A method for detecting R waves of human electrocardiogram

Abstract

The present invention discloses a method for detecting the R wave of human electrocardiogram, which relates to the technical field of electrocardiogram signal detection. The system comprises the following components: Gene expression data and electrocardiogram signal acquisition: Obtain the gene expression level information of the individual's whole genome through gene detection technology. At the same time, use a multi-channel electrocardiogram sensor to collect electrocardiogram signals, and ensure that the acquisition time interval between the two is short. By introducing gene expression data, the present invention can perform personalized electrocardiogram R wave detection for individuals with specific gene mutations or a tendency to have familial genetic cardiovascular diseases. This method not only considers the conventional characteristics of the electrocardiogram signal itself, but also combines information at the gene level, so as to be able to more accurately identify the R wave, reduce the situation of false detection and missed detection, and is of great significance for improving the accuracy of cardiovascular disease diagnosis, early detection of cardiovascular abnormalities, and formulating personalized treatment plans.

Description

Technical Field

[0001] The present invention relates to the technical field of electrocardiogram signal detection, and particularly to a method for detecting human electrocardiogram R waves. Background Art

[0002] The detection of electrocardiogram R waves plays a crucial role in the fields of cardiovascular disease diagnosis, arrhythmia analysis, etc. As a direct indicator reflecting the electrophysiological activities of the heart, the accurate detection of electrocardiogram signals is of great significance for evaluating heart function and diagnosing heart diseases. With the continuous progress of medical technology, the methods for detecting electrocardiogram R waves are also constantly improved and perfected, aiming to improve the accuracy and reliability of detection, so as to provide more accurate diagnosis and treatment plans for patients with cardiovascular diseases.

[0003] Although the traditional methods for detecting electrocardiogram R waves have achieved certain results, there are still some obvious limitations in practical applications. The traditional methods usually identify the electrocardiogram signals based on their conventional characteristics and use general thresholds or algorithms to detect R waves. However, due to genetic differences among different individuals, especially those with specific gene mutations or a family genetic tendency to cardiovascular diseases, their cardiovascular physiological characteristics are significantly different, which leads to significant differences in the characteristics of R waves in electrocardiogram signals. The traditional general detection methods are difficult to fully consider these individual specificities, often resulting in inaccurate detection results, which in turn affects the subsequent accurate diagnosis and condition assessment of cardiovascular diseases. In addition, electrocardiogram signals are extremely vulnerable to various noise interferences during the acquisition process, such as electromyographic noise, power frequency interference, etc. These noises will seriously interfere with the accuracy of R wave detection. The traditional detection algorithms lack the ability of adaptive error correction. Once false detections, missed detections, etc. occur, they can only output incorrect results and cannot adjust in real time according to the actual situation during the detection process to obtain the correct results, which is particularly disadvantageous in the complex and changeable human physiological state and the actual detection environment.

[0004] In view of the above problems, it is necessary to detect the existing methods for detecting human electrocardiogram R waves. By deeply analyzing the correlation between gene expression data and electrocardiogram signals, the personalized adjustment of the R wave detection algorithm can be realized. Therefore, it is of great significance to develop a method for detecting human electrocardiogram R waves that can comprehensively achieve the above characteristics. Summary of the Invention

[0005] The object of the present invention is to make up for the deficiencies of the prior art and provide a method for detecting the R wave of human electrocardiogram. By deeply analyzing the correlation between gene expression data and electrocardiogram signals, personalized adjustment of the R wave detection algorithm is realized, significantly improving the accuracy and specificity of detection. At the same time, by adopting an advanced combined preprocessing method of adaptive noise suppression and feature enhancement, the noise in the electrocardiogram signal is effectively suppressed and the R wave features are enhanced, further improving the detection accuracy. In addition, an adaptive fault-tolerant R wave detection algorithm automatically identifies and corrects possible false detections, missed detections, etc. during the detection process, ensuring that the detection algorithm still maintains high accuracy and reliability in complex environments and changing human physiological states.

[0006] To solve the above technical problems, the present invention provides the following technical solution: A method for detecting the R wave of human electrocardiogram, the method comprising the following specific steps:

[0007] Collection of gene expression data and electrocardiogram signals: Obtain the gene expression level information of the individual's whole genome through gene detection technology. At the same time, use a multi-channel electrocardiogram sensor to collect electrocardiogram signals, and ensure that the time interval between the two collections is short. Transmit the collected gene expression data and electrocardiogram signals to the computer system;

[0008] Gene-electrocardiogram signal correlation analysis: Carry out a combined preprocessing operation of adaptive noise suppression and feature enhancement on the collected gene expression data and electrocardiogram signals. Design a filtering algorithm that can automatically adjust parameters according to the real-time noise level of the electrocardiogram signal. Specifically, perform a short-time Fourier transform on the collected electrocardiogram signal to convert it to the frequency domain, and synchronously analyze the energy distribution of the frequency domain signal to initially determine the type and distribution range of the noise. According to the determined noise type and its distribution, set initial filtering parameters for different types of noise. During the filtering process, monitor the noise level of the filtered electrocardiogram signal in real time by calculating the root mean square error of the filtered signal, and use the multi-resolution analysis method based on wavelet transform to enhance the feature information of the R wave. In the correlation analysis stage, construct a gene-electrocardiogram feature mapping model, and train it based on the known gene expression conditions and the corresponding electrocardiogram signal feature data. Input the preprocessed gene expression data of the current individual into the trained gene-electrocardiogram feature mapping model to obtain the R wave feature prediction result based on gene expression;

