FPGA based High Speed ECG Signal Diagnosis for Artifacts

Abstract

The paper is dedicated to the design of multiband digital FIR filter for removal of various interference signals present in the ECG signal which does not allow the correct diagnosis of the patient. Filtering is finding its applications in wide domains and one of those domains is the biomedical science. ECG is the electrical response of the heart with respect to time. Heart diseases especially heart attacks are way common today and to detect this very precise detailing is required. There is no scope for allowed interference level or noise which corrupts the signal. Seven band filter is designed which will remove the frequencies that are prone to disturbances i.e. the various spectral components where particular type of noise from different sources dominates like noise from power line, respiration, muscle movement etc. The filter is developed and designed in Matlab along with Xilinx DSP tools synthesized with XST using Spartan 6 and Virtex 5 as target device. Filter is optimized using DA based architecture to increase the speed and maximum area utilization is obtained. The processing speed is efficiently optimized up to 28.36% with Virtex 5 as compared to Spartan 6 with the maximum area utilization in the presented paper.