Adaptive algorithms for real-time filtering of the electrocardiogram

Published: 3 October 2019| Version 1 | DOI: 10.17632/d8byfd8d99.1
Contributor:
Nataliya Tulyakova

Description

The algorithms codes have been written in Pascal by Nataliya Tulyakova. To call the program, you need a 32-bit operating system or Free Pascal emulation. At the command prompt, enter the name of the program (for example, AHRT.exe) then the name of the test signal (clean.txt). Filter parameters are read from the “filters.txt” text file. The program asks for the signal length (you need to enter 798), then via space you need to set the variance of the additive, multiplicative Gaussian noise, the probability and amplitude of the outliers and the number of realizations for statistical averaging (for example: 0.00001 0 0 0 200), then select the menu item “Load from file”, by pressing the number 6. The results are written to the “MSE.res” and “SNR.res” files and together with the input and output signals in text format are placed in the RESULT folder. It have been created three separate programs: AHRT.pas - a single-pass algorithm that simulates ECG signal real-time processing; AHTPRT.pas - two-pass algorithm; AHMTPRT.pas - an algorithm with selective application of re-filtering. These algorithms are described in the article. The procedure that implements the algorithm is called adHampMod, adHampModTP. The text file “clean.txt” is the test ECG signal.

Files

Categories

Applied Computing, Computer Program

Licence