What is the clinical evidence from this study?

Study design: Other. Population: Arrhythmia (n=47). Intervention: ACoFCN and ACoBi-LSTM deep learning models vs. Existing classification methods. Primary outcome: Overall classification accuracy.

May 29, 2025Discover Artificial Intelligence15 citationsOpen Access

FADLEC: feature extraction and arrhythmia classification using deep learning from electrocardiograph signals

SLSumita Lamba SKSatender Kumar MDManoj Diwakar

Key Result

The proposed ACoFCN and ACoBi-LSTM deep learning models achieved an overall accuracy of 99.1% and 98.9%, respectively, for 5-class arrhythmia classification using the MIT-BIH dataset.

Structured PICO

Population

ECG recordings from the MIT-BIH Arrhythmia Dataset, balanced to 55,000 samples per class (total 275,000 samples) for 5 arrhythmia classes (Normal, Supraventricular, Ventricular, Fusion, Unknown).

Intervention

FADLEC model using Ant Colony Optimization (ACO) to fine-tune Bidirectional Long Short-Term Memory (Bi-LSTM) and Fully Convolutional Network (FCN) architectures, with Stationary Wavelet-Hilbert transform (SW-HT) for feature extraction and SMOTE for data balancing.

Outcome

Overall accuracy of 5-class arrhythmia classification

An optimized deep learning model using ACO to fine-tune Bi-LSTM and FCN architectures achieved over 98.9% accuracy in classifying five types of arrhythmias from ECG signals.

Limitations

Performance might diverge when applied to real-time ECG signal processing or other, more diverse datasets
Computational intensity of Ant Colony Optimization (ACO)

Abstract

Abstract Classifying arrhythmia is an essential step in the diagnosis and monitoring of cardiovascular illness. Deep learning (DL) models are trained on the electro-cardiogram recordings found in the ECG signal dataset to accurately classify arrhythmia into five groups: Normal (N), Fusion (F), Supraventricular (S), Ventricular (V), and Unknown (Q). In the proposed work, Ant Colony Optimization (ACO) to fine-tune the hyperparameter of two potent Deep Learning (DL) architectures, Bidirectional Long Short-Term Memory (Bi-LSTM) and Fully Convolutional Network (FCN) is utilised. Initially, ECG signals are pre-processed, where Multi-Resolution Wavelet-based techniques are applied for noise removal. Afterwards, the Stationary Wavelet-Hilbert transform (SW-HT) is applied for feature extraction. Next, training, validation, and testing sets are created from the extracted feature set. After performing data balancing using the SMOTE (Synthetic Minority Over-sampling Technique) algorithm, classification using optimized deep learning models is performed. With an overall accuracy of 98.9% (ACoBi-LSTM) and 99.1% (ACoFCN) on the 5-Class (N, S, V, F, Q) arrhythmia classification in the MIT-BIH dataset, the proposed model’s performance is compared and analyzed against the existing methods.

问 AI

Bookmark

View Full Paper