[0009] Personalized adjustment of the R wave detection algorithm: According to the R wave feature prediction result, respectively adjust the amplitude threshold, slope judgment parameter, and adjacent R wave time interval judgment parameter in the detection algorithm according to the predicted changes in the R wave amplitude, morphology, and occurrence time to adapt to individual differences;

[0010] Adaptive Fault-Tolerant R-Wave Detection: Use a personalized adjusted R-wave detection algorithm for electrocardiogram (ECG) R-wave detection, and implement an adaptive fault-tolerant mechanism during the detection process. Monitor for false detections and missed detections by comparing with historical data. If the threshold is not exceeded, adjust the parameters or algorithm in real-time according to the current ECG signal characteristics and historical data and re-detect. If the threshold is exceeded, output a prompt message to indicate the need to re-collect the signal or check the device;

[0011] Detection Result Output and Storage: Output the detected R-wave characteristic information for subsequent arrhythmia analysis and cardiovascular disease diagnosis. At the same time, store the gene expression data, ECG signal, and R-wave detection results during this detection process in the local hard drive and cloud server.

[0012] Furthermore, in the gene expression data and ECG signal acquisition step, obtain the gene expression level information of the individual's whole genome through gene detection techniques. The gene detection techniques include gene chip technology and high-throughput sequencing technology. When using gene chip technology, process the individual's blood sample or tissue sample and hybridize it with the gene chip, and detect the fluorescence signal intensity of each gene locus on the chip to obtain the gene expression level information. When using high-throughput sequencing technology, perform nucleic acid extraction and library construction operations on the individual's blood sample or tissue sample and then put it into a high-throughput sequencing instrument for sequencing. Calculate the expression level of each gene through bioinformatics analysis methods on the gene sequence information obtained by sequencing to form gene expression data.

[0013] Even further, in the gene expression data and ECG signal acquisition step, use a multi-channel ECG sensor to collect ECG signals. Select a three-channel or twelve-channel ECG sensor and determine the number of channels according to the detection requirements and accuracy requirements. Paste the electrode pads of the ECG sensor at the ECG electrode placement positions on the individual's chest. The ECG sensor receives the weak electrical signals generated by the heart activity through the electrode pads and converts them into a digital signal form that can be processed by the computer system. During the acquisition process, ensure that the time interval between the gene expression data and ECG signal acquisition is short to maintain the relevance of the two sets of data.

[0014] Even further, in the gene-ECG signal correlation analysis step, design a filtering algorithm that can automatically adjust parameters according to the real-time noise level of the ECG signal. The algorithm formula of the filtering algorithm is: where G(f,t) represents the gain function of the ECG signal after filtering at frequency f and time t, f ∈ (t) is the center frequency reference value that changes dynamically with time t and is used to locate the main noise frequency components, B Ψ(t) is a bandwidth parameter that varies dynamically with time t and is used to control the frequency range of filtering. n is an exponential parameter that determines the shape of the filtering curve, with a value range of 3 ≤ n ≤ 8. μ is a weight coefficient that balances the effects of noise suppression and R-wave feature enhancement, with a value range of 0 < μ < 1. S represents the number of scale levels of wavelet transform decomposition, and ω s is a weighting factor at the s-th wavelet scale level, and X s (f) is the frequency-domain representation of the electrocardiogram signal at the s-th wavelet scale level, obtained by performing wavelet transform on the electrocardiogram signal.

[0015] Furthermore, in the gene-electrocardiogram signal association analysis step, training is carried out based on known gene expression conditions and corresponding electrocardiogram signal feature data. The model training formula is: where y k represents the predicted value related to a specific feature of the R wave output by the model. σ is an activation function used to introduce non-linearity, enabling the model to learn more complex relationships between gene expression and electrocardiogram signal features. w kl is the weight parameter connecting the l-th input feature and the k-th output feature. During the training process, the weights are continuously adjusted through the backpropagation algorithm to enable the model to accurately learn the potential association relationship between gene expression and electrocardiogram signal features. b k is the bias term corresponding to the k-th output feature, which is adjusted during the training process.

[0016] Furthermore, in the adaptive fault-tolerant R-wave detection step, according to the R-wave feature prediction results obtained from the gene-electrocardiogram signal association analysis, the R-wave detection algorithm is optimized and adjusted specifically. For the prediction results showing changes in the amplitude of the R wave, when it is predicted that the amplitude of the R wave will decrease, the lower amplitude threshold in the R-wave detection algorithm is decreased; when it is predicted that the amplitude of the R wave will increase, the upper amplitude threshold in the R-wave detection algorithm is increased. For the prediction results indicating changes in the morphology of the R wave, when it is predicted that the morphology of the R wave becomes flatter, the slope threshold used to judge the R-wave peak is decreased; when it is predicted that the morphology of the R wave becomes sharper, the slope threshold is increased. For the prediction that the occurrence time of the R wave is advanced or delayed, when it is predicted that the occurrence time of the R wave is advanced, the upper limit of the judgment of the time interval between adjacent R waves is shortened; when it is predicted that the occurrence time of the R wave is delayed, the lower limit of the judgment of the time interval between adjacent R waves is extended.

