Fast Walsh transform data-compression algorithm: E.c.g. applications

Kuklinski, Walter S.

doi:10.1007/bf02442635

Cited by 66 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus original samples of ECG are subjected to a transformation and the compression is performed in the entirely new domain like Fourier transform (FT), DCT and wavelet etc [8,12,16,28,31]. These techniques pose higher CR than direct techniques and are insensitive to noise present in ECG signals.…”

Section: Transform-domain Techniquesmentioning

confidence: 99%

A Review of ECG Data Compression Techniques

Singh¹,

Kaur²,

Singh³

2015

IJCA

View full text Add to dashboard Cite

Electrocardiogram (ECG) data compression reduced the storage requirements to develop a more efficient telecardiology system for cardiac analysis and diagnosis. The ECG compression without loss of diagnostic information is based on the fact that consecutive samples of the digitized ECG carry redundant information that can be removed with very less computing effort. This paper focuses on providing a comparison of the major techniques (direct, transform, parameter extraction and 2D approaches) of ECG data compression which are intended to attain a lossless compressed data with relatively high compression ratio (CR) and low percent root mean square difference (PRD).The paper concludes with the presentation of a framework for evaluation and comparison of ECG compression schemes.

show abstract

Section: Transform-domain Techniquesmentioning

confidence: 99%

A Review of ECG Data Compression Techniques

Singh¹,

Kaur²,

Singh³

2015

IJCA

View full text Add to dashboard Cite

show abstract

“…Different types of transformation based compression techniques are-Karhunen-Loeve transform (KLT) [4], Fourier transform (FT), Discrete cosine transform (DCT), Walsh transform (WT) [5], and Haar transform (HT). Among these, wavelet transform based compression technique is a popular time-frequency analysis tool.…”

Section: Introductionmentioning

confidence: 99%

A DPCM based Electrocardiogram coder with thresholding for real time telemonitoring applications

Alam¹,

Gupta

2014

2014 International Conference on Communication and Signal Processing

View full text Add to dashboard Cite

This paper illustrates a DPCM based approach for real-time compression of ECG data for real time telemonitoring application. For real time implementation a 'frame' is considered with one original sample followed by 64 first difference elements. The coder compresses the non-QRS regions of an ECG data stream through stages of first difference, joint sign and magnitude coding, and run length encoding. A hard thresholding at the equipotential regions have been applied to enhance the RLE efficiency. For testing 10 second ECG data from Physionet has been used with 10-bit quantization level. The CR, PRD and PRDN achieved with PTB Database (ptbdb) are 6.42, 9.77 and 9.77 respectively. With MIT-BIH arrhythmia data (mitdb), these values are 5.92, 8.19 and 8.19 respectively. With MIT-BIH ECG Compression test data (cd b), these values are 4.25, 5.37 and 6.65 respectively. The frame wise compression rises to a value of 12-14 in flat (TP) segments and low 1-2 in QRS regions. The computational simplicity of the algorithm provides an opportunity to implement the coder using a low cost microcontroller.D

show abstract

“…Bien qu'il existe un algorithme rapide pour la transformation de Fourier (FFT), la charge de calcul reste élevée puisque le signal est décomposé en parties réelle et imaginaire. Les transformations quasi-spectrales à base des fonctions de Wash (WaT) ou de Haar (HT) ont été proposées pour la compression de l'ECG dans la référence [16]. La charge de calcul est faible dans ces cas, mais le signal reconstruit contient trop de distorsions, à cause de la discontinuité de WaT et de HT.…”

Section: Introductionunclassified