Electrocardiosignal prediction method, system and equipment for acute heart failure patient and medium

Abstract

The invention discloses an electrocardiosignal prediction method, system and device for an acute heart failure patient and a medium, and the method comprises the steps: obtaining a historical electrocardiosignal of the acute heart failure patient, generating a multi-modal signal comprising a feature enhanced electrocardiosignal, a dynamic feature and a power spectrum density, and constructing an acute heart failure sample data set; obtaining a pre-trained ECG signal prediction model, and training the model by using the acute heart failure sample data set based on transfer learning to obtain an acute heart failure ECG signal prediction model; the method comprises the following steps: acquiring an electrocardiosignal of a patient with acute heart failure in real time and preprocessing the electrocardiosignal to obtain a real-time multi-mode signal, and taking the real-time multi-mode signal as an input to predict an ECG signal within a future preset duration; by extracting a multi-modal signal as model input, the prediction precision and robustness of the ECG signal are remarkably improved; based on transfer learning, efficient and accurate ECG signal prediction is realized in an acute heart failure scene.

Description

Technical Field

[0001] The present invention relates to the technical field of cardiac motion prediction, and in particular to a method, system, device and medium for predicting electrocardiogram signals of patients with acute heart failure. Background Art

[0002] Acute heart failure is a rapidly developing and highly dangerous cardiovascular emergency, usually triggered by events such as myocardial infarction, myocarditis or severe arrhythmia. The patient's heart pumping function drops suddenly and cannot meet the body's metabolic needs, which can easily lead to complications such as acute pulmonary edema, hypoperfusion shock and multiple organ failure, which directly threaten life.

[0003] Mechanical circulatory support (MCS) plays an important role in the treatment of patients with acute heart failure. It can partially or completely replace the heart's pumping function and provide effective support for improving hemodynamic status and maintaining organ perfusion. However, the current MCS technology is mostly used in patients with chronic heart failure, especially the pulsation pattern based on ECG-R peak prediction, which significantly optimizes hemodynamic performance by synchronizing with the patient's natural heart beat. This technology usually relies on personalized modeling of a single patient, requiring long-term collection of patient physiological data and complex model training.

[0004] For patients with acute heart failure, this personalized modeling has significant limitations. Since acute attacks often occur without warning, there is not enough time to complete modeling and training. At the same time, the patient's physiological parameters may change rapidly with the disease, increasing the difficulty of real-time synchronization. Therefore, achieving accurate ECG signal prediction in acute situations requires a technology with high robustness and rapid adaptability to meet the physiological characteristics of different patients. Summary of the invention

[0005] The present invention provides a method, system, device and medium for predicting electrocardiogram signals of patients with acute heart failure, so as to solve the problem that it is difficult to accurately predict electrocardiogram signals of patients with acute heart failure in a timely manner.

[0006] In a first aspect, a method for predicting an electrocardiogram signal of a patient with acute heart failure is provided, comprising the following steps: S1: Obtain historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, and then construct an acute heart failure sample dataset; S2: Obtain a pre-trained ECG signal prediction model, the ECG signal prediction model input is a multimodal signal, and the output is an ECG signal within a preset time period in the future; based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using the acute heart failure sample data set to obtain an acute heart failure ECG signal prediction model; S3: Acquire the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

[0007] Furthermore, in step S1, the preprocessing process includes: S1.1: Using the adaptive noise cancellation method, the ECG signal is denoised by the least mean square adaptive filter; S1.2: resampling the denoised ECG signal to the target sampling rate, and then obtaining the feature enhanced ECG signal through wavelet transform; S1.3: Extract dynamic features based on the time change rate of the feature-enhanced ECG signal to obtain dynamic features; S1.4: Use fast Fourier transform to calculate the frequency domain features of the feature-enhanced ECG signal and obtain the power spectral density.