[0017] Furthermore, in the adaptive fault-tolerant R-wave detection step, according to the R-wave feature prediction results obtained from the gene-electrocardiogram signal association analysis, the R-wave detection algorithm is optimized and adjusted specifically. For the prediction results showing changes in the amplitude of the R wave, the formula Adjust the amplitude threshold in the R-wave detection algorithm, where is the adjusted amplitude threshold, is the original amplitude threshold, and α A is the preset amplitude threshold adjustment coefficient, and ΔA pred is the change in R-wave amplitude predicted based on the gene-ECG feature mapping model. For the prediction result indicating that the morphology of the R-wave will change, use the formula Adjust the slope judgment parameter in the R-wave detection algorithm, where is the adjusted slope threshold, is the original slope threshold, and α S is the preset slope judgment parameter adjustment coefficient, and ΔS pred is the change in slope corresponding to the change in R-wave morphology predicted based on the gene-ECG feature mapping model. For the prediction that the occurrence time of the R-wave is advanced or delayed, use the formula Adjust the judgment parameter for the time interval between adjacent R-waves in the R-wave detection algorithm, where is the adjusted judgment parameter for the time interval between adjacent R-waves, is the original judgment parameter for the time interval between adjacent R-waves, and α I ×ΔT pred is the preset judgment parameter adjustment coefficient for the time interval between adjacent R-waves, and ΔT pred is the change in the occurrence time of the R-wave predicted based on the gene-ECG feature mapping model.

[0018] Furthermore, in the self-adaptive fault-tolerant R-wave detection step, it is judged whether there are misdetection and missed detection abnormalities by comparing with historical data using the formula: where, is the currently detected distance value between adjacent R-waves, is the average value of the distances between adjacent R-waves in the historical detection data, obtained by averaging the distance values between adjacent R-waves detected multiple times before, and θ D is the deviation threshold of the distance between adjacent R-waves, with the value range of 0.15 ≤ θ D ≤ 0.35, is the currently detected R-wave amplitude, is the R-wave amplitude predicted according to the gene-ECG feature mapping model, and θ A is the R-wave amplitude deviation threshold, with the value range of 0.25 ≤ θ A ≤ 0.45, is the currently detected time interval value between adjacent R-waves, is the average value of the time intervals between adjacent R-waves in the historical detection data, and θ T is the deviation threshold of the time interval between adjacent R-waves, with the value range of 0.1 ≤ θT ≤ 0.3.

[0019] Further, in the self - adaptive fault - tolerant R - wave detection step, by comparing with historical data to monitor whether there are abnormal misdetections and missed detections. If the threshold is not exceeded, the parameters or algorithms are adjusted in real - time according to the current ECG signal characteristics and historical data and redetection is performed. The adjustment formula for the time interval is as follows: Where, is the parameter for judging the time interval between adjacent R - waves currently in use, is the detected distance between adjacent R - waves, and the average value of the distances between adjacent R - waves in the historical detection data, is the deviation situation, β I is the adjustment coefficient, and the value range is - 0.3 ≤ β I ≤ 0.3. Calculate the adjusted parameter for judging the time interval between adjacent R - waves according to the formula For other detection parameters, including amplitude thresholds and filtering parameters, when adjusting, determine the adjusted parameter values according to the deviation situation between the current characteristics and historical data and the set adjustment coefficient, and use the adjusted parameters or algorithms to perform redetection to obtain correct results.

[0020] Compared with the prior art, this method for detecting the R - wave of human electrocardiogram has the following beneficial effects:

[0021] First, by introducing gene expression data, the present invention can perform personalized electrocardiogram R - wave detection for individuals with specific gene mutations or a genetic tendency to cardiovascular diseases. This method not only considers the conventional characteristics of the electrocardiogram signal itself but also combines information at the gene level, thereby being able to more accurately identify R - waves, reducing the situations of misdetections and missed detections, and is of great significance for improving the accuracy of cardiovascular disease diagnosis, early detection of cardiovascular abnormalities, and formulating personalized treatment plans.

[0022] Second, by adopting the combined pre - processing method of adaptive noise suppression and feature enhancement, and the self - adaptive fault - tolerant R - wave detection algorithm, the present invention can maintain a high detection accuracy in complex detection environments and changing human physiological states. By real - time monitoring the noise level and R - wave characteristics of the electrocardiogram signal, dynamically adjusting the filtering parameters and detection algorithms, it can effectively cope with common interference factors such as electromyogram noise and power - frequency interference, and at the same time enhance the feature information of R - waves.

[0023] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 It is a flowchart of an operation for a method of detecting the R wave of human electrocardiogram.

[0026] Figure 2 It is a flowchart of a method for detecting the R wave of human electrocardiogram. Specific embodiments

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0028] Embodiment 1

[0029] This embodiment details the specific application of a method for detecting the R wave of human electrocardiogram in the field of early screening of cardiovascular diseases. By introducing the present invention, for individuals with a family history of cardiovascular diseases, more accurate detection of the R wave of human electrocardiogram can be achieved.

