Coronary Artery Bypass Risk Prediction Using Neural Networks Richard P Lippmann, PhD, David M Shahian, MD The Annals of Thoracic Surgery Volume 63, Issue 6, Pages 1635-1643 (June 1997) DOI: 10.1016/S0003-4975(97)00225-7
Fig. 1 Two-layer, multilayer perceptron neural network using random weight initialization and back propagation. The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)
Fig. 2 Block diagram of risk prediction system with confidence intervals. The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)
Fig. 3 (A) Receiver operating characteristic curve for committee classifier. Area (C-index) = 76.4%. (B) Fifty superimposed receiver operating characteristic curves generated using bootstrap sampling. Average receiver operating characteristic curve area (AVE AREA) = 75.4% ± 0.3%, suggesting little variability in neural network output secondary to random weight initialization and stochastic descent gradient training. The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)
Fig. 4 Calibration, by mortality bins, of four classifiers. (MLP = multilayer sigmoid neural network.) The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)
Fig. 5 Block diagram of bootstrap method for determination of confidence intervals for multilayer perceptron. (MLP = multilayer sigmoid neural network.) The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)
Fig. 6 Confidence intervals for two-layer multilayer perceptron classifier. The Annals of Thoracic Surgery 1997 63, 1635-1643DOI: (10.1016/S0003-4975(97)00225-7)