Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

A new approach to adaptive threshold based method for QRS detection with fuzzy clustering

Pander Tomasz

Biocybernetics and Biomedical Engineering

2022

Vol. 42, no. 1

404--425

The most crucial requirements for a QRS complex detection algorithm are accuracy, precision and repeatability. Most methods of detecting QRS complexes use the approach based on exceeding a certain amplitude threshold. However, the presence of noise in the electro-cardiographic signal can inhibit the accuracy and precision of detection especially for low amplitude QRS-complexes. The proposed algorithm uses a new approach for the amplitude threshold determination and in the decision stage. The fuzzy c-median clustering method is used to determine the amplitude threshold values for each sliding window across the composed detection function waveform. It allows us to adjust threshold value to noise variations in the ECG signal. When a specified amplitude threshold is exceeded by the detection function and finding the peaks in its waveform, potential QRS complexes can be identified. Then the identified peaks are evaluated on the basis of the speed of rising and falling slopes of detection function peak. It enables identification of only those peaks of the detection function whose location corresponds to QRS complexes. ECG recordings taken from the standard-available eight databases are used to evaluate the performance quality of the proposed method. The proposed QRS detector achieved sensitivity of 99.82%, positive predictivity of 99.88% over the validation MIT-BIH Arrhythmia Database. The overall sensitivity and positive predictivity are respectively 99.81% and 99.67%. Advantages of the proposed method are the robustness against noise, the accuracy and the simplicity of the algorithm that evaluates the candidate peaks of the detection function which indicate the QRS complexes.