Skip to main content

Table 3 Classification results of the prospective study

From: Automated prediction of early spontaneous miscarriage based on the analyzing ultrasonographic gestational sac imaging by the convolutional neural network: a case-control and cohort study

Methods Characteristics Accuracy,% (95%CI) Sensitivity,% (95%CI) Specificity,% (95%CI) PPV,% (95%CI) NPV,% (95%CI) AUC (95%CI)
CNN VGG19 78.10 (76.21, 79.99) 80.39 (78.59, 82.18) 94.52 (89.15, 99.88) 94.89 (89.85, 99.88) 77.00(75.27, 78.72) 0.885 (0.846, 0.925)
ultrasound characteristics CRL 66.34 (56.65, 74.82) 66.67 (29.57, 90.75) 66.32 (56.32, 75.04) 11.11 (3.82, 25.91) 96.92 (88.83, 99.78) 0.665 (0.564–0.756)
HR 82.18 (73.49, 88.51) 50.00 (18.76, 81.24) 84.21 (75.46, 90.31) 16.67 (5.01, 40.05) 96.39 (89.47, 99.20) 0.671 (0.570–0.761)
CRL + HR 83.17 (74.59, 89.31) 50.00 (18.76, 81.24) 87.37 (79.06, 92.77) 20.00 (6.28, 45.95) 96.51(89.82, 99.23) 0.687 (0.587–0.775)
  1. VGG19: a model of the convolutional neural network
  2. Auc Area Under Curve, CI Confidence Interval, CNN convolutional neural networks, CRL crown-rump length, HR heart rate, PPV positive predictive value, NPV negative predictive value