Related Experiment Video
Updated: Aug 23, 2025
Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice
Published on: May 23, 2021
A novel multi-scale 2D CNN with weighted focal loss for arrhythmias detection on varying-dimensional ECGs
Pan Xia1,2, Zhengling He1,2, Zhongrui Bai1
1State Key Laboratory of Transducer Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, People's Republic of China.
Insights
A novel algorithm automatically identifies 30 arrhythmias using multi-scale 2D CNNs and reduced-lead electrocardiograms (ECGs). This approach shows potential for accurate arrhythmia detection, even with fewer ECG leads.
Area of Science:
- Cardiology
- Artificial Intelligence
- Signal Processing
Background:
- Electrocardiograms (ECGs) are crucial for diagnosing arrhythmias, impacting cardiovascular disease prevention and treatment.
- Accurate and early arrhythmia detection is vital for patient outcomes.
- Current methods may benefit from advanced algorithmic approaches for improved accuracy and efficiency.
Purpose of the Study:
- To develop a multi-scale 2D Convolutional Neural Network (CNN) algorithm for automatic identification of 30 arrhythmias.
- To evaluate the algorithm's performance using varying-dimensional ECG signals (12-lead down to 2-lead).
- To investigate the impact of reduced-lead ECGs on arrhythmia detection accuracy.
Main Methods:
- Proposed a novel multi-scale 2D CNN to capture pathological ECG information at various scales.
- Explored the influence of different convolution kernel sizes and subnetwork architectures.
- Implemented weighted focal loss to address class imbalance in multi-label arrhythmia classification.
- Compared model performance across 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead ECGs.
Main Results:
- Achieved competitive scores on a hidden test set, comparable to top teams in the PhysioNet/Computing in Cardiology Challenge 2021.
- Demonstrated strong performance across all tested ECG dimensions, including reduced-lead configurations.
- Validated the effectiveness of multi-scale perception and weighted focal loss for arrhythmia detection.
Conclusions:
- A multi-scale framework effectively detects 30 arrhythmias from varying-dimensional ECGs.
- Multi-scale perception fields are potentially necessary for comprehensive pathological information capture.
- Weighted focal loss aids in managing class imbalance and improving cross-dataset generalization.
- Reduced-lead ECGs (e.g., 4-lead, 3-lead) can achieve performance comparable to 12-lead ECGs, highlighting potential for streamlined diagnostics.
Abstract:
Objective. The ECG is a standard diagnostic tool for identifying many arrhythmias. Accurate diagnosis and early intervention for arrhythmias are of great significance to the prevention and treatment of cardiovascular disease. Our objective is to develop an algorithm that can automatically identify 30 arrhythmias by using varying-dimensional ECG signals.Approach. In this paper, we firstly proposed a novel multi-scale 2D CNN that can effectively capture pathological information from small-scale to large-scale from ECG signals to identify 30 arrhythmias from 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead ECGs. Secondly, we explored the effects of varying convolution kernels sizes and branch subnetworks on the model's performance for each arrhythmia. Thirdly, we introduced the weighted focal loss to alleviate the positive-negative class imbalance problem in the multi-label arrhythmias classification. Fourthly, we explored the utility of reduced-lead ECGs in detecting arrhythmias by comparing the performances of models on varying-dimensional ECGs.Main results. As a follow-up entry after the PhysioNet/Computing in Cardiology Challenge (2021), our proposed approach achieved the official test scores of 0.52, 0.47, 0.53, 0.51, and 0.50 for the 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead ECGs on the hidden test set (comparable to that of 6th, 11th, 4th, 5th, and 7th out of 39 teams in the Challenge).Significance. A multi-scale framework capable of detecting 30 arrhythmias from varying-dimensional ECGs was proposed in our work. We preliminarily verified that the multi-scale perception fields may be necessary to capture more comprehensive pathological information for arrhythmias detection. Besides, we also verified that the weighted focal loss may alleviate the positive-negative class imbalance and improve the model's generalization performance on the cross-dataset. In addition, we observed that some reduced-lead models, such as the 4-lead and 3-lead models, can even achieve performance that is almost comparable to that of the 12-lead model. The excellent performance of our proposed framework demonstrates its great potential in detecting a wide range of arrhythmias.
Related Concept Videos
01:25ECG Interpretation of Arrhythmias II: Atrial, Junctional and Ventricular Arrhythmias
01:25Correlation between ECG and Cardiac Cycle
A cardiac action potential originates in the SA node and spreads throughout the atria and the AV node in approximately 0.03 seconds. This results in the P wave in an ECG and triggers atrial contraction. The action potential is then briefly slowed at the AV node, allowing the atria to contract and fill the ventricles with blood before...
01:30Dysrhythmias V: Evaluating Dysrhythmias
01:29Electrocardiogram
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and...