Related Experiment Video
Updated: Nov 2, 2025
Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice
Published on: May 23, 2021
Identification of 27 abnormalities from multi-lead ECG signals: an ensembled SE_ResNet framework with Sign Loss
Zhaowei Zhu1, Xiang Lan2, Tingting Zhao1
1Ping An Technology, Beijing, People's Republic of China.
Insights
An advanced algorithm accurately identifies 27 cardiac abnormalities from 12-lead electrocardiograms (ECGs), improving diagnosis for cardiovascular disease. This AI framework combines deep learning and clinical rules for robust classification across diverse datasets.
Area of Science:
- Cardiology
- Artificial Intelligence
- Biomedical Engineering
Background:
- Cardiovascular disease (CVD) is a leading global cause of mortality.
- The 12-lead electrocardiogram (ECG) is a vital, accessible diagnostic tool for cardiac conditions.
- Accurate and early ECG interpretation is crucial for preventing severe CVD complications.
Purpose of the Study:
- To develop an automated algorithm for identifying 27 distinct ECG abnormalities using 12-lead ECG data.
- To enhance the accuracy and generalizability of ECG abnormality classification.
- To create a robust framework integrating deep learning and clinical expertise.
Main Methods:
- Applied pre-processing techniques to harmonize diverse ECG data sources.
- Ensembled two SE_ResNet models with a rule-based model for classification.
- Introduced a Sign Loss function to address class imbalance and improve model generalizability.
Main Results:
- Achieved a 3rd place ranking out of 40 participants in the PhysioNet/Computing in Cardiology Challenge (2020).
- Attained a challenge validation score of 0.682 and a full test score of 0.514.
- Demonstrated robust performance across multiple datasets from various countries.
Conclusions:
- The developed framework accurately and reliably classifies multiple ECG abnormalities from multi-lead signals.
- The approach effectively handles data discrepancies and class imbalance issues.
- The combined deep neural network and clinical knowledge framework shows significant promise for automated ECG analysis.
Abstract:
Objective. Cardiovascular disease is a major threat to health and one of the primary causes of death globally. The 12-lead ECG is a cheap and commonly accessible tool to identify cardiac abnormalities. Early and accurate diagnosis will allow early treatment and intervention to prevent severe complications of cardiovascular disease. Our objective is to develop an algorithm that automatically identifies 27 ECG abnormalities from 12-lead ECG databases.Approach. Firstly, a series of pre-processing methods were proposed and applied on various data sources in order to mitigate the problem of data divergence. Secondly, we ensembled two SE_ResNet models and one rule-based model to enhance the performance of various ECG abnormalities' classification. Thirdly, we introduce a Sign Loss to tackle the problem of class imbalance, and thus improve the model's generalizability.Main results. In the PhysioNet/Computing in Cardiology Challenge (2020), our proposed approach achieved a challenge validation score of 0.682, and a full test score of 0.514, placed us 3rd out of 40 in the official ranking.Significance. We proposed an accurate and robust predictive framework that combines deep neural networks and clinical knowledge to automatically classify multiple ECG abnormalities. Our framework is able to identify 27 ECG abnormalities from multi-lead ECG signals regardless of discrepancies in data sources and the imbalance of data labeling. We trained our framework on five datasets and validated it on six datasets from various countries. The outstanding performance demonstrate the effectiveness of our proposed framework.
