Multichannel ECG Data Compression Based on Multiscale Principal Component Analysis

Abstract

In this paper, multiscale principal component analysis (MSPCA) is proposed for multichannel electrocardiogram (MECG) data compression. In wavelet domain, principal components analysis (PCA) of multiscale multivariate matrices of multichannel signals helps reduce dimension and remove redundant information present in signals. The selection of principal components (PCs) is based on average fractional energy contribution of eigenvalue in a data matrix. Multichannel compression is implemented using uniform quantizer and entropy coding of PCA coefficients. The compressed signal quality is evaluated quantitatively using percentage root mean square difference (PRD), and wavelet energy-based diagnostic distortion (WEDD) measures. Using dataset from CSE multilead measurement library, multichannel compression ratio of 5.98:1 is found with PRD value 2.09% and the lowest WEDD value of 4.19%. Based on, gold standard subjective quality measure, the lowest mean opinion score error value of 5.56% is found.

Quantification of Diagnostic Information from Electrocardiogram Signal: A Review

“…The DCT based MECG compression was proposed in [43]. In this work, first the PCA is applied to MECG signal and according to diagnostic importance the first six principal components (PCs) are retained [40] and others PCs are discarded. The DCT is applied to the PCA domain multivariate signal and due to energy compaction property the first few samples are retained.…”

“…The evaluation of sample entropy for a time series data is given below. [40], multivariate empirical mode decomposition [41], compressive sensing based MECG compression [42], DCT and karhunen loeve transform (KLT) based MECG compression [43] etc. are reported from literature.…”

Multiscale Principal Component Analysis to Predict Atrial Fibrillation Reversion to Sinus Rhythm

“…MSPCA has already been successfully implemented in biomedical studies, such as multichannel ECG data compression [6] or in milling tool wear monitoring method [7]. Moreover, MSPCA has been introduced for multivariate statistical process control [8] and it has been also used for real process fault detection and diagnosis [9,10].…”

Multilead ECG data compression using SVD in multiresolution domain

“…Recently, wavelet transform (WT) based compression methods have become popular [13][14][15][16][17][18]. The multiresolution property of the WT grossly segments the morphological features (P-wave, QRS-complex and T-wave) of the ECG signal into different subbands [15,16,19].…”