Related Experiment Video
Updated: Jun 14, 2025
12:10
Semi-automated Optical Heartbeat Analysis of Small Hearts
Published on: September 16, 2009
12.2K
[Early classification and recognition algorithm for sudden cardiac arrest based on limited electrocardiogram data
Xingzeng Cha1, Yue Zhang1, Yifei Zhang1
1School of Electronic Science and Engineering, University of Electronic Science and technology, Chengdu 610054, P. R. China.
Summary
This study introduces a deep transfer learning algorithm for early prediction of sudden cardiac arrest (SCA) using limited electrocardiogram (ECG) data. The method accurately identifies high-risk ECG signals minutes before SCA onset.
Area of Science:
- Cardiology
- Artificial Intelligence
- Biomedical Engineering
Context:
- Sudden cardiac arrest (SCA) is a critical arrhythmia with limited clinical ECG data for research.
- Existing deep learning models require extensive datasets, which are scarce for SCA prediction.
Purpose:
- To develop an early prediction and classification algorithm for SCA using deep transfer learning.
- To overcome the challenge of limited ECG data for training predictive models.
Summary:
- A two-stage deep transfer learning approach using a lightweight convolutional neural network was employed.
- Heart rate variability features were extracted from limited ECG data for pre-training and fine-tuning.
- The algorithm achieved high accuracy (91.79%), sensitivity (87.00%), and specificity (96.63%) in predicting SCA onset within 30 minutes.
Impact:
- Enables early and accurate detection of high-risk ECG signals preceding SCA.
- Addresses the need for large datasets in deep learning for SCA prediction.
- Offers a promising tool for improving patient outcomes and reducing mortality from SCA.
Keywords:
Deep transfer learning.ElectrocardiogramHeart rate featuresRapid response systemsSudden cardiac arrest