[0008] Further, in step S2, the ECG signal prediction model includes a multimodal signal feature embedding module, a parallel multi-scale convolution module, an attention module and a linear output layer; The multimodal signal feature embedding module is used to embed the multimodal signal; the parallel multi-scale convolution module is used to perform implicit ECG segmentation on the embedded multimodal signal and extract its multi-scale physiological features; the attention module is used to perform weighted fusion on the multi-scale physiological features, and the linear output layer outputs the ECG signal within a preset time length in the future.

[0009] Furthermore, the multimodal signal feature embedding module uses a one-dimensional convolutional layer and two linear layers to map the feature-enhanced ECG signal, dynamic features, and power spectral density to a higher-dimensional latent space, sets the embedding dimension, aligns the multimodal signals, and then merges them; The parallel multi-scale convolution module includes three independent convolution branches, which respectively extract features at three different time scales, corresponding to capturing the short-term features of P wave and QRS wave, the mid-term features of T wave and PR interval, and the long-term features of QT interval; each convolution branch includes a first convolution layer, a pooling layer, and a second convolution layer connected in sequence; The attention module performs global average pooling on the feature map output by each convolution branch to obtain a global feature vector for each convolution branch; the global feature vector of each convolution branch is input into two fully connected neural networks to generate the weight corresponding to each convolution branch, and then the feature maps of the three convolution branches are weightedly fused to obtain a fused feature.

[0010] Furthermore, in step S2, when the pre-trained ECG signal prediction model is trained using the acute heart failure sample dataset, the parameters of the multimodal signal feature embedding module and the parallel multi-scale convolution module are frozen, and the parameters of the attention module and the linear output layer are adjusted using the acute heart failure sample dataset.

[0011] Furthermore, the ECG signal data of patients with acute heart failure is continuously acquired, and the acquired data is used to continue online training of the ECG signal prediction model for acute heart failure.

[0012] Furthermore, it also includes: The predicted ECG signal of the acute heart failure patient within a preset time period in the future is input into an R peak detector to detect the moment when the future R peak appears.

[0013] In a second aspect, a system for predicting an electrocardiogram signal of a patient with acute heart failure is provided, comprising the following steps: The data processing module is used to obtain the historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, thereby constructing an acute heart failure sample data set; A transfer learning module is used to obtain a pre-trained ECG signal prediction model, the input of the ECG signal prediction model is a multimodal signal, and the output is an ECG signal within a preset time period in the future; based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using an acute heart failure sample data set to obtain an acute heart failure ECG signal prediction model; The signal prediction module is used to obtain the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

[0014] In a third aspect, an electronic device is provided, including: Memory on which computer programs or instructions are stored; A processor is used to load and execute the computer program or instructions to implement the electrocardiogram signal prediction method for patients with acute heart failure as described above.

[0015] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program or instruction is stored, including: when the computer program or instruction is executed by a processor, the method for predicting electrocardiogram signals of patients with acute heart failure as described above is implemented.

[0016] The present invention proposes a method, system, device and medium for predicting electrocardiogram signals of patients with acute heart failure, which has the following beneficial effects: (1) The present invention extracts multimodal signals including feature-enhanced ECG signals, dynamic features and power spectral density as model input, which has more dimensional information and enhances the model's ability to respond to signal changes. Feature-enhanced ECG signals ensure high-quality signal input, and dynamic features and power spectral density further improve the distinguishability and comparability of features, providing the model with more comprehensive feature input and significantly improving the prediction accuracy and robustness of ECG signals. (2) In the absence of a large amount of personalized data, the present invention can, based on transfer learning and multimodal signal processing, enable the model to quickly adapt to the physiological changes of patients with acute heart failure, ensure the accuracy and real-time nature of the prediction results, avoid the problems of complex modeling and long-term training in traditional methods, and achieve efficient and accurate ECG signal prediction in acute heart failure scenarios, thereby providing reliable support for the pulsation pattern of mechanical circulatory support devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 This is a flow chart of a method for predicting electrocardiogram signals of patients with acute heart failure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0020] In the treatment of acute heart failure, since acute attacks often occur without warning and the patient's physiological parameters change rapidly, the existing ECG-R peak prediction method (usually used for chronic heart failure) is difficult to cope with these situations. The existing technology often relies on personalized modeling of long-term data, which has the problem of difficulty in real-time prediction. Based on this, the present invention proposes an ECG signal (electrocardiogram signal) prediction scheme that can quickly adapt to the physiological characteristics of patients with acute heart failure. In the absence of a large amount of personalized data, based on transfer learning and multimodal signal processing, it can achieve efficient and accurate ECG signal prediction in acute heart failure scenarios, thereby providing reliable support for the pulsation mode of mechanical circulatory support devices.

