ECG Signal Denoising That Preserves QRS Wave Peak Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for removing noise from electrocardiography signals, particularly those acquired by wearable devices, often distort critical features like the P wave, T wave, and R wave, compromising the signal's reference value due to noise interference such as electromyography and motion noise.
Innovation Solution
A method that identifies sampling points on the QRS complex in electrocardiography signals by analyzing the continuous abrupt change performance, using feature sequences and thresholds to determine if noise should be removed, thereby preserving the signal's reference value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional noise removal methods are applied to electrocardiography signals, then noise is reduced, but wave peaks of P wave, T wave and R wave are disturbed
Solution Approach 1:
The patent applies different processing strategies to different regions of the ECG signal. By calculating the gradient magnitude at each sampling point and comparing it to a threshold, the method identifies regions with significant signal changes (likely containing wave peaks) and preserves them, while applying noise removal to regions with smaller gradients. This local differentiation allows noise reduction without distorting critical wave peak features.
Solution Approach 2:
The patent uses a dynamic thresholding approach where the gradient threshold is determined adaptively based on the local signal characteristics. The threshold is set as a proportion of the maximum gradient value in the signal, allowing the noise removal process to adapt to varying signal amplitudes and noise levels across different segments of the ECG recording, thereby preserving wave peaks while removing noise.
2Reliability
If noise removal is applied to stationary segments of the signal, then noise effects are reduced, but critical features may be lost
Solution Approach 1:
The patent transforms the ECG signal from the time domain to the frequency domain using Fast Fourier Transform (FFT), applies frequency-based filtering to remove noise components, and then transforms back to the time domain using inverse FFT. This parameter transformation allows selective removal of noise frequencies while preserving the frequency components corresponding to critical ECG features, thereby maintaining signal reliability without losing important information.
3Object-affected harmful factors
If aggressive filtering is used to remove noise, then noise is effectively reduced, but the electrocardiography signal loses reference value
Solution Approach 1:
The patent applies a gradient-based thresholding criterion that processes only the portions of the signal where the gradient exceeds a certain threshold. By setting the threshold as a proportion (e.g., 0.1) of the maximum gradient, the method applies noise removal selectively rather than uniformly across the entire signal. This partial action approach removes noise effectively while preserving the reference value of critical features that exhibit steep gradients.
Data Source
Figure 1~2
Figure 3(A)~3(C)
Figure 4(A)~4(B)
AI summary
A method for removing noise from a signal and an electronic device are provided. According to the method, by virtue of a continuous abrupt change performance of a wave peak having reference significance in an electrocardiography signal, whether a signal point is a point on a signal wave having reference significance can be identified during noise removal, and whether to remove noise from the signal point can be determined. The method is implemented, so that noise can be effectively removed from the electrocardiography signal on the premise that a reference value of the electrocardiography signal is guaranteed.