Session P74.5

Disorder Classification in the Regulatory Mechanism of the Cardiovascular System

A Jalali*, H SadAbadi, M Ghasemi, A Ghaffri, P Bahaman Bijari

K.N. Toosi University of Technology
Tehran, Iran

An approach to classify disorders in autonomic control of the cardiovascular system is proposed in this paper. The Autonomic Nervous System (ANS) provides second-to-second adjustment of blood pressure and heart rate, allowing humans great flexibility in posture and environment. Since many clinicians are unfamiliar with disorders in the autonomic nervous system, it is important to automatically detect these abnormalities. The target of this study is to highlight the main features of malfunctions in the cardiovascular system due to autonomic disorder. Collecting the data from the PhysioNet archive, we divided patients into two groups of normal and abnormal, based on their having autonomic disorder in their cardiovascular system or not. We then extracted features of each group to detect differences between them. Collected data of continuous arterial blood pressure (ABP) and ECG signals of patients in two groups are preprocessed to remove artifacts and noises. Systolic blood pressure (SBP) and heart rate (HR) are evaluated for each beat. We then plot the diagram of SBP against HR for all patients in a single figure. Fuzzy c-means clustering (FCM) method is also applied to cluster data into two groups. The plot and the clustering results indicate distinguished differences between the two groups. A neural network is then implemented to classify and to distinguish the two groups. A multi-layer perceptron (MLP) network is trained with data of a normal patient and is tested with data of other normal and abnormal patients. It is shown that the root mean square error (RMSE) for all abnormal patients is significantly larger then for normal patients. It is concluded that we can use this network to separate normal patients from abnormal ones. Results show that selected features can clearly detect disorders in the Autonomic Nervous System.

(Abstract Control Number: 194)