[0021] like Figure 1 As shown, this embodiment provides a method for predicting an ECG signal of a patient with acute heart failure, comprising the following steps: S1: Obtain historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, and then construct an acute heart failure sample data set; the preprocessing process includes denoising, interpolation, dynamic feature extraction or frequency domain feature extraction; S2: Obtain a pre-trained ECG signal prediction model, the ECG signal prediction model input is a multimodal signal, and the output is an ECG signal within a preset time period in the future; based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using the acute heart failure sample data set to obtain an acute heart failure ECG signal prediction model; S3: Acquire the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

[0022] The ECG signal prediction method for patients with acute heart failure provided in the above embodiment has the following advantages: (1) Multimodal signal processing: The present invention extracts multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density as model input, which has more dimensional information and enhances the model's ability to respond to signal changes. Feature-enhanced ECG signals ensure high-quality signal input, and dynamic features and power spectral density further improve the distinguishability and comparability of features, providing the model with more comprehensive feature input, significantly improving the prediction accuracy and robustness of ECG signals. (2) Rapid adaptation and optimization of acute scenarios based on transfer learning: A key innovation of the present invention is the ability to adapt to acute scenarios based on transfer learning. Transfer learning allows the model to quickly adapt to the real-time data of patients with acute heart failure after being pre-trained with a large-scale ECG signal data set. Therefore, in the absence of a large amount of personalized data, the present invention can enable the model to quickly adapt to the physiological changes of patients with acute heart failure based on transfer learning and multimodal signal processing, ensuring the accuracy and real-time nature of the prediction results, avoiding the problems of complex modeling and long-term training in traditional methods, and achieving efficient and accurate ECG signal prediction in acute heart failure scenarios, thereby providing reliable support for the pulsation pattern of mechanical circulatory support devices.

[0023] In some embodiments, in step S1, the preprocessing process includes: S1.1: Use the adaptive noise cancellation method to remove the noise (such as myoelectric noise, power supply interference and baseline drift) in the ECG signal through the least mean square adaptive filter; the specific process is as follows: The reference noise signal is recorded as r(t), where t represents the discrete time index and the filter order is set to M. The noise estimate output by the adaptive filter at time t can be expressed as: ; In the formula, represents the noise estimate; Represents the mth coefficient of the filter, which is updated as follows: ; in, Represents the learning rate, which is used to control the magnitude of each iteration coefficient update; is the error signal, which is determined by the following formula: ; in, Represents the main channel ECG signal (containing useful components and noise). When the filter gradually converges to the noise estimation, the estimated noise can be filtered out from the ECG signal: ; In the formula, Represents the denoised ECG signal.

[0024] S1.2: resampling the denoised ECG signal to the target sampling rate, and then obtaining the feature-enhanced ECG signal through wavelet transform; specifically including: For different acquisition devices, the original ECG signal sampling rate , uniformly resample the collected signals to the target sampling rate , while preserving the global and local characteristics of the signal; Timeline of the original ECG signal for: ; Feature Enhanced ECG Signal Timeline for: ; Where N is the number of sampling points of the original ECG signal, L is the number of sampling points of the feature enhanced ECG signal, is the original ECG signal sampling rate, Target sampling rate of ECG signal for feature enhancement; By interpolation ( ) or extract ( ) method, the denoised ECG signal Corresponding to the time axis The resampled signal is obtained , and then perform discrete wavelet decomposition, expressed as follows: ; In the formula, is the number of decomposition levels, is the current level of decomposition, is the translation index of the wavelet singular function, No. The detail factor at the layer scale, For the The approximate coefficient at the layer scale, is the scale function, is the wavelet function; Finally, the approximate coefficients and detail coefficients obtained by discrete wavelet decomposition are used to generate feature-enhanced ECG signals through the following mapping relationship: : ; In the formula, The coefficients are fitted with a function, which may be IDWT or other predetermined rules, for mapping the approximation coefficients combined with the detail coefficients back to the time domain.

