Evaluating a decision making system for cardiovascular dysautonomias diagnosis

Autonomic nervous system (ANS) is the part of the nervous system that is involved in homeostasis of the whole body functions. A malfunction in this system can lead to a cardiovascular dysautonomias. Hence, a set of dynamic tests are adopted in ANS units to diagnose and treat patients with cardiovascular dysautonomias. The purpose of this study is to develop and evaluate a decision tree based cardiovascular dysautonomias prediction system on a dataset collected from the ANS unit of the Moroccan university hospital Avicenne. We collected a dataset of 263 records from the ANS unit of the Avicenne hospital. This dataset was split into three subsets: training set (123 records), test set (55 records) and validation set (85 records). C4.5 decision tree algorithm was used in this study to develop the prediction system. Moreover, Java Enterprise Edition platform was used to implement a prototype of the developed system which was deployed in the Avicenne ANS unit so as to be clinically validated. The performance of the decision tree-based prediction system was evaluated by means of the error rate criterion. The error rates were measured for each classifier and have achieved an average value of 1.46, 2.24 and 0.89 % in training, test, and validation sets respectively. The results obtained were encouraging but further replicated studies are still needed to be performed in order to confirm the findings of this study.