A Generic and Robust System for Automated Patient-Specific Classification of ECG Signals

By Turker Ince, Serkan Kiranyaz, Moncef Gabbouj. IEEE Transactions on Biomedical Engineering, 2009

485 citations recorded by the source corpus.

This paper presents a generic and patient-specific classification system designed for robust and accurate detection of ECG heartbeat patterns. The proposed feature extraction process utilizes morphological wavelet transform features, which are projected onto a lower dimensional feature space using principal component analysis, and temporal features from the ECG data. For the pattern recognition unit, feedforward and fully connected artificial neural networks, which are optimally designed for each patient by the proposed multidimensional particle swarm optimization technique, are employed. By using relatively small common and patient-specific training data, the proposed classification system can adapt to significant interpatient variations in ECG patterns by training the optimal network structure, and thus, achieves higher accuracy over larger datasets. The classification experiments over a benchmark database demonstrate that the proposed system achieves such average accuracies and sensitivities better than most of the current state-of-the-art algorithms for detection of ventricular ectopic beats (VEBs) and supra-VEBs (SVEBs). Over the entire database, the average accuracy-sensitivity performances of the proposed system for VEB and SVEB detections are 98.3%-84.6% and 97.4%-63.5%, respectively. Finally, due to its parameter-invariant nature, the proposed system is highly generic, and thus, applicable to any ECG dataset.

Subfields: Cardiology and Cardiovascular Medicine, Cognitive Neuroscience, Signal Processing

View Paper: https://www.semanticscholar.org/paper/4d35546b45d83f8f1fdc676fac60c24758e56ab6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cardiology and Cardiovascular Medicine

In Cognitive Neuroscience

In Signal Processing

Citations in the corpus

Heartbeat Classification Using Feature Selection Driven by Database Generalization Criteria (2011, methodology): “…in the reviewed papers, from simple linear discriminant functions based on the Gaussian assumption of the data [4], [8] to more elaborated ones, as arti-ﬁcial neural networks (ANN’s), self-organizing maps (SOM), and learning vector quantization (LVQ) among others [2], [3], [5], [6], [9]–[11].”
Heartbeat Classification Using Morphological and Dynamic Features of ECG Signals (2012, methodology): “For most of the previous works [7], [13], [14], [16], two instantaneous RR features, also regarded as heartbeat interval features, were utilized namely, previous RR and post-RR features.”
Heartbeat classification using disease-specific feature selection (2014, methodology): “Recently, the manual expert-assisted approach [12,13] and unsupervised machine learning methods [15] have been employed for research in this direction.”
1D Convolutional Neural Networks and Applications: A Survey (2019, methodology): “In the literature, there have been some attempts to change MLPs by modifying the neuron model and/or the conventional Back Propagation (BP) algorithm [9–11], or the MLP configuration [12–14] or even the way to update the network parameters (weights and biases) [15].”
Arrhythmia recognition and classification using combined linear and nonlinear features of ECG signals (2016, methodology): “(Hu et al.,1997) Time domain features Mixture of experts 2 94.00% (Ince et al.,2009) DWT+PCA MDPSO 5 95.58% (Martis et al., 2012) PCA SVM-RBF 5 98.11% (Martis et al., 2013c) DWT+PCA SVM-RBF 5 96.92% DWT+PCA NN 5 98.78%”
Deep learning approach for active classification of electrocardiogram signals (2016, background): “Basically, these approaches start by training a global classifier on a large dataset and another local-classifier on the first few minutes from the test record labeled by an expert [5,15,41,32,34].”