[0025] The method of resampling to the target sampling rate is resampling or interpolation, which is useful for data sets with different sampling frequencies. The resampled signal is then input into the wavelet transform to remove useless information and highlight key ECG features. It should be noted that the number of retained layers in the wavelet transform process must be less than the number of decomposition layers. For example, the number of decomposition layers in the wavelet transform is 7, and the number of retained layers is 1-6. The number of retained layers is not fixed and can be selected as needed, but it must be less than the number of decomposition layers.

[0026] S1.3: Extract dynamic features based on the time change rate of the feature-enhanced ECG signal to obtain dynamic features; specifically including: Computational features to enhance ECG signals The time rate of change of , extract the first-order difference features: ; In the formula, is the first-order difference of the time rate of change; Then the first-order difference results are standardized to highlight the comparability of features: ; In the formula, is the first-order difference of the normalized time-varying rate; exist A copy padding value is added at the end of to align the feature enhanced ECG signal length.

[0027] S1.4: Use fast Fourier transform to calculate the frequency domain features of the feature-enhanced ECG signal and obtain the power spectrum density. The process is shown as follows: ; In the formula, represents the power spectral density, stands for Fast Fourier Transform.

[0028] S1.5: Sample construction: The three feature quantities of feature-enhanced ECG signals, dynamic features, and power spectral density are integrated according to the category dimension, and the sliding window is divided to obtain a number of samples with multimodal signals as input and ECG signals of a preset future duration as output, thereby constructing an acute heart failure sample data set.

[0029] The signal is reconstructed through adaptive noise cancellation and wavelet transform to ensure high-quality signal input; the extraction of dynamic features and frequency domain features further improves the distinctiveness and comparability of features, providing more comprehensive feature input for subsequent models. This innovation significantly improves the prediction accuracy and robustness of ECG signals.

[0030] In some embodiments, the ECG signal prediction model can select a basic model, including a multimodal signal feature embedding module, a convolution block and a linear output layer. However, in order to further improve the prediction accuracy of the model, in some preferred embodiments, the basic model is further improved, and the ECG signal prediction model includes a multimodal signal feature embedding module, a parallel multiscale convolution module, an attention module and a linear output layer; the multimodal signal feature embedding module is used to embed the multimodal signal; the parallel multiscale convolution module is used to perform implicit ECG segmentation on the embedded multimodal signal and extract its multiscale physiological features; the attention module is used to perform weighted fusion on the multiscale physiological features, and the linear output layer outputs the ECG signal within a preset time in the future.

[0031] Specifically, the multimodal signal feature embedding module uses a one-dimensional convolutional layer and two linear layers to respectively embed the feature enhanced ECG signal , Dynamic Features and power spectral density Mapping to a higher-dimensional latent space to achieve more adequate feature expression, the process is expressed as follows: ; In the formula, , , They represent the embedded feature-enhanced ECG signal, dynamic features, and power spectral density, respectively; Conv1D represents a one-dimensional convolutional layer, and FC represents a linear layer; The embedding dimension is usually set to D. The multimodal signals are aligned and merged through embedding, which can be expressed as follows: ; In the formula, Represents the merged multimodal signal.

[0032] In order to fully capture the temporal features of different physiological time scales in the ECG signal, the parallel multi-scale convolution module includes three independent convolution branches, named Conv Block 1, Conv Block 2 and Conv Block 3. The three independent convolution branches perform feature extraction at three different time scales, that is, feature extraction is performed for the information in the short time window, medium time window and long time window, respectively, corresponding to capturing the short-term features of the P wave and QRS wave, the medium-term features of the T wave and PR interval, and the long-term features of the QT interval. Each convolution branch includes a first convolution layer, a pooling layer and a second convolution layer connected in sequence, which gradually expands the receptive field of the convolution. Specifically, the structure of the convolution branch is as follows: ; in, and Represent the convolution kernel sizes of the first and second convolution layers respectively, Indicates the pooling window size.

