Development of an automated diagnosis of complications after operation on mitral valve

Abstract

There is always a risk of postoperative complications after heart surgery. Doctors need carry out comprehensive analysis of many data in order to detect and to foresee this aftereffect. This analysis takes a lot of time and effort. Automated identification of postoperative complications in the patient's mitral valve was developed for avoiding this problem. Statistical models of post-operative complications of the patient were engineered based on discriminant analysis and logistic regression models with use cluster analysis. The results of calculations at this stage of the study showed a high diagnostic performance of logistic regression with clustering.