Table 3 Performance of the deep-learning system on the external retrospective 184 videos in 184 patients.

From: A deep-learning pipeline to diagnose pediatric intussusception and assess severity during ultrasound scanning: a multicenter retrospective-prospective study

Patients

Accuracy (95%CI)

Sensitivity (95%CI)

Specificity (95%CI)

AUC (95%CI)

FK (95%CI)

Average-AUC (95%CI)

FPS Median (range)

NON

0.978 (0.978–0.978)

0.929 (0.833–1.000)

0.987 (0.970–1.000)

0.958 (0.909–1.000)

0.916 (0.834–0.997)

0.956 (0.0.913–0.998)

91(83–101)

NSI

0.951 (0.951–0.952)

0.949 (0.912–0.986)

0.957 (0.900–1.000)

0.953 (0.919–0.988)

0.876 (0.797–0.955)

SIP

0.962 (0.962–0.962)

0.947 (0.847–1.000)

0.964 (0.935–0.992)

0.956 (0.902–1.000)

0.816 (0.684–0.948)

Back to article page