In this paper we present a method to predict Sudden Cardiac Arrest (SCA) with higher order spectral (HOS) and linear (Time) features extracted from heart rate variability (HRV) signal. Predicting the occurrence of SCA is important in order to avoid the probability of Sudden Cardiac Death (SCD). This work is a challenge to predict five minutes before SCA onset. The method consists of four steps: pre-processing, feature extraction, feature reduction, and classification. In the first step, the QRS complexes are detected from the electrocardiogram (ECG) signal and then the HRV signal is extracted. In second step, bispectrum features of HRV signal and time-domain features are obtained. Six features are extracted from bispectrum and two features from time-domain. In the next step, these features are reduced to one feature by the linear discriminant analysis (LDA) technique. Finally, KNN and support vector machine-based classifiers are used to classify the HRV signals. We used two database named, MIT/BIH Sudden Cardiac Death (SCD) Database and Physiobank Normal Sinus Rhythm (NSR). In this work we achieved prediction of SCD occurrence for six minutes before the SCA with the accuracy over 91%.

Abstract:In this paper we present a method to predict sudden cardiac death (SCD) based on the heart rate variability (HRV) signal and recurrence plots and Poincaré plot-extracted features. This work is a challenge since it is aimed to devise a method to predict SCD 5 min before its onset. The method consists of four steps: preprocessing, feature extraction, feature reduction, and classification. In the first step, the QRS complexes are detected from the electrocardiogram signal and then the HRV signal is extracted. In the second step, the recurrence plot of the HRV signal and Poincaré plot-extracted features are obtained. Four features from the recurrence plot and three features from the Poincaré plot are extracted. The features are recurrence rate, determinism, entropy and averaged diagonal line length, and SD1, SD2, and SD1/SD2. In the next step, these features are reduced to one feature by the linear discriminant analysis technique.Finally, K-nearest neighbor and support vector machine-based classifiers are used to classify the HRV signals. We use two databases, the MIT/BIH Sudden Cardiac Death Database and PhysioBank Normal Sinus Rhythm Database. We manage to predict SCD occurrence 5 min before the SCD with accuracy of over 92%.

This paper introduces the wireless sensor networks (WSNs) as an ad-hoc network and their structure in general and provides a small survey on sensors (nodes) as an embedded mechanism. The survey mainly focuses on the main challenges of these networks, such as the network layers of WSNs, data transfer over WSNs, the approaches for routing and packet management methods and localization of nodes.