[0033] Short-term dependency (corresponding to Conv Block 1): aims to capture the timing dependency of about 0.05 s, mainly used to extract short-term features in the process of cardiac depolarization such as P wave and QRS wave; Mid-time dependency (corresponding to Conv Block 2): The goal is to capture the timing dependency of about 0.12 s, which corresponds to mid-term features of the repolarization process such as atrioventricular node conduction and T waves; Long-term dependency (corresponding to Conv Block 3): The target is set to a timing dependency of about 0.20 s, which is used to extract long-term features such as the QT interval from the end of cardiac depolarization to repolarization.

[0034] Convolution branch input data: Segment the signal using a sliding window strategy The size of each window is , the step length is .

[0035] Sampling time calibration: If Each data corresponds to t seconds, so the time represented by each data is .

[0036] Convolution branch parameter description: The first convolution layer The convolution kernel size is (3, k 1 ), stride 1, same padding (in the convolution operation, zero padding is added to the edge of the input to make the output size after the convolution operation consistent with the input size), the number of output channels is ; The pooling layer uses maximum pooling, and the pooling window size and step length are , shortening the time dimension to ; The second convolutional layer The convolution kernel size is (3, k 2 ), the stride is 1, the same padding is used, and the number of output channels is .

[0037] The receptive field of each convolution branch is: ; In the formula, Represents the receptive field of the convolution branch; corresponding to the time scale: ; In the formula, Represents the receptive field at the time scale of the convolution branch; In order to ensure that short-term, medium-term, and long-term dependencies can be accurately extracted, the size relationship of the convolution kernel needs to satisfy the following formula: ; In the following embodiments, B=250 and t=0.5s are taken as an example for description. The corresponding three convolution branches take 0.05s, 0.12s, and 0.20s respectively; k in Conv Block 1 1 , p, k 2 are 5, 7, and 3 respectively; k in Conv Block 2 1 , p, k 2 are 7, 9, and 6 respectively; k in Conv Block 3 1 , p, k 2 They are 11, 15 and 6 respectively. The output feature map of each convolution branch is .

[0038] The attention module performs global average pooling on the feature map output by each convolution branch, extracts global features, and calculates the mean in the time dimension, thereby compressing the feature map of each channel into a scalar, which is expressed as follows: ; In the formula, is the feature map of the i-th convolution branch, is the global feature vector of the i-th convolution branch. After completing the global average pooling, The input is shared by two layers of fully connected neural network ( , the number of neurons are and ) to generate a dynamic weight vector. The specific formula is as follows: ; In the formula, Represents the weight corresponding to the i-th convolution branch; The weights automatically obtained by training are then weighted fusion of features: ; In the formula, Represents fusion features.

[0039] The linear output layer performs single channel dimensionality reduction to achieve prediction output : ; In the formula, , is the weight matrix of the linear output layer, T is the length to be predicted, that is, the preset time in the future; b is the bias term.

[0040] The above embodiment uses a parallel multi-scale convolution module to extract various timing features in the ECG signal through three different convolution paths (short-time, medium-time, and long-time dependent features). This design innovation solves the complexity problem of ECG signals at different time scales. For example, the short-time dependent features mainly capture rapidly changing ECG waveforms such as P waves and QRS waves, the medium-time dependent features focus on the features of longer time windows such as PR intervals and T waves, and the long-time dependent features focus on long-term changes in cardiac repolarization such as the QT interval. Through this innovative design, the model can accurately capture the characteristics of ECG changes at different time scales, improving the prediction accuracy of the model.

