A book on Compression of Electrocardiogram signal through PCA and SVD

The ECG is an electronic method of recording the heart beat of the patient. The characteristic of heart beat is determined by PQRST curve. By attaching additional leads while recording ECG signals they require higher sampling rates and finer amplitude resolution, for Non-cardiac signals such as blood pressure and respiration it requires huge amount of storage space. For the fast transmission of ECG signals across wireless network, from ambulance to hospital, telephone networks, hospital networks is not possible without compressing them. Principal Component Analysis (PCA) and (Singular Value Decomposition) SVD method has been applied for ECG data compression. Two lead MIT-BIH Arrhythmia Database and a fifteen lead Physikalisch-Technische Bundesanstalt (PTB) Data Base is compressed and the quality of reconstructed signal is evaluated on the basis of Percentage Root Mean Square Difference (PRD) and Compression Ratio (CR).PCA of ECG signal is obtained using SVD method. ECG signal is reconstructed using different number of Eigen Vectors. CR of 4.26 is achieved with PRD of 8.1944 with this method for 2-lead ECG signal.