Skip to main content
. 2022 Jul 12;12:11803. doi: 10.1038/s41598-022-16154-x

Table 5.

Classifications scores of Model A and Model B. Values (except F1-score and AUC) are shown in percentage. The Confidence level was set to 95%. (P) = Pediatric patients only, (A) = Adult patients only.

Model Precision Sensitivity Specificity Accuracy F1-score AUC
Model A

86.8%

(83.3–90.3)

88.5%

(87.0–90.1)

90.2%

(87.1–93.3)

89.5%

(88.0–91.0)

0.876

(0.86–0.89)

0.951

(0.94–0.96)

Model B

77.9%

(75.1–80.8)

82.2%

(76.8–87.5)

83.1%

(79.4–86.7)

82.7%

(81.8–83.6)

0.797

(0.78–0.81)

0.906

(0.89–0.91)

Model A, (P)

86.4%

(83.7–89.1)

84.9%

(82.4–87.3)

85.3%

(81.7–88.9)

85.1%

(83.6–86.5)

0.856

(0.84–0.87)

0.924

(0.91–0.93)

Model A, (A)

87.3%

(82.7–91.8)

91.7%

(89.9–93.4)

92.4%

(89.2–95.6)

92.1%

(90.2–94.0)

0.893

(0.87–0.92)

0.966

(0.96–0.97)

Model B, (P)

83.4%

(81.0–85.7)

75.7%

(73.0–78.3)

83.3%

(80.1–86.4)

79.3%

(77.7–81.0)

0.793

(0.78–0.81)

0.866

(0.85–0.88)

Model B, (A)

83.3%

(80.4–86.2)

78.0%

(74.5–81.5)

91.3%

(89.1–93.6)

86.7%

(85.8–87.5)

0.804

(0.79–0.82)

0.924

(0.92–0.93)