Table 2.

Performance metrics of the machine learning algorithms

	ROC AUC	PR AUC	Accuracy	Sensitivity	Specificity	NPV	PPV	F1 score	p value
TE75	0.64 (0.42–0.82)	0.63 (0.40–0.83)	0.62 (0.45–0.79)	0.80 (0.57–1.00)	0.43 (0.15–0.71)	0.67 (0.30–1.00)	0.60 (0.37–0.80)	0.69 (0.48–0.83)	0.092
TE300	0.57 (0.33–0.78)	0.52 (0.28–0.70)	0.42 (0.23–0.62)	0.30 (0.08–0.58)	0.54 (0.27–0.80)	0.44 (0.18–0.69)	0.40 (0.10–0.73)	0.33 (0.10–0.56)	0.312
b0	0.66 (0.46–0.84)	0.73 (0.48–0.89)	0.55 (0.39–0.71)	0.81 (0.60–1.00)	0.26 (0.07–0.50)	0.60 (0.17–1.00)	0.54 (0.35–0.74)	0.65 (0.44–0.81)	0.069
b10	0.53 (0.33–0.74)	0.58 (0.37–0.79)	0.49 (0.31–0.66)	0.68 (0.47–0.89)	0.24 (0.06–0.47)	0.40 (0.09–0.71)	0.52 (0.32–0.71)	0.59 (0.40–0.75)	0.415
b200	0.79 (0.61–0.93)	0.72 (0.51–0.91)	0.68 (0.52–0.84)	0.56 (0.31–0.81)	0.81 (0.57–1.00)	0.63 (0.41–0.84)	0.75 (0.50–1.00)	0.64 (0.40–0.82)	0.002
b800	0.64 (0.44–0.82)	0.74 (0.55–0.88)	0.56 (0.42–0.72)	0.64 (0.40–0.83)	0.47 (0.23–0.71)	0.53 (0.27–0.79)	0.57 (0.37–0.77)	0.59 (0.40–0.76)	0.071
ADC	0.72 (0.50–0.90)	0.75 (0.54–0.94)	0.68 (0.53–0.82)	0.80 (0.60–0.95)	0.54 (0.27–0.78)	0.69 (0.36–0.91)	0.68 (0.50–0.87)	0.74 (0.56–0.87)	0.019