Preoperational Time Prediction for Percutaneous Coronary Intervention Using Machine Learning Techniques

This paper addresses the prediction of preoperational time for patients with the acute coronary syndrome. Health records contain personal information, life and disease anamnesis, test results. Using this data, we tried to predict time before the coronary stent operation with regression methods. During the preprocessing, we divided health records into three clusters with k-means method and compared the results of cluster's prediction for five different classification methods. The results show that it is possible to classify initial data with the accuracy of 68.31% on average. Pre-classification of health records has helped to improve the results of regression almost twice on average, although the accuracy of prediction is needed to be further increased.