Cardiac Arrhythmia Diagnosis with an Intelligent Algorithm using Chaos Features of Electrocardiogram Signal and Compound Classifier

Q: What is the clinical evidence from this study?

Population: Cardiac Arrhythmias (Atrial Fibrillation, Ventricular Tachycardia, Paroxysmal Supraventricular Tachycardia) (n=72). Intervention: Intelligent algorithm using chaos features (Lyapunov exponent and fractal dimension) and a compound classifier (MLP, SVM, KNN) vs. Individual classifiers (MLP, SVM, KNN) and time-frequency/time-domain features. Primary outcome: Classification accuracy.

Q: What does this research mean for the field?

Combining chaos features of ECG signals with a compound classifier achieves cardiac arrhythmia classification with 99.1% accuracy, 99.6% sensitivity, and 99.3% specificity. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

E. Zarei; N Barimani; G. Nazari Golpayegani

doi:10.22044/jadm.2022.12165.2370

November 1, 2022Open Access

Cardiac Arrhythmia Diagnosis with an Intelligent Algorithm using Chaos Features of Electrocardiogram Signal and Compound Classifier

View Full Paper

Ask AI

Bookmark

Share

Key result

ECG chaos features and a compound classifier classify cardiac arrhythmias with ~99% accuracy.

n=72

Why the study?

Intelligent algorithms can reduce ECG processing time and prevent diagnostic mistakes caused by physician fatigue, motivating an intelligent algorithm to separate three cardiac arrhythmias.

Key Points

To develop an intelligent algorithm for classifying three types of cardiac arrhythmias alongside normal heart rhythms using chaos features extracted from electrocardiogram signals and a compound machine learning classifier.
Gathered normal electrocardiogram (ECG) signals and records of atrial fibrillation, ventricular tachycardia, and post supraventricular tachycardia from the MIT-BIH arrhythmia database, totaling 1382 segments.
Extracted non-linear dynamic chaos metrics including Lyapunov exponent values and fractal dimension algorithms (Higuchi and Petrosian), selecting five features with p < 0.05 via statistical t-test.
Trained and evaluated a compound classifier combining multilayer perceptron, support vector machine, and k-nearest neighbor networks over 20 iterations, benchmarking against individual classifiers and time-frequency methods.
The compound classifier using ECG chaos features attained an accuracy of 99.1% ± 0.2, sensitivity of 99.6% ± 0.1, and specificity of 99.3% ± 0.1.
Statistical t-tests confirmed p < 0.05 for maximum, minimum, average, and standard deviation of Lyapunov exponents and the Higuchi coefficient, whereas the Petrosian coefficient had p > 0.05 and was discarded.
Combining the three classifiers yielded superior accuracy, sensitivity, and specificity compared to individual evaluations using multi-layer perceptron, support vector machine, or k-nearest neighbor alone.

Structured PICO

P

Population

72 ECG signals (18 each of normal, atrial fibrillation, ventricular tachycardia, and paroxysmal supraventricular tachycardia) from the MIT-BIH database, segmented into 1728 beats for processing.

I

Intervention

Intelligent algorithm using chaos features (minimum, maximum, standard deviation, and average of Lyapunov exponent, and fractal dimension via Higuchi and Petrosian algorithms) combined with a compound classifier (multi-layer perceptron neural network, support vector machine, and K-Nearest Neighbor combined via Behavior Knowledge Space).

C

Comparator

Classifying methods based on time-domain and time-frequency domain features, as well as the individual classifiers (MLP, SVM, KNN) used separately.

O

Outcome

Classification accuracy, sensitivity, and specificity for differentiating the three cardiac arrhythmias and normal signals.

An intelligent algorithm combining chaos features of ECG signals with a compound classifier achieved over 99% accuracy, sensitivity, and specificity in distinguishing normal rhythms from atrial fibrillation, ventricular tachycardia, and paroxysmal supraventricular tachycardia.

Limitations

Small number of original ECG signals (72 total) used for segmentation
Retrospective analysis on a public database
The Petrosian algorithm did not show an acceptable level of significance for feature extraction

Implication

High reported accuracy supports chaos-feature ECG classifier development; leaves open clinical adoption pending prospective validation.

Abstract

Cardiac Arrhythmias are known as one of the most dangerous cardiac diseases. Applying intelligent algorithms in this area, leads into the reduction of the ECG signal processing time by the physician as well as reducing the probable mistakes caused by fatigue of the specialist. The purpose of this study is to introduce an intelligent algorithm for the separation of three cardiac arrhythmias by using chaos features of ECG signal and combining three types of the most common classifiers in these signal’s processing area. First, ECG signals related to three cardiac arrhythmias of Atrial Fibrillation, Ventricular Tachycardia and Post Supra Ventricular Tachycardia along with the normal cardiac signal from the arrhythmia database of MIT-BIH were gathered. Then, chaos features describing non-linear dynamic of ECG signal were extracted by calculating the Lyapunov exponent values and signal’s fractal dimension. finally, the compound classifier was used by combining of multilayer perceptron neural network, support vector machine network and K-Nearest Neighbor. Obtained results were compared to the classifying method based on features of time-domain and time-frequency domain, as a proof for the efficacy of the chaos features of the ECG signal. Likewise, to evaluate the efficacy of the compound classifier, each network was used both as separately and also as combined and the results were compared. The obtained results showed that Using the chaos features of ECG signal and the compound classifier, can classify cardiac arrhythmias with 99.1% ± 0.2 accuracy and 99.6% ± 0.1 sensitivity and specificity rate of 99.3 % ± 0.1

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Zarei et al. (2022) studied Cardiac Arrhythmias (Atrial Fibrillation, Ventricular Tachycardia, Paroxysmal Supraventricular Tachycardia) (n=72). Intelligent algorithm using chaos features (Lyapunov exponent and fractal dimension) and a compound classifier (MLP, SVM, KNN) vs. Individual classifiers (MLP, SVM, KNN) and time-frequency/time-domain features was evaluated on Classification accuracy. Using chaos features of the ECG signal and a compound classifier classified cardiac arrhythmias with 99.1% accuracy, 99.6% sensitivity, and 99.3% specificity.

synapsesocial.com/papers/6a2e4779b1cc60ccdea8bfc9 https://doi.org/10.22044/jadm.2022.12165.2370

View Full Paper

Ask AI

Bookmark

Share

Cardiac Arrhythmia Diagnosis with an Intelligent Algorithm using Chaos Features of Electrocardiogram Signal and Compound Classifier

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Cardiac Arrhythmia Diagnosis with an Intelligent Algorithm using Chaos Features of Electrocardiogram Signal and Compound Classifier

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading