Cardiac events detection using curvelet transform

Abstract

Cardiac event detection is one of the essential steps in cardiac signal processing, analysis and disease diagnosis. Complete morphology of cardiac waves (P-QRS-T) extracted from the location of R-peak is helpful for feature extraction of many applications related to cardiac diseases classification. Therefore cardiac event detection is a prerequisite for reliable cardiac disease diagnosis, and hence it should be robust and timeefficient so that it can be used for real-time signal processing. This work proposes a novel method for R-peak detection using curvelet transform (CT). It demonstrates the use of curvelet energy with an adaptive threshold to estimate the boundaries around R-peak. The exact R-peak locations are then detected from the input signal with the predefined estimated boundaries. The proposed method is evaluated and analysed with all 48 records from the MIT-BIH arrhythmia database. The experimental analysis result yields an average sensitivity of 99.62%, average positive productivity of 99.74% and average detection error rate of 0.6%. The results obtained have higher than or comparable indices to those in literature. Thus, the proposed system yields high accuracy, low complexity and high processing speed.

Prediction and Analysis of Heart Disease Using SVM Algorithm

“…Different technologies and the rich survey is available for the heart disease prediction model. Alka S. Barhatte et al [1] propose ECG signal analysis and classification method using wavelet energy histogram method and support vector machine (SVM). The classification of cardiac heart disease in the ECG signal consists of three stages including ECG signal preprocessing, feature extraction and heartbeats classification.…”

“…The paper gives the use of methods like fourth order wavelet decomposition, wavelet decomposition used for time-frequency representation and feature extraction. For classification, support vector machine is used for detection kinds of ECG signals validated by the data MIT BIH [14] [15] arrhythmia database. This method uses fourth-order wavelet decomposition, wavelet decomposition used for time-frequency representation and feature extraction.…”