Table 4.

Classification confusion matrix for results with LOO CV on the IMBALANCED data set using top 15 DTCWT features and three clinical features. The overall per-image classification accuracy achieved is 74.39%

		Predicted class
		Normal (%)	CLD (%)	CON (%)	PTX (%)	RDS (%)	TTN (%)
True Class	Normal	17.84	0	0	7.03	0	75.14
	CLD	0	100	0	0	0	0
	CON	0	0	99.02	0	0.98	0
	PTX	8.57	0	12.38	17.14	25.71	36.19
	RDS	2.86	0	8.98	5.31	57.55	25.31
	TTN	4.53	0	0	2.26	2.83	90.38

The key metrics, overall classification accuracy, the weighted F1-score, and the classification accuracy for each class were shown in bold

The weighted F1-score is 0.740