[0030] In the step of gene expression data and electrocardiogram signal acquisition, patients undergoing cardiac rehabilitation treatment are selected as the monitoring objects. High-throughput sequencing technology is used to collect the gene expression data of the patients. First, blood samples of the patients are obtained, and pretreatment operations such as nucleic acid extraction and library construction are performed on the samples to prepare a DNA or RNA library suitable for sequencing. The constructed library is placed into a high-throughput sequencing instrument, and sequencing is carried out according to the operating procedures of the instrument. After sequencing is completed, the gene sequence information obtained by sequencing is processed by bioinformatics analysis methods to calculate the expression levels of each gene, forming the gene expression data of the patients. At the same time, a three-channel electrocardiogram sensor is used to collect the electrocardiogram signals of the same patients. Electrode patches are pasted on the patients' chests according to the standard electrocardiogram electrode placement positions to ensure good contact between the electrode patches and the skin, reduce signal interference, collect electrocardiogram signals for a period of time (such as 10 minutes), and transmit the collected gene expression data and electrocardiogram signals to a computer system through a data cable for subsequent processing.

[0031] In the step of gene-electrocardiogram signal correlation analysis, the collected electrocardiogram signals are jointly preprocessed by adaptive noise suppression and feature enhancement. First, the designed filtering algorithm is used Perform short-time Fourier transform on the electrocardiogram (ECG) signal, convert it to the frequency domain, analyze the spectral characteristics and energy distribution, and find that there is a certain degree of power frequency interference and electromyogram (EMG) noise. For power frequency interference, according to the power frequency of the region (such as 50 Hz), set f in the filtering algorithm ∈ (t) to be close to 50 Hz, and determine the appropriate B according to the actual situation of the noise Ψ (t). For EMG noise, determine the corresponding f by analyzing its energy distribution ∈ (t) and B Ψ (t). During the filtering process, calculate the root mean square error (RMSE) of the filtered signal to monitor the noise level in real time. If the RMSE value is found to be high, appropriately increase the value of B according to the parameter adjustment rule in the formula Ψ (t) to further suppress the noise. At the same time, use the multi-resolution analysis method based on wavelet transform to enhance the R-wave characteristics. Perform wavelet transform on the ECG signal, decompose it into different scales (such as setting S = 5 scale layers), and obtain the frequency domain representation X of the ECG signal at each scale s (f). According to the characteristic performance of the R-wave at different scales obtained from the analysis of a large number of normal and abnormal ECG signals in the past, determine the weighting factor ω at each scale layer s . For the scale layer where the R-wave peak is located, set a relatively large ω s value, and perform weighted processing on the decomposed coefficients according to the part in the formula to make the R-wave more prominent and complete in the reconstructed signal. In the correlation analysis stage, construct a gene-ECG feature mapping model, select the convolutional neural network (CNN) as the training algorithm, collect a large amount of gene expression data and corresponding ECG signal feature data of individuals with a known family history of cardiovascular diseases as the training set, and determine that the input features of the model include the expression level values of each gene in the preprocessed gene expression data and the relevant feature values in the ECG signal feature data (such as historical data such as the amplitude, slope, and time interval between adjacent R-waves of the R-wave), and the output is the predicted value related to a specific feature of the R-wave (such as the predicted change ratio of the R-wave amplitude, the degree of morphological change, the advance or delay amount of the occurrence time, etc.). Perform model training according to the formula . Among them, the activation function σ selects RelU. During the training process, adjust the weight parameter w by minimizing the mean square error loss function (N is the number of training samples, y k,i is the predicted value of the i-th training sample, is the actual observed value of the eighth training sample) and the bias term b kl k ​, continuously optimize the model, input the gene expression data of the current individual after preprocessing into the trained gene-ECG feature mapping model, and obtain the R-wave feature prediction result based on gene expression. For example, it is predicted that the amplitude of the R wave in the ECG signal of this individual may decrease by a certain proportion, and the shape of the R wave may become smoother.

[0032] In the step of personalized adjustment of the R-wave detection algorithm, according to the above prediction results, the traditional R-wave detection algorithm is personalized adjusted. Since it is predicted that the amplitude of the R wave may decrease, the formula is used to adjust the amplitude threshold in the R-wave detection algorithm. The original amplitude threshold in the traditional R-wave detection algorithm is obtained by statistically analyzing the amplitudes of R waves in a large number of normal ECG signal samples to get the average amplitude reference value A0 of the R wave under normal conditions. Based on the predicted change in the amplitude of the R wave ΔA pred (assuming that it is predicted that the amplitude of the R wave will decrease by 20%, then ΔA pred =-0.2A0), let the amplitude threshold adjustment coefficient α A =-0.5, and calculate the adjusted amplitude threshold for R-wave detection according to the formula Accordingly, the lower limit threshold of the amplitude in the R-wave detection algorithm is reduced. Since it is predicted that the shape of the R wave becomes smoother, the formula is used to adjust the slope judgment parameter in the R-wave detection algorithm. The original slope threshold in the traditional R-wave detection algorithm is based on the predicted change in the slope ΔS corresponding to the change in the shape of the R wave pred (assuming that it is predicted that the shape of the R wave becomes smoother and the slope will decrease by a certain value, then ΔS pred is the corresponding negative slope change amount), let the slope judgment parameter adjustment coefficient α S =-0.4, and calculate the adjusted slope threshold for judging the R-wave peak according to the formula and reduce the slope threshold for judging the R-wave peak.

