Related Experiment Video
Updated: Jun 18, 2025
Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice
Published on: May 23, 2021
Advanced integration of 2DCNN-GRU model for accurate identification of shockable life-threatening cardiac
Abduljabbar S Ba Mahel1, Shenghong Cao1, Kaixuan Zhang1
1School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China.
Insights
A new hybrid deep learning method accurately detects dangerous arrhythmias from short ECG fragments. This approach converts electrocardiogram (ECG) signals into images for improved classification of life-threatening heart conditions.
Area of Science:
- Cardiology
- Biomedical Engineering
- Artificial Intelligence
Background:
- Cardiovascular diseases pose significant health risks, necessitating early detection of arrhythmias.
- Accurate identification of dangerous arrhythmias is critical for patient outcomes.
- Existing methods may struggle with the complexity and variability of electrocardiogram (ECG) data.
Purpose of the Study:
- To develop a novel hybrid method for automatic detection of dangerous arrhythmias using short ECG segments.
- To classify four types of shockable arrhythmias: ventricular flutter, ventricular fibrillation, Torsade de Pointes, and high-rate ventricular tachycardia.
- To improve the accuracy and efficiency of arrhythmia diagnosis in cardiac patients.
Main Methods:
- Continuous Wavelet Transform (CWT) to convert ECG signals into time-frequency images (scalograms).
- A novel hybrid deep learning neural network architecture for image classification.
- Utilized real ECG data from PhysioNet and synthetic data (SMOTE) to address class imbalance.
Main Results:
- The proposed method achieved high performance metrics: 97.75% accuracy, sensitivity, precision, and F1-score, with 99.25% specificity.
- Demonstrated superior performance compared to traditional arrhythmia detection methods.
- The model showed adaptability and generality across multiple datasets.
Conclusions:
- The developed hybrid deep learning approach offers a highly accurate and effective tool for detecting dangerous arrhythmias.
- This method holds significant clinical value for enhancing the diagnosis and treatment of life-threatening arrhythmias.
- The model's performance and adaptability suggest potential for widespread clinical application in cardiac care.
Abstract:
Cardiovascular diseases remain one of the main threats to human health, significantly affecting the quality and life expectancy. Effective and prompt recognition of these diseases is crucial. This research aims to develop an effective novel hybrid method for automatically detecting dangerous arrhythmias based on cardiac patients' short electrocardiogram (ECG) fragments. This study suggests using a continuous wavelet transform (CWT) to convert ECG signals into images (scalograms) and examining the task of categorizing short 2-s segments of ECG signals into four groups of dangerous arrhythmias that are shockable, including ventricular flutter (C1), ventricular fibrillation (C2), ventricular tachycardia torsade de pointes (C3), and high-rate ventricular tachycardia (C4). We propose developing a novel hybrid neural network with a deep learning architecture to classify dangerous arrhythmias. This work utilizes actual electrocardiogram (ECG) data obtained from the PhysioNet database, alongside artificially generated ECG data produced by the Synthetic Minority Over-sampling Technique (SMOTE) approach, to address the issue of imbalanced class distribution for obtaining an accuracy-trained model. Experimental results demonstrate that the proposed approach achieves high accuracy, sensitivity, specificity, precision, and an F1-score of 97.75%, 97.75%, 99.25%, 97.75%, and 97.75%, respectively, in classifying all the four shockable classes of arrhythmias and are superior to traditional methods. Our work possesses significant clinical value in real-life scenarios since it has the potential to significantly enhance the diagnosis and treatment of life-threatening arrhythmias in individuals with cardiac disease. Furthermore, our model also has demonstrated adaptability and generality for two other datasets.
