A Low-Complexity ECG Feature Extraction Algorithm for Mobile Healthcare Applications

Abstract

Abstract-This paper introduces a low-complexity algorithm for the extraction of the fiducial points from the Electrocardiogram (ECG). The application area we consider is that of remote cardiovascular monitoring, where continuous sensing and processing takes place in low-power, computationally constrained devices, thus the power consumption and complexity of the processing algorithms should remain at a minimum level. Under this context, we choose to employ the Discrete Wavelet Transform (DWT) with the Haar function being the mother wavelet, as our principal analysis method. From the modulus-maxima analysis on the DWT coefficients, an approximation of the ECG fiducial points is extracted. These initial findings are complimented with a refinement stage, based on the time-domain morphological properties of the ECG, which alleviates the decreased temporal resolution of the DWT. The resulting algorithm is a hybrid scheme of time and frequency domain signal processing. Feature extraction results from 27 ECG signals from QTDB, were tested against manual annotations and used to compare our approach against the state-of-the art ECG delineators. In addition, 450 signals from the 15-lead PTBDB are used to evaluate the obtained performance against the CSE tolerance limits. Our findings indicate that all but one CSE limits are satisfied. This level of performance combined with a complexity analysis, where the upper bound of the proposed algorithm, in terms of arithmetic operations, is calculated as 2.423N + 214 additions and 1.093N + 12 multiplications for N ≤ 861 or 2.553N + 102 additions and 1.093N + 10 multiplications for N > 861 (N being the number of input samples), reveals that the proposed method achieves an ideal trade-off between computational complexity and performance, a key requirement in remote CVD monitoring systems.

A Modular Low-Complexity ECG Delineation Algorithm for Real-Time Embedded Systems

“…As shown in next section, approximately 90% of the computation time is devoted to FIR (50%) and morphological (40%) filtering. These two parts are the most computationally intensive and their number of operations gives roughly the complexity of the algorithm [40], since the delineation of the ECG points mainly requires comparisons and means the 10% of the computation time.…”

Detection of Heart Beat Positions in ECG Recordings: A Lead-Dependent Algorithm

“…Several algorithms achieve beat detection rates greater than 99.5% (Afonso et al, 1999;Adnane et al, 2009). Recent technical advances have made possible to achieve low execution times (even real-time) or to eliminate the need for specific hardware (Quero et al, 2005;Mazomenos et al, 2013). Most of these proposals are based on single-channel detection, in some cases applied independently to two channels of the ECG.…”

Classification Methodology of CVD with Localized Features Analysis Using Phase Space Reconstruction Targeting Personalized Remote Health Monitoring

“…The State-of-the Art techniques may enable to extract the P and T waves correctly in following diseases (sinus arrhythmia, atrioventricular conduction variation and many more) [4], [10] where more likely P and T waves still be detected even though these are absent in the real life signals for the following diseases (Ventricular Arrhythmias, Atrial fibrillation and flutter) resulting in wrong diagnosis and therapy followed. Whereas the proposed methodology will detect the abnormal variation in the intervals and produce the corresponding contour and shows the disparity in the box count.…”

“…As shown in Fig.1, ECG samples are initially stored in the memory, for the analysis of stored healthy ECG and unhealthy ECG signals, the boundaries are extracted using our proposed Boundary Detection (BD) methodology [7] and the output of BD is applied to FE block [4] to get the following features QRS complex and PR interval of each and every individual beat. The proposed localized feature analysis methodology takes the features from FE block and plot the PSR of all the intervals and calculate the box count.…”