Skip to main content

Table 3 Performance of the Cox regression model and an ANN on 398 patients with SARS-CoV-2 in a West London hospital, during March 1–April 24, 2020

From: Comparison of deep learning with regression analysis in creating predictive models for SARS-CoV-2 outcomes

 

Cox regression model (95% CI)

ANN model (95% CI)

Cox regression vs ANN model

Sensitivity

50.0% (28.2–71.8)

64.7% (38.3–85.8)

 

Specificity

96.6% (88.1–99.6)

96.8% (89–99.6)

 

Positive predictive value

84.6% (57.0–95.8)

84.6% (57.4–95.7)

 

Negative predictive value

83.6% (77.0–88.6)

91.4% (84.2–95.1)

 

Accuracy

83.75% (73.8–91.1)

90.0% (81.2–95.6)

 

AUROC

86.9% (85.7–88.2)

92.6% (91.1–94.1)

Z = 12.021, p < 0.001

  1. ANN artificial neural network, AUROC area under the receiver operating characteristic curve
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us