Table 2.

Comparison of diagnostic performance metrics between the ML system model and reference interval in suspected cases.

Methods	TNs (n)	TPs (n)	FNs (n)	FPs (n)	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	Accuracy (%)
ML system model	22	142	10	6	93.42	78.57	95.95	68.75	91.11
Reference interval	2	144	8	26	94.74	7.14	84.71	20	81.11

ML, machine learning; TNs, true-positives; TPs, true-negatives; FNs, false-negatives; FPs, false-positives; PPV, positive predictive value; NPV, negative predictive value.