Electrocardiogram Signal Compression Based on 2D-Transforms: A Research Overview

Off-Person ECG Biometrics Using Spatial Representations and Convolutional Neural Networks

“…Transforming ECG time series into images has been previously considered for compression purposes [30], building upon clever using of both intra-segment and intersegment correlation of data divided into individual heartbeats. Encoding (ECG) time series as images brings a series of advantages including the extent of the range of texture features to be further classified (repeated convolutions extract relevant information using the local spatial coherence of such features), the identification of regions of interest in the resulting bidimensional representation (allowing interpretation if the encoding enables time series reconstruction), and proper use of the transfer learning principle (by considering already available high quality models to be tuned according to the specific ECG-based dataset).…”

EPOWT: A denoising technique of the electrocardiography signal transmission via 5G wireless communications

“…Mostly, measurement of signal compression in term of compression ratio (CR), nevertheless there are much post compression problems which have been required to be detected and measured to calculate the reconstructed signal quality and towards checking the diagnostic information. 37 Another one is Percentage Root Mean Square Difference (PRD) that checks the distortion among reconstructed/filtered and original ECG signal. These two parameters are mostly applied to calculate the performance of ECG compression algorithms, and are termed as: CR: CR = Amount of encoded transform coefficients Total amount of transform coefficients .…”

“…Mostly, measurement of signal compression in term of compression ratio (CR), nevertheless there are much post compression problems which have been required to be detected and measured to calculate the reconstructed signal quality and towards checking the diagnostic information . Another one is Percentage Root Mean Square Difference (PRD) that checks the distortion among reconstructed/filtered and original ECG signal.…”

Efficient compression technique based on temporal modelling of ECG signal using principle component analysis

“…In this context, telecardiography is one type of telemonitoring/telemedicine health assistive tool especially for cardiac health diagnosis. Basically, telecardiography is a key solution for regular/continuous cardiac health monitoring from remote station as well as ambulatory moving station [1], which has several essential parts based on information and communication technology that makes possible connection between the patient and health centres. These are signal acquisition system, processing unit and communication system as shown in Fig.…”

“…In telecardiography, a huge amount of data is generated during ambulatory monitoring that is very expensive with such type of communication system. Therefore, less expensive/lower bandwidth communication channels or data shrinkage/compression maybe the solution for cost effective as well as space saving systems [1–6].…”