[0033] In the step of adaptively fault-tolerant R-wave detection, initialize the detection parameters and related variables, set the false detection times threshold to 3 and the missed detection times threshold to 5. According to the personalized adjusted R-wave detection algorithm, search for R waves in the preprocessed ECG signal, identify the position, amplitude, shape and other feature information of the R waves, and record the results of each detection and related intermediate data, such as the distance between adjacent R waves, the stability index of the overall signal, etc. During the detection process, monitor in real time whether abnormal situations such as false detection and missed detection occur, and use the formula for judgment, where is the value of the distance between adjacent R waves detected currently, is the average value of the adjacent R - wave intervals in the historical detection data. Let θ D = 0.25, is the amplitude of the currently detected R - wave, is the amplitude of the R - wave predicted according to the gene - electrocardiogram feature mapping model. Let θ A = 0.35, is the value of the currently detected adjacent R - wave time interval, is the average value of the adjacent R - wave time intervals in the historical detection data. Let θ T = 0.2. If an abnormal situation is detected and the number of false detections and missed detections does not exceed the set threshold, for example, if the relative deviation of the adjacent R - wave interval from the historical average interval exceeds θ D , taking the adjustment of the adjacent R - wave time interval judgment parameter as an example, the formula is used for adjustment. Given the currently used adjacent R - wave time interval judgment parameter Let the adjustment coefficient β I = - 0.2. According to the deviation situation between the currently detected adjacent R - wave interval and the historical average interval, the adjusted adjacent R - wave time interval judgment parameter is calculated according to the formula, and then the detection is retried using the adjusted parameter or algorithm to obtain the correct result.

[0034] In the detection result output and storage step, the characteristic information such as the position, amplitude, and morphology of the detected R - wave is output through the display screen in the form of a visual chart. The position information is displayed with the time axis as the abscissa and the electrocardiogram signal amplitude as the ordinate. The amplitude information is marked with specific numerical values at the corresponding positions. The morphology information is presented by drawing the R - wave curve and marking the key feature points. The gene expression data, electrocardiogram signal, and R - wave detection results in the current detection process are stored in the local hard disk and classified and labeled according to information such as detection time and individual identification for subsequent reference and further analysis.

[0035] In summary, the present invention can achieve more accurate detection of the human electrocardiogram R - wave for individuals with a family history of cardiovascular diseases in the field of early screening of cardiovascular diseases, helping to detect potential cardiac rhythm abnormalities and signs of cardiovascular lesions earlier, and providing strong support for subsequent diagnosis and treatment.

[0036] Embodiment Two

[0037] This embodiment details the specific application of a human electrocardiogram R - wave detection method in the field of personalized cardiac rehabilitation monitoring. By introducing the present invention, accurate detection and dynamic monitoring of the R - wave are achieved based on the gene expression situation and real - time electrocardiogram signal of the patient individual, which helps rehabilitation therapists and doctors better grasp the changes in the patient's cardiac condition and adjust the rehabilitation plan in a timely manner.

[0038] In the gene expression data and electrocardiogram (ECG) signal acquisition step, patients undergoing cardiac rehabilitation treatment are selected as the monitoring objects. High-throughput sequencing technology is used to collect the gene expression data of the patients. First, blood samples of the patients are obtained, and preprocessing operations such as nucleic acid extraction and library construction are performed on the samples to prepare DNA or RNA libraries suitable for sequencing. The constructed libraries are placed into high-throughput sequencing instruments and sequenced according to the operating procedures of the instruments. After sequencing is completed, the gene sequence information obtained by sequencing is processed through bioinformatics analysis methods to calculate the expression levels of each gene, forming the gene expression data of the patients. At the same time, the ECG signals of the same patients are collected using a three-channel ECG sensor. Electrode patches are pasted on the patients' chests according to the standard ECG electrode placement positions to ensure good contact between the electrode patches and the skin, reduce signal interference, collect ECG signals for a certain period of time (such as 10 minutes), and transmit the collected gene expression data and ECG signals to a computer system through a data cable for subsequent processing.

[0039] In the gene-ECG signal correlation analysis step, joint preprocessing of adaptive noise suppression and feature enhancement is performed on the collected ECG signals. Using the designed filtering algorithm, first, the short-time Fourier transform is performed on the ECG signals to convert them into the frequency domain, analyze the spectral characteristics and energy distribution, and it is found that there is a small amount of power frequency interference and relatively obvious electromyogram (EMG) noise. For the power frequency interference, set f c (t) close to the power frequency of the local area (such as 60 Hz), and determine the appropriate B w (t) according to the actual situation of the noise. For the EMG noise, determine the corresponding f c (t) and B R (t) by analyzing its energy distribution. During the filtering process, calculate the signal-to-noise ratio (SNR) of the filtered signal to monitor the noise level in real time. If it is found that the SNR value is low, appropriately increase the value of B w (t) according to the parameter adjustment rules in the formula to further suppress the noise. At the same time, use the multi-resolution analysis method based on wavelet transform to enhance the R-wave feature. Perform wavelet transform on the ECG signals and decompose them into different scales (such as setting S = 4 scale layers) to obtain the frequency-domain representation X s (f) of the ECG signals at each scale. According to the R-wave characteristics at different scales obtained from the analysis of a large number of normal and abnormal ECG signals in the past, determine the weighting factor ω s at each scale layer. For the scale layer where the R-wave peak is located, set a relatively large ω s value, and according to the formula in The partial coefficients after decomposition are weighted to make the R wave more prominent and complete in the reconstructed signal. In the correlation analysis stage, a gene-ECG feature mapping model is constructed, and the support vector machine (SVM) is selected as the training algorithm, with the radial basis function (RBF) chosen as its kernel function. A large amount of gene expression data and corresponding ECG signal feature data of known patients undergoing cardiac rehabilitation treatment are collected as the training set. It is determined that the input features of the model include the expression level values of each gene in the preprocessed gene expression data and the relevant feature values in the ECG signal feature data (such as historical data like the amplitude, slope, and time interval between adjacent R waves of the R wave), and the output is the predicted value related to a specific feature of the R wave (such as the predicted change ratio of the R wave amplitude, the degree of morphological change, the advance or delay amount of the occurrence time, etc.). According to the formula the model is trained, where the specific form of the activation function f is determined according to the selected algorithm. During the training process, by minimizing the mean squared error loss function (N is the number of training samples, y k,i is the predicted value of the i-th training sample, is the actual observed value of the i-th training sample to adjust the weight parameter w kl and the bias term b k , so that the model is continuously optimized. The preprocessed gene expression data of the current patient is input into the trained gene-ECG feature mapping model to obtain the R wave feature prediction result based on gene expression. For example, it is predicted that the occurrence time of the R wave in the ECG signal of this patient may be delayed by a certain amount, and the amplitude of the R wave may increase by a certain proportion.

