基于对抗网络的R波提取方法
By employing an adversarial network-based R-wave extraction method, which combines G and D networks with a genetic algorithm, the problem of R-wave detection being susceptible to noise and sample data interference was solved, achieving high accuracy and interference resistance in ECG signal R-wave extraction.
CN117752343BActive Publication Date: 2026-07-17SHANGHAI SID MEDICAL CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI SID MEDICAL CO LTD
- Filing Date
- 2023-12-29
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
In existing technologies, R-wave detection is easily affected by noise interference, and supervised deep learning methods are easily affected by sample data, leading to a decrease in detection accuracy.
Method used
An R-wave extraction method based on adversarial networks is adopted. By constructing G and D networks, a genetic algorithm is used to solve the R-wave position of the electrocardiogram signal, avoiding conventional filtering and threshold judgment, and improving anti-interference and accuracy.
Benefits of technology
It improves the accuracy of R-wave extraction, reduces noise interference, enhances the universality of the algorithm, avoids false detections and missed detections, and achieves higher signal accuracy.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117752343B_ABST
Abstract
本发明公开了基于对抗网络的R波提取方法，属于心电信号处理领域，基于对抗网络的R波提取方法，采用以下步骤操作：步骤一，采集至少N条心电信号；步骤二，构建出输入为心电信号长度Fs*T*1,输出为Fs*T*1的深度学习网络；步骤三，以心电信号作为输入，以0‑1之间的数字作为输出，构建D网络；步骤四，训练D网络；步骤五，训练G网络；步骤六，用遗传算法得出心电信号R波发生位置，完成提取；本发明中提升了心电信号R波提取的准确率，不会出现错检和漏检情况，降低了噪声的干扰性，且不会受到样本数据标注不准确及样本导联等情况的影响，信号的准确率更高，抗干扰性更好，能够提高现有算法的普适性。
Need to check novelty before this filing date? Find Prior Art