Artificial intelligence as a precision tool for predicting risk of in-hospital death after coronary artery bypass graft surgery

Abstract

Abstract Background/Introduction The prediction of risk of in-hospital death associated with cardiac surgery still has important gaps. In this scenario, the computational tools and mathematical techniques that constitute data science and provide machine learning, pillars of artificial intelligence, can represent an effective solution to this problem. Purpose To develop an in-hospital death prediction model for isolated CABG based on an artificial intelligence constituted by an artificial neural network (ANN). Methods 3,124 patients consecutively submitted to isolated CABG between 2010 and 2020 were included. Altogether, 30 baseline and operative variables were evaluated. Initially, the Extra Tree Classifier machine learning algorithm was used to select the attributes with the highest association with death. With the application of the algorithm, it was possible to identify the 13 variables with the greatest weight associated with hospital death. After selecting the variables and dividing the dataset into training (70%) and testing (30%), a risk prediction model was structured through an ANN with multiple layers. The ReLU activation function was used in the hidden layers and the SoftMax activation function was used in the output layer to extract the specific probability of death and survival. As an optimizing function of the ANN, the Nadam function was used. In addition, a thousand cycles of propagation and data return (Epochs) were performed to induce machine learning based on the cyclic adjustment of the weights of each of the independent variables included in the model. Accuracy assessments were performed using the ROC curve in the test dataset. The model was developed using the Python programming language. Results After consolidating machine learning based on the training dataset with 70% of the general sample, it was possible to observe that through the artificial intelligence technique, a predictive accuracy of 83.86% (AUC 0.8386) was obtained for the occurrence of in-hospital death in the test dataset. When comparing the performance of traditional risk scores, also tested only in the test dataset, we found that the ANN-based model was significantly superior to the scores (EuroScore I = 71.4% (AUC 0.714); EuroScore II = 71.9% (AUC 0.719), STS Score = 71.1% (AUC 0.714). The area under the curve of the model based on the ANN was significantly higher when compared to the areas of the scores using the DeLong test (p<0.05) Conclusion The application of artificial intelligence modelling is feasible for the creation of prediction models in the health area. In this study, the accuracy of the ANN was significantly higher than that of the other traditional risk scores. These findings demonstrate the great potential that representative datasets have when accessed through artificial intelligence techniques. The demand for massive volumes of information is mitigated when well-structured datasets with extreme data quality is used. Funding Acknowledgement Type of funding sources: None.