[0040] In the personalized adjustment step of the R wave detection algorithm, according to the above prediction results, the traditional R wave detection algorithm is personalized adjusted. Since it is predicted that the amplitude of the R wave may increase, the formula is used to adjust the amplitude threshold in the R wave detection algorithm. The original amplitude threshold in the traditional R wave detection algorithm is obtained by statistically analyzing the R wave amplitudes in a large number of normal ECG signal samples to get the average amplitude reference value A0 of the R wave under normal conditions. Based on the predicted change amount ΔA pred of the R wave amplitude predicted by the gene-ECG feature mapping model (assuming that it is predicted that the amplitude of the R wave will increase by 30%, then ΔA pred = 0.3A0), and setting the amplitude threshold adjustment coefficient α A = 0.4, the adjusted R wave detection amplitude threshold is calculated according to the formula. Correspondingly, the upper amplitude threshold in the R wave detection algorithm is increased. Since it is predicted that the occurrence time of the R wave may be delayed, the formula Adjust the judgment parameter of the adjacent F-wave time interval in the R-wave detection algorithm. The original judgment parameter of the adjacent R-wave time interval in the traditional R-wave detection algorithm is known. By statistically analyzing the adjacent R-wave time intervals in a large number of normal electrocardiogram signal samples, obtain the reference value T0 of the average time interval of adjacent R-waves under normal conditions. Based on the gene-electrocardiogram feature mapping model, the change amount ΔT of the R-wave occurrence time is predicted. pred (Assume that the predicted delay in the R-wave occurrence time is 0.2 seconds, then ΔT pred = 0.2T0). Let the adjustment coefficient α of the adjacent R-wave time interval judgment parameter I = 0.3. Calculate the adjusted adjacent R-wave time interval judgment parameter according to the formula.

[0041] In the adaptive fault-tolerant R-wave detection, initialize the detection parameters and related variables. Set the false detection times threshold to 2 and the missed detection times threshold to 4. According to the personalized adjusted R-wave detection algorithm, search for R-waves in the preprocessed electrocardiogram signal, identify the position, amplitude, morphology and other feature information of the R-waves, and record the results of each detection and related intermediate data, such as the spacing between adjacent R-waves, the stability index of the overall signal, etc. During the detection process, monitor in real time whether abnormal situations such as false detection and missed detection occur, and use the formula to make a judgment. Among them, is the adjacent spacing value of the currently detected adjacent R-waves, is the average value of the adjacent R-wave spacings in the historical detection data. Let θ D = 0.2, is the amplitude of the currently detected R-wave, is the amplitude of the R-wave predicted according to the gene-electrocardiogram feature mapping model. Let θ A = 0.3, is the adjacent R-wave time interval value currently detected, is the average value of the adjacent R-wave time intervals in the historical detection data. Let θ T = 0.15. If an abnormal situation is detected and the number of false detections and missed detections does not exceed the set threshold, for example, it is found that the relative deviation of the adjacent R-wave spacing from the historical average spacing exceeds θ D , taking the adjustment of the adjacent R-wave time interval judgment parameter as an example, use the formula to make an adjustment. The currently used adjacent R-wave time interval judgment parameter is known Let the adjustment coefficient β I = -0.25. Calculate the adjusted adjacent R-wave time interval judgment parameter according to the relative deviation situation of the currently detected adjacent R-wave spacing from the historical average spacing according to the formula Then use the adjusted parameter or algorithm to re-detect and try to obtain the correct result.

