Performance Assessment of a Clinical Decision Support System for Analysis of Heart Failure

Abstract

In this paper we compare five machine learning techniques in dealing with typical Heart Failure (HF) data. We developed a Clinical Decision Support System (CDSS) for the analysis of Heart Failure patient that provides various outputs such as an HF severity evaluation, an HF type prediction, as well as a management interface that compares the various patient’s follow-ups. To realize these smart functions we used machine learning techniques and in this paper we compare the performance of a neural network, a support vector machine, a system with fuzzy rules genetically produced, a Classification and regression tree and its direct evolution which is the Random Forest, in analyzing our database. Best performances (intended as accuracy and less critical errors committed) in both HF severity evaluation and HF type prediction functions are obtained by using the Random Forest algorithm.