QRS detection algorithm based on Hilbert transform

Elektrokardiogram (EKG) je grafički prikaz električne aktivnosti srca koji se generira tijekom polarizacije i depolarizacije atrija i ventrikula. Karakteristični dijelovi EKG signala su P val, QRS kompleks i T val. Jedna od obrada EKG signala je detekcija QRS kompleksa, posebno R zupca. U radu je prikazan novi algoritam za detekciju QRS kompleksa koji koristi svojstva Hilbertove transformacije. Ovaj algoritam smanjuje utjecaj velikih amplituda P i T valova na točnost detekcije. Također, prednost ovog algoritma je dobra detekcija QRS kompleksa u uvjetima velikog šuma. Algoritam je razvijen pomoću softverskog paketa Matlab, a dobiveni rezultati su uspoređeni s Pan-Tompkins algoritmom za detekciju QRS kompleksa. Oba algoritma su testirana pomoću MIT-BIH standardne baze EKG signala snimljenih na pacijentima s aritmijama.The electrocardiogram (ECG) is a grapich record of the electrical activity of the heart that is generated during polarization and depolarization of the atria and ventricles. Typical parts of the ECG signal are the P wave, QRS complex and T wave. On of the ECG signal processing is QRS detection, especially the R wave. In this paper is proposed a new algorithm for QRS detection who using the properties of the Hilbert transform. This algorithm minimizes the influence of the large amplitude of P and T waves on the detection accuary. Another adventage of this algorithm is a good detection of the QRS complex in conditions of the large noise. The algorithm was developed using the software package MATLAB, and the results were compared with the Pan-Tompkins algorithm for the detection of QRS complex. Both algorithms were tested using the MIT-BIH standard database of ECG signals recorded on patients with arrhythmia