[0042] In the detection result output and storage step, the characteristic information such as the position, amplitude, and morphology of the detected R wave is output in the form of a visual chart through the display screen. The position information is presented with the time axis as the abscissa and the electrocardiogram signal amplitude as the ordinate. The amplitude information is marked with specific numerical values at the corresponding positions. The morphology information is presented by drawing the R wave curve and marking the key feature points. The rehabilitation treatment team (including doctors, rehabilitation therapists, etc.) can intuitively view the relevant characteristics of the R wave to accurately evaluate the changes in the cardiac electrophysiological state of the patient during cardiac rehabilitation, and then adjust the rehabilitation plan. The gene expression data, electrocardiogram signal, and R wave detection results during this detection process are stored in the cloud server and classified and marked according to information such as the detection time and patient identification for subsequent reference and further analysis. Cloud storage facilitates professionals at different locations to access the data at any time for remote collaborative diagnosis or tracking of the patient's rehabilitation progress.

[0043] In summary, in the field of personalized cardiac rehabilitation monitoring, the present invention can achieve accurate detection of the human electrocardiogram R wave based on the gene expression situation and real-time electrocardiogram signal of the patient individual, and through an adaptive adjustment and monitoring mechanism, timely capture the changes in the R wave characteristics, providing a more targeted monitoring and evaluation means for cardiac rehabilitation treatment, helping to optimize the rehabilitation plan, improve the rehabilitation effect, and ensure the recovery of the patient's heart health.

[0044] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A method for detecting the R wave of human electrocardiogram, characterized in that, The method includes the following specific steps: Gene expression data and electrocardiogram (ECG) signal acquisition: Obtain the gene expression level information of the individual's whole genome through gene detection technology. At the same time, use a multi-channel ECG sensor to collect ECG signals, and ensure that the time interval between the two acquisitions is short. Transmit the collected gene expression data and ECG signals to the computer system; Gene-ECG Signal Association Analysis: Perform joint preprocessing operations of adaptive noise suppression and feature enhancement on the collected ECG signals. Design a filtering algorithm that can automatically adjust parameters according to the real-time noise level of the ECG signals. The algorithm formula of the filtering algorithm is: where G(f,t) represents the gain function of the ECG signal after filtering at frequency f and time t, and f ∈ (t) is the reference value of the center frequency that changes dynamically with time t, used to locate the main noise frequency components, and B Ψ (t) is the bandwidth parameter that changes dynamically with time t, used to control the frequency range of filtering. n is the exponential parameter that determines the shape of the filtering curve, and its value range is 3 ≤ n ≤ 8. μ is the weight coefficient that balances the effects of noise suppression and R-wave feature enhancement, and its value range is 0 < μ < 1. S represents the scale level of wavelet transform decomposition, and ω s is the weighting factor at the s-th wavelet scale level. X s (f) is the frequency-domain representation of the ECG signal at the s-th wavelet scale level, obtained by performing wavelet transform on the ECG signal. Specifically, perform short-time Fourier transform on the collected ECG signal to convert it to the frequency domain, and synchronously analyze the energy distribution of the frequency-domain signal to preliminarily determine the type and distribution range of the noise. According to the determined noise type and its distribution situation, set initial filtering parameters for different types of noise. During the filtering process, monitor the noise level of the filtered ECG signal in real time by calculating the root mean square error of the filtered signal, and use the multi-resolution analysis method based on wavelet transform to enhance the feature information of the R wave. In the association analysis stage, construct a gene-ECG feature mapping model, and train it based on the known gene expression conditions and the corresponding ECG signal feature data. Determine that the input features of the model include the values of each gene expression level in the preprocessed gene expression data and the relevant feature values in the ECG signal feature data, and the output is the predicted value related to a specific feature of the R wave. The model training formula is: where y k represents the predicted value related to a specific feature of the R wave output by the model. σ is the activation function, used to introduce non-linearity so that the model can learn more complex relationships between gene expression and ECG signal features. w kl is the weight parameter connecting the l-th input feature and the k-th output feature. During the training process, continuously adjust the weights through the backpropagation algorithm so that the model can accurately learn the potential association relationship between gene expression and ECG signal features. b k is the bias term corresponding to the k-th output feature, and x l For each gene expression level value in the preprocessed gene expression data and the relevant feature values in the electrocardiogram signal feature data, adjustments are made during the training process. The preprocessed gene expression data of the current individual is input into the trained gene-electrocardiogram feature mapping model, so as to obtain the R-wave feature prediction result based on gene expression. Personalized adjustment of the R-wave detection algorithm: According to the R-wave feature prediction results, for the changes in the predicted R-wave amplitude, morphology, and occurrence time, respectively adjust the amplitude threshold, slope judgment parameter, and adjacent R-wave time interval judgment parameter in the detection algorithm to adapt to individual differences; Adaptive fault-tolerant R-wave detection: According to the R-wave feature prediction results obtained from the gene-ECG signal correlation analysis, conduct targeted optimization and adjustment of the R-wave detection algorithm. Specifically, for the prediction results showing changes in the R-wave amplitude, when it is predicted that the R-wave amplitude will decrease, lower the lower amplitude threshold in the R-wave detection algorithm; when it is predicted that the R-wave amplitude will increase, raise the upper amplitude threshold in the R-wave detection algorithm. For the prediction results indicating that the R-wave morphology will change, when it is predicted that the R-wave morphology becomes flatter, lower the slope threshold for judging the R-wave peak; when it is predicted that the R-wave morphology becomes sharper, raise the slope threshold. For the prediction that the occurrence time of the R-wave is advanced or delayed, when it is predicted that the occurrence time of the R-wave is advanced, shorten the upper limit of the judgment of the adjacent R-wave time interval; when it is predicted that the occurrence time of the R-wave is delayed, extend the lower limit of the judgment of the adjacent R-wave time interval; Output and storage of the detection results: Output the detected R-wave feature information for subsequent arrhythmia analysis and cardiovascular disease diagnosis. At the same time, store the gene expression data, ECG signals, and R-wave detection results during this detection process in the local hard disk and cloud server.

