A method for detecting atrial fibrillation based on the time-frequency characteristics of single-lead ECG signals
A technology of time-frequency characteristics and detection methods, which is applied in diagnostic recording/measurement, medical science, diagnosis, etc., can solve the problems of low detection accuracy and differentiation of atrial fibrillation, and achieve high accuracy, improved accuracy, and good robustness Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0026] Specific Embodiment 1: This embodiment describes a method for detecting atrial fibrillation based on the time-frequency characteristics of single-lead ECG signals, which divides the original ECG signals into normal sinus rhythm, atrial fibrillation, other arrhythmias or excessive noise Four classes, described method comprises the steps:
[0027] Step 1: For the single-lead body surface ECG signal (the signal length is within the interval of 10 to 60 seconds), first perform a preprocessing operation to remove the baseline drift and part of the noise interference in the ECG signal;
[0028] Step 2: Calculate the change of signal energy in the time domain based on the sliding window method, and find the first window whose energy is not higher than the median value of the subsequent window energy. The corresponding part is removed from the signal, proceed to step 3; if the window does not exist, the signal is considered to be too noisy, and the detection process ends;
[0...
specific Embodiment approach 2
[0035] Specific embodiment two: the method for detecting atrial fibrillation based on the time-frequency characteristics of single-lead electrocardiographic signals described in specific embodiment one, the specific steps of the described step two are as follows:
[0036] (1) Divide the ECG signal into a series of windows, each window length is 1.8 seconds, and there is an overlapping area of 0.9 seconds between adjacent windows;
[0037] (2) Calculate the energy of the signal in each window, the formula is as follows:
[0038]
[0039] Among them, x is the discrete signal sequence in the window, x[n] represents the value of the nth sampling point in the window, T is the window duration, and N represents the number of sampling points in the window;
[0040] (3) Starting from the first window, compare its energy with the median value of the energy of each window in the signal until a window is found so that its energy is not higher than the median energy of the subsequent ...
specific Embodiment approach 3
[0044] Specific embodiment three: the method for detecting atrial fibrillation based on the time-frequency characteristics of single-lead electrocardiographic signals described in specific embodiment one, the specific steps of the described step five are as follows:
[0045] (1) Define the lower limit of the similarity within the group, that is, the minimum Pearson correlation coefficient of the heartbeat waveform within the group, and the value is about 0.8 (adjusted according to the strength of the signal noise);
[0046] (2) Calculate the Pearson correlation coefficient between the heartbeat waveforms in the same signal, the formula is as follows
[0047]
[0048] Among them, X and Y respectively represent the sampling point sequence corresponding to the two heartbeat waveforms, E represents the mathematical expectation, μ X Indicates the average value of each sampling point in X, μ Y Indicates the average value of each sampling point in Y, σ X Indicates the standard d...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More