Computational Efficiency and Accuracy for QRS Detection Algorithms on Clinical Long Term Multilead Monitoring

Abstract

A number of relevant clinical measurements are derived from QRS detection. As a consequence, the fast and accurate calculation becomes a key factor to meet this target, which is especially relevant for the huge amount of beats recorded in the increasingly IntroductionThe main purpose of cardiac monitoring is the detection of cardiac pathologies with transient manifestations or symptoms. The 24h holter is the device more frequently used for ECG monitoring in clinical environments. To carry out this test, the patient is subjected to monitoring for twenty-four consecutive hours, while performing their daily activities normally carrying the said device. Research in recent years has proven the long term monitoring (LTM) to be an important prognostic tool. It has been shown that diagnostic indicators are seen more clearly in longer periods. For this reason, recent works are focusing in incremental LTM protocols and systems: 7 days, 15 days, or 21 days. Inside this new scenario, one of the most important challenges is the extraction of different parameters that can be used to improve the accuracy of the diagnosis. The main challenges are in the massive amount of data that the clinician must analyse and validate for each patient, if the traditional protocols and signal processing techniques were to be applied. New protocols and new and more advanced signal processing techniques are required to serve this new framework, and that is where relevant literature is currently focused [1][2][3][4][5][6].One of the key parameters to examine the heart behaviour is the heart rate (HR) variation observation, whose analysis supports much heart disease analysis, namely, arrhythmias, transient ischemic episodes, silent myocardial ischemia, and arrhythmic risk assessment of patients, among others [7]. In all cases, and in order to create the heart rate variability signal, the detection of the QRS complex, the R-peak and the time intervals among two consecutive beats, are required.The objective of this paper is to develop a QRS detection algorithm capable of running with the highest possible computational efficiency and accuracy over LTM records. This goal has been accomplished by using an algorithm that takes into account both the inter-lead and the intra-lead information by adjusting the threshold level and optimizing the computational requirements. Accuracy and computational efficiency of algorithms were tested over, a trusted and solid set of 48h holter

QRS Fragmentation Index as a New Discriminator for Early Diagnosis of Heart Diseases

“…After this, a characteristic signal is computed, which contains enhanced QRS complexes and where the R-waves can be easily thresholded. Finally, the detection of QRS complexes is performed by using the characteristic signal, the threshold, and the medical knowledge about the heart behavior and for this purpose we used previous published work from our team [6].…”