2. The method for detecting human electrocardiogram R wave according to claim 1, characterized in that In the step of gene expression data and ECG signal acquisition, obtain the gene expression level information of the individual's whole genome through gene detection technology. The gene detection technology includes gene chip technology and high-throughput sequencing technology. When using gene chip technology, process the individual's blood sample or tissue sample and then perform a hybridization reaction with the gene chip, and detect the fluorescence signal intensity of each gene locus on the chip to obtain the gene expression level information. When using high-throughput sequencing technology, perform nucleic acid extraction and library construction operations on the individual's blood sample or tissue sample and then put it into a high-throughput sequencing instrument for sequencing. Calculate the expression level of each gene through bioinformatics analysis methods for the gene sequence information obtained by sequencing to form gene expression data.

3. A method for detecting the R wave of human electrocardiogram according to claim 1, characterized in that, In the step of gene expression data and ECG signal acquisition, use a multi-channel ECG sensor to collect ECG signals. Select a three-channel or twelve-channel ECG sensor, and determine the number of channels according to the detection requirements and accuracy requirements. Paste the electrode pads of the ECG sensor at the positions of the ECG electrodes on the individual's chest. The ECG sensor receives the weak electrical signals generated by the heart activities through the electrode pads and converts them into a digital signal form that can be processed by the computer system. During the acquisition process, ensure that the time interval between the gene expression data and ECG signal acquisitions is short to maintain the relevance of the two sets of data.

4. A method for detecting the R wave of human electrocardiogram according to claim 1, characterized in that, In the above-mentioned R-wave detection step with adaptive fault tolerance, according to the R-wave feature prediction result obtained from the gene-ECG signal correlation analysis, the R-wave detection algorithm is adjusted specifically. Specifically, for the prediction result indicating that the amplitude of the R-wave changes, the formula is used to adjust the amplitude threshold in the R-wave detection algorithm, where is the adjusted amplitude threshold, is the original amplitude threshold, and α A is the preset amplitude threshold adjustment coefficient, and ΔA pred is the R-wave amplitude change amount predicted based on the gene-ECG feature mapping model. For the prediction result indicating that the morphology of the R-wave will change, the formula is used to adjust the slope judgment parameter in the R-wave detection algorithm, where is the adjusted slope threshold, is the original slope threshold, and α S is the preset slope judgment parameter adjustment coefficient, and ΔS pred is the slope change amount corresponding to the R-wave morphology change predicted based on the gene-ECG feature mapping model. For the prediction that the occurrence time of the R-wave is advanced or delayed, the formula is used to adjust the judgment parameter regarding the time interval between adjacent R-waves in the R-wave detection algorithm, where is the adjusted judgment parameter for the time interval between adjacent R-waves, is the original judgment parameter for the time interval between adjacent R-waves, and α I ×ΔT pred is the preset judgment parameter adjustment coefficient for the time interval between adjacent R-waves, and ΔT pred is the R-wave occurrence time change amount predicted based on the gene-ECG feature mapping model.

5. A method for detecting the R wave of human electrocardiogram according to claim 1, characterized in that, In the above-mentioned self-adaptive fault-tolerant R-wave detection step, it is judged whether there are false detections and missed detections by comparing with historical data and using a formula: where, is the currently detected adjacent R-wave interval value, is the average value of adjacent R-wave intervals in historical detection data, which is obtained by averaging the adjacent R-wave interval values obtained from multiple previous detections, and θ D is the adjacent R-wave interval deviation threshold, and the value range is 0.15 ≤ θ D ≤ 0.35, is the currently detected R-wave amplitude, is the R-wave amplitude predicted according to the gene-ECG feature mapping model, and θ A is the R-wave amplitude deviation threshold, and the value range is 0.25 ≤ θ A ≤ 0.45, is the currently detected adjacent R-wave time interval value, is the average value of adjacent R-wave time intervals in historical detection data, and θ T is the adjacent R-wave time interval deviation threshold, and the value range is 0.1 ≤ θ T ≤ 0.

3.

6. A method for detecting human electrocardiogram R wave according to claim 1, characterized in that, In the step of adaptively fault-tolerant R-wave detection, whether false detection and missed detection anomalies occur is monitored by comparing with historical data. If the threshold is not exceeded, the parameters or algorithms are adjusted in real time according to the characteristics of the current electrocardiogram signal and historical data and redetection is performed. The adjustment formula for the time interval is as follows: where is the parameter for judging the time interval between adjacent R-waves currently in use, is the detected distance between adjacent R-waves and the average value of the distances between adjacent R-waves in the historical detection data of the deviation situation, β I is the adjustment coefficient, and the value range is -0.3 ≤ β I ≤ 0.

3. The adjusted parameter for judging the time interval between adjacent R-waves is calculated according to the formula For other detection parameters, including amplitude thresholds and filtering parameters, when adjusting, the adjusted parameter values are determined according to the deviation situation between the current characteristics and historical data and the set adjustment coefficient, and redetection is performed using the adjusted parameters or algorithms to obtain the correct results.

Images