TABLE 5.

Classification of clinical isolates based on ML-based AMCA method.

Target	N	TP	TN a	FP	FN	SEN (%)	SPE (%)	Accuracy (CI)
blaIMP	45	45	32	0	0	100.0	100.0	100.0% (95.32–100.00%)
blaKPC	9	9	32	0	0	100.0	100.0	100.0% (91.40–100.00%)
blaNDM	74	73	32	1 b	0	100.0	96.97	99.06% (94.86–99.98%)
blaOXA-48	84	84	32	0	0	100.0	100.0	100.0% (96.87–100.00%)
blaVIM	8	8	32	0	0	100.0	100.0	100.0% (91.19–100.00%)
blaOXA-48 and blaNDM	1	1	32	0	0	100.0	100.0	100.0% (97.24–100.00)
Total	221	220	32	1	0	100.0	96.97	99.60% (97.82 to 99.99%)

Abbreviations-N, Number of samples; TP, True Positive; TN, True Negative; FP, False Positive; FN, False Negative; SEN, Sensitivity; SPE, Specificity; CI, Confidence Interval.

^aA total 32 negatives samples are considered across all the groups for sensitivity, specificity and accuracy calculation.

^bThis isolate was misclassified as bla_NDM, and bla_KPC, double infection.