Noise-tolerant modular neural network system for classifying ECG signal

Abstract

Millions of electrocardiograms (ECG) are interpreted every year, requiring specialized training for accurate interpretation. Because automated and accurate classification ECG signals will improve early diagnosis of heart condition, several neural network (NN) approaches have been proposed for classifying ECG signals. Current strategies for a critical step, the preprocessing for noise removal, are still unsatisfactory. We propose a modular NN approach based on artificial noise injection, to improve the generalization capability of the resulting model. The NN classifier initially performed a fairly accurate recognition of four types of cardiac anomalies in simulated ECG signals with minor, moderate, severe, and extreme noise, with an average accuracy of 99.2%, 95.1%, 91.4%, and 85.2% respectively. Ultimately we discriminated normal and abnormal heartbeat patterns for single lead of raw ECG signals, obtained 95.7% of overall accuracy and 99.5% of Precision. Therefore, the propose approach is a useful tool for the detection and diagnosis of cardiac abnormalities.Povzetek: V članku je opisana metoda modularnih nevronskim mrež za prepoznavanje šumnih ECG signalov. Materials and methodsThe NN classification comprises five stages: (i) simulation of ECG signal, (ii) extraction of features that indicate cardiac abnormalities, (iii) computer generation of normal and abnormal heartbeat patterns, (iv) artificial noise injection, and (v) cardiac rhythm classification on simulated and real ECG signals.

Classification of Cardiac Arrhythmia using improved Feature Selection methods and Ensemble Classifiers

“…Generated feed-forward neural network is also used for doing the similar work [6]. Along with this, some other approaches like Bayesian artificial neural networks [7], modular neural networks [8] and perceptron having multiple layers with one-against-all method [9] are also used in a similar fashion for categorizing cardiac arrhythmia into sixteen different classes. Apart from these, a contemporary approach for categorizing cardiac arrhythmia is lodged in [10] whose operating principle revolves around picking up the most appropriate characteristics from the UCI electrocardiogram data by using correlation-based feature selection technique.…”