Electrocardiogram Based Heartbeat Detection Using Deep Learning

Q: What is the clinical evidence from this study?

Population: Arrhythmia (n=47). Intervention: Convolutional neural network with multimodal image fusion vs. LSTM, RNN, and existing methods. Primary outcome: Heartbeat detection accuracy.

Q: What does this research mean for the field?

An automated heartbeat detection framework combining convolutional neural networks with multimodal image fusion from transformed ECG signals achieves high arrhythmia recognition performance, reaching 99.63% accuracy and a 99.64% F1-score. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Minh Tuan Nguyen; Anh Nguyen

doi:10.21553/rev-jec.401

July 2, 2025REV Journal on Electronics and CommunicationsOpen Access

Electrocardiogram Based Heartbeat Detection Using Deep Learning

View Full Paper

Ask AI

Bookmark

Share

Key result

Multimodal image fusion CNN detects ECG heartbeats with ~100% accuracy.

n=47

Why the study?

Early and accurate arrhythmia diagnosis requires improved automated heartbeat detection methods using ECG signals.

Does a CNN model using three-channel transformed images improve heartbeat detection accuracy in ECG signals?

Key Points

To develop an automated heartbeat classification system that transforms raw electrocardiogram signals into multimodal fused images for robust arrhythmia diagnosis using deep learning.
Cleaned electrocardiogram signals using [0.5-40] Hz band-pass filtering and wavelet transform, identified R-peaks with X-wave QRS detection, and extracted heartbeats via a 360-sample sliding window.
Converted segmented heartbeats into three-channel images by fusing Gramian angular field, Markov transition field, and recurrence plot representations, while applying synthetic minority oversampling.
Optimized convolutional neural network hyperparameters through grid search and evaluated diagnostic performance using a 5-fold cross-validation procedure.
The proposed deep learning algorithm achieved an average classification accuracy of 99.63% across 5-fold cross-validation.
Arrhythmia detection yielded an average precision of 99.41%, recall of 99.52%, and an F1-score of 99.64%.

Structured PICO

Does a CNN model using three-channel transformed images improve heartbeat detection accuracy in ECG signals?

P

Population

47 subjects with 48 ECG recordings from the MIT-BIH Arrhythmia dataset used to train and test a deep learning model for heartbeat detection.

I

Intervention

Automated heartbeat detection algorithm using a convolutional neural network (CNN) and three-channel images constructed via multimodal image fusion (Gramian angular field, Markov transition field, and Recurrence plot) from preprocessed ECG signals.

C

Comparator

Other deep learning models (LSTM, RNN) and existing computational methods.

O

Outcome

Average accuracy, precision, recall, and F1-score of heartbeat classification.surrogate

A novel deep learning algorithm using a CNN and multimodal image fusion of ECG signals achieved over 99% accuracy in heartbeat classification, offering a potential tool for automated arrhythmia detection.

Limitations

Implementation of transformation technique to convert the ECG signals into multiple image formats increases the computational complexity.
The analysis is limited to a single dataset which raises concerns about the generalization of the results to other clinical scenarios.
Implementation of transformation technique to convert ECG signals into multiple image formats increases computational complexity.
Analysis is limited to a single dataset, raising concerns about generalization to other clinical scenarios.

Implication

Promising benchmark accuracy merits prospective clinical validation; leaves open utility for automated arrhythmia detection in practice.

Abstract

Cardiovascular diseases remain a leading cause of death worldwide, which results in important requirements for early and accurate arrhythmia diagnosis. This work proposes a novel design of automated heartbeat detection, which consists of a convolutional neural network and three-channel images using the electrocardiogram (ECG) signals. A combination of various preprocessing is applied for the elimination of interferences of the ECG signals such as band-pass filtering and wavelet transform for R-peak identification using a sliding window. Multimodal image fusion method is used to construct three-channel images from different grayscale images, which are transformed from the heartbeats by three transformation techniques namely Gramian angular field, Markov transition field, and Recurrence plot. Grid-search based optimization method in combination with 5-fold cross validation procedure are implemented for selection of the optimal hyper-parameters of the CNN models using the input three-channel images. The proposed algorithm including CNN models and MIF images is estimated the detection performance using 5-fold cross validation, which produces average accuracy of 99.63%, precision of 99.41%, recall of 99.52%, and F1-score of 99.64%. The relatively high performance of the proposed algorithm confirms the effectiveness for the arrhythmia recognition on the ECG signals.

Discussion

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

Cite This Study

Nguyen et al. (2025) studied Arrhythmia (n=47). Convolutional neural network with multimodal image fusion vs. LSTM, RNN, and existing methods was evaluated on Heartbeat detection accuracy. The proposed convolutional neural network with multimodal image fusion achieved an average accuracy of 99.62% for heartbeat detection on the MIT-BIH dataset.

synapsesocial.com/papers/6ac19d70aeea5716e8c6c60b https://doi.org/10.21553/rev-jec.401

View Full Paper

Ask AI

Bookmark

Share

Electrocardiogram Based Heartbeat Detection Using Deep Learning

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Electrocardiogram Based Heartbeat Detection Using Deep Learning

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading