87
Pneumonia 0.65 0.84 0.73 0.93
Pneumothorax 0.52 0.81 0.63 0.89
Effusion 0.66 0.82 0.73 0.87
Cardiomegaly 0.59 0.88 0.71 0.94
Fold(3)—validation accuracy 0.89 – – – –
Class average No Finding 0.69 0.76 0.72 0.87
Pneumonia 0.69 0.87 0.77 0.94
Pneumothorax 0.51 0.81 0.63 0.89
Effusion 0.69 0.81 0.74 0.88
Cardiomegaly 0.62 0.87 0.72 0.95
Class standard deviation No Finding 0.0101 0.0031 0.0041 0.0030
Pneumonia 0.0310 0.0224 0.0254 0.0101
Pneumothorax 0.0188 0.0179 0.0095 0.0046
Effusion 0.0159 0.0184 0.0070 0.0049
Cardiomegaly 0.0355 0.0133 0.0193 0.0049
Average validation accuracy 0.89 – – – –