[0041] In addition, a multi-scale feature fusion method based on the attention mechanism is used to extract the global features of each convolution branch through global average pooling, and then the attention mechanism is used to weightedly fuse the features of different scales. This method effectively solves the problem of how to mine the most important features in multi-scale information, and improves the model's adaptability and prediction accuracy to complex signals. This innovation further enhances the model's adaptability to data from patients with acute heart failure through feature weighting and fusion.

[0042] In some embodiments, when a pre-trained ECG signal prediction model is trained using an acute heart failure sample dataset, the parameters of the multimodal signal feature embedding module and the parallel multi-scale convolution module are frozen, and the parameters of the attention module and the linear output layer are fine-tuned using the acute heart failure sample dataset.

[0043] In some embodiments, the following steps are also included: S4: inputting the predicted ECG signal of the acute heart failure patient within a preset time period in the future into the R peak detector to detect the time when the future R peak will appear; and then the pulsation of the mechanical circulatory support device can be modulated according to the time when the future R peak will appear.

[0044] The entire implementation process can be further described as follows: (1) Model initialization and pre-training.

[0045] On a large-scale historical ECG signal dataset, the ECG signal prediction model is fully pre-trained to lay the foundation for subsequent rapid adaptation. The specific process is as follows: Dataset construction: A large dataset of ECG signals from public healthy individuals and patients with chronic heart failure is used. Data enhancement techniques (such as adding random noise, signal translation and scaling, etc.) are used to simulate signal interference and individual differences in actual scenarios, improve data diversity and the generalization ability of the model, and use the method in step S1 to construct the dataset.

[0046] Feature learning process: The above data set is input into the ECG signal prediction model for model pre-training. In the pre-training stage, the model first extracts time series features of different time scales through parallel multi-scale convolution modules: short-term features focus on the rapidly changing signal structure (such as P wave, QRS wave) during cardiac depolarization, medium-term features (such as PR interval, T wave) focus on the medium-term regularity of cardiac atrioventricular node conduction and repolarization, and long-term features (such as QT interval) focus on the long-term signal changes of the heart from depolarization to the end of repolarization. Finally, the attention layer is used to perform weighted fusion of these multi-scale features to capture the long-term temporal dependencies and cross-band feature interactions in the signal, and further optimize the feature expression capabilities to cope with the diversity of complex ECG waveforms.

[0047] Frozen convolution module: After pre-training is completed, the parameters of the multimodal signal feature embedding module and the parallel multi-scale convolution module are frozen to ensure the stability of feature extraction and the universal application of ECG waveforms in the transfer learning process, providing a high-quality feature foundation for subsequent real-time adaptation and transfer learning.

[0048] (2) Real-time adaptation and fine-tuning for acute scenarios.

[0049] In the treatment of acute heart failure, the model needs to adapt to real-time patient data based on transfer learning. The specific process is as follows: Real-time signal preprocessing: The real-time ECG signal of the patient with acute heart failure is processed according to the method of step S1 to perform denoising, interpolation, dynamic feature extraction and frequency domain feature calculation to generate a multimodal signal and provide complete feature information for subsequent processing.

[0050] Feature transfer and fine-tuning: The pre-processed real-time multimodal signal is input into the ECG signal prediction model for training, and features are extracted through the frozen parallel multi-scale convolution module and input into the attention module for feature weighting and adjustment. The attention module dynamically fine-tunes the weight of the feature according to the real-time signal characteristics of acute patients, so as to quickly adapt to individual physiological changes and ensure that the model can capture the complex changes and instantaneous fluctuations in the ECG signals of patients with acute heart failure in real time.

[0051] Fine-tuning the linear output layer: After fine-tuning the attention module, the generated features are input into the linear output layer. By fine-tuning the linear output layer, the model can make accurate predictions based on the physiological characteristics of acute patients and provide high-quality output for subsequent prediction tasks.

[0052] (3) Prediction and continuous adaptation.

[0053] After completing real-time adaptation and fine-tuning, the model enters the prediction phase and optimizes the prediction results through continuous adaptation. The specific process includes: Prediction task: For an input ECG signal, use the real-time fine-tuned model to accurately output a signal of a preset duration in the future, providing support for treatment decisions for patients with acute heart failure.

[0054] Continuous adaptation: As more real-time data is input, the model continues to be trained and fine-tuned online to adapt to the dynamic changes in the patient's physiological state, ensuring continuous improvement in prediction accuracy. Through this continuous adaptability, the model can cope with the physiological changes and uncertainties of patients with acute heart failure during the treatment process and maintain high-precision prediction results.

[0055] The embodiment of the present invention also provides an ECG signal prediction system for patients with acute heart failure, comprising the following steps: The data processing module is used to obtain the historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, thereby constructing an acute heart failure sample data set; A transfer learning module is used to obtain a pre-trained ECG signal prediction model, the input of the ECG signal prediction model is a multimodal signal, and the output is an ECG signal within a preset time period in the future; based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using an acute heart failure sample data set to obtain an acute heart failure ECG signal prediction model; The signal prediction module is used to obtain the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

[0056] It should be understood that the functional unit modules in various embodiments of the present invention may be concentrated in one processing unit, or each unit module may exist physically separately, or two or more unit modules may be integrated in one unit module, and may be implemented in the form of hardware or software.

[0057] An embodiment of the present invention further provides an electronic device, including: Memory on which computer programs or instructions are stored; A processor is used to load and execute the computer program or instructions to implement the electrocardiogram signal prediction method for patients with acute heart failure as described above.

[0058] An embodiment of the present invention further provides a computer-readable storage medium having a computer program or instruction stored thereon, including: when the computer program or instruction is executed by a processor, the method for predicting electrocardiogram signals of patients with acute heart failure as described above is implemented.

[0059] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0060] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0062] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0064] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for predicting electrocardiogram signals of patients with acute heart failure, characterized in that: The steps include: S1: Obtain historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, and then construct an acute heart failure sample dataset; S2: Obtain a pre-trained ECG signal prediction model, the ECG signal prediction model input is a multimodal signal, and the output is an ECG signal within a preset time period in the future; based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using the acute heart failure sample data set to obtain an acute heart failure ECG signal prediction model; S3: Acquire the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

2. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 1, characterized in that: In step S1, the preprocessing process includes: S1.1: Using the adaptive noise cancellation method, the ECG signal is denoised by the least mean square adaptive filter; S1.2: resample the denoised ECG signal to the target sampling rate, and combine it with wavelet transform to obtain the feature enhanced ECG signal; S1.3: Extract dynamic features based on the time change rate of the feature-enhanced ECG signal to obtain dynamic features; S1.4: Use fast Fourier transform to calculate the frequency domain features of the feature-enhanced ECG signal and obtain the power spectral density.

3. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 1 or 2, characterized in that: In step S2, the ECG signal prediction model includes a multimodal signal feature embedding module, a parallel multi-scale convolution module, an attention module and a linear output layer; The multimodal signal feature embedding module is used to embed the multimodal signal; the parallel multiscale convolution module is used to perform implicit ECG segmentation on the embedded multimodal signal to extract its multiscale physiological features; The attention module is used to perform weighted fusion on multi-scale physiological features, and the linear output layer outputs ECG signals within a future preset time length.

4. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 3, characterized in that: The multimodal signal feature embedding module uses a one-dimensional convolutional layer and two linear layers to map the feature-enhanced ECG signal, dynamic features, and power spectral density to a higher-dimensional latent space, sets the embedding dimension to align the multimodal signals, and then merges them; The parallel multi-scale convolution module includes three independent convolution branches, which respectively extract features at three different time scales, corresponding to capturing the short-term features of P wave and QRS wave, the mid-term features of T wave and PR interval, and the long-term features of QT interval; each convolution branch includes a first convolution layer, a pooling layer, and a second convolution layer connected in sequence; The attention module performs global average pooling on the feature map output by each convolution branch to obtain a global feature vector for each convolution branch; the global feature vector of each convolution branch is input into two fully connected neural networks to generate the weight corresponding to each convolution branch, and