Predicting adverse outcomes of cardiac surgery with the application of artificial neural networks

Abstract

SummaryRisk-stratification models based on pre-operative patient and disease characteristics are useful for providing individual patients with an insight into the potential risk of complications and mortality, for aiding the clinical decision for surgery vs non-surgical therapy, and for comparing the quality of care between different surgeons or hospitals. Our study aimed to apply artificial neural networks (ANN) models to predict mortality and morbidity after cardiac surgery, and also to compare the efficacy of this model to that of the logistic regression model and Parsonnet score. The accuracy of the ANN, logistic regression and Parsonnet score in predicting mortality was 83.8%, 87.9% and 78.4%. The accuracy of the ANN, logistic regression and Parsonnet score in predicting major morbidity was 79.0%, 74.3% and 68.6%. The area under the receiver operating characteristic curves (AUC) of the ANN, logistic regression and Parsonnet score in predicting in-hospital mortality were 0.873, 0.852 and 0.829. The AUCs of the ANN, logistic regression and Parsonnet score in predicting major morbidity were 0.852, 0.789 and 0.727. The results showed the ANN models have the best discriminating power in predicting in-hospital mortality and morbidity among these models. Advances in medical treatment, the advent of thrombolytic therapy, and the availability of a range of percutaneous angiographic interventions has changed the profile of patients referred for cardiac surgery [1,2]. Referred patients tend to be older, with a substantial increase in the proportion of high-risk patients [3][4][5]. Risk-stratification models based on pre-operative patient and disease characteristics are useful for providing patients with an insight into the potential risk of complications and mortality, for aiding the clinical decision for surgery vs non-surgical therapy, and for comparing the quality of care between different surgeons and hospitals.Parsonnet and colleagues first designed a risk stratification model for evaluation of the mortality after cardiac surgery in 1989 [6]. The Parsonnet score is a simple scoring system which derived from large international statistics tries, on the basis of pre-operative risk factors, to assess and predict the mortality of patients with coronary and heart valve operations. Thereafter, numerous multifactorial risk scores have been developed to predict the outcomes following cardiac surgery [7][8][9][10][11][12]. Most of the scoring systems have been designed to predict mortality, but postoperative morbidity is recognised as a major determinant of hospital costs and quality of life after surgery [13]. Relatively few studies have involved the prediction of postoperative morbidity following cardiac surgery [8,14].Logistic regression models the probability of some event occurring as a linear function of a set of predictor variables. The actual state of the dependent variable is determined by looking at the estimated probability.The artificial neural network (ANN) is a computational model with parallel non-linear pr...

Mortality Predicted Accuracy for Hepatocellular Carcinoma Patients with Hepatic Resection Using Artificial Neural Network

“…In all training sets and in most validation sets, accuracy, sensitivity, specificity, and AUROC were higher in the 1-, 3-, and 5-year survival models constructed by ANN than in those constructed by LR, which is consistent with other reports that ANN outperforms LR in both training [15, 31–35] and validation [14, 36, 37]. …”

Neural Network Prediction of ICU Length of Stay Following Cardiac Surgery Based on Pre-Incision Variables

“…They also provide an excellent introduction to the mathematical foundation and design of neural networks, and how they are suggested to simulate the “learning” and “generalization” properties of human neural networks. For example, such networks have been applied to diagnosis of coronary artery syndromes [ 2 , 3 ], interpretation of electrocardiograms [ 4 , 5 ] classification of hemodynamic states in pregnancy [ 6 ], prediction of intensive care unit length of stay (ICU LOS) in trauma patients[ 7 ], and prediction of an adverse outcome in cardiac surgery [ 8 ].…”

Addition of Frailty and Disability to Cardiac Surgery Risk Scores Identifies Elderly Patients at High Risk of Mortality or Major Morbidity

“…1 Advanced age is frequently accompanied by a larger burden of comorbid conditions and greater illness severity. In the setting of cardiac surgery, elderly patients are more likely to have extensive coronary artery disease and concomitant valvular disease and are more likely to require urgent or emergent surgery.…”