Title :
Multichannel ECG Data Compression Based on Multiscale Principal Component Analysis
Author :
Sharma, L.N. ; Dandapat, S. ; Mahanta, Anil
Author_Institution :
Dept. of Electron. & Electr. Eng., Indian Inst. of Technol. Guwahati, Guwahati, India
fDate :
7/1/2012 12:00:00 AM
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.
Keywords :
data compression; eigenvalues and eigenfunctions; electrocardiography; matrix algebra; medical signal processing; principal component analysis; wavelet transforms; MECG data compression; MSPCA; PCA coefficients; PRD; WEDD; average fractional energy contribution; compressed signal quality; data matrix eigenvalue; dimensional reduction; entropy coding; multichannel electrocardiogram; multichannel signals; multiscale PCA; multiscale multivariate matrices; percentage root mean square difference; principal component analysis; redundant information removal; uniform quantizer; wavelet domain; wavelet energy based diagnostic distortion; Covariance matrix; Eigenvalues and eigenfunctions; Electrocardiography; Lead; Matrix decomposition; Principal component analysis; Wavelet transforms; Multiscale principal component analysis (MSPCA); multichannel electrocardiogram (MECG); percentage root mean square difference (PRD); wavelet; wavelet energy-based diagnostic distortion (WEDD); Algorithms; Databases, Factual; Electrocardiography; Humans; Multivariate Analysis; Principal Component Analysis; Signal Processing, Computer-Assisted;
Journal_Title :
Information Technology in Biomedicine, IEEE Transactions on
Conference_Location :
4/19/2012 12:00:00 AM
DOI :
10.1109/TITB.2012.2195322
