Body Sensor Network Based ECG Segmentation and Analysis

In this paper, a body sensor network based ECG signal segmentation approach is presented. Hidden Markov modeling (HMM) technique is employed. Since people's heart rates vary a lot, the corresponding characteristic waveform intervals and durations change with time as well. For patients with cardiac diseases, such as arrhythmia, the heart beat interval may even change abruptly and irregularly. Because traditional HMM parameter adaptation is conservative and slow to respond to beat interval changes, inadequate and slow parameter adaptation is largely responsible for the low positive predictivity rate (+P). To solve the problem, we introduce an active HMM parameter adaptation and ECG segmentation algorithm, which includes three parts: the pre-segmentation and classification, the HMM model training, and the detailed segmentation. Body sensor networks are used to collect and pre-segment the raw ECG data by performing QRS detection. Then the R-R interval information that directly reflects the beat variation is extracted and used to classify the raw ECG data into several groups. One specific HMM is trained for each of the groups. Instead of one single generic HMM, multiple individualized HMMs are set up and each HMM is only responsible for extracting the characteristic waveforms of the ECG signals with similar temporal features in the detailed segmentation, so that the temporal parameter adaptation can be naturally achieved.

I. INTRODUCTION and Related Works

Traditionally, exercise test and ambulatory ECG are two of the most widely accepted noninvasive procedures for identifying patients with probable heart disease [2]. Measurements of the width or duration of characteristic waveforms and analysis of their morphology variations in the ECG are used to define ectopic electrical activity in the heart, to detect myocardial damage, and to classify patients at risk of cardiac diseases [8]. However, these tests and ECG processing are usually performed in a lab environment and are under the supervision of specially trained physicians. This may cause so-called White Coat Syndrome, which may affect the accuracy of the test result due to mental stress. Moreover, since the test can only be performed occasionally due to the cost and schedule constraint, some important symptoms can be missed which may lead to delayed diagnosis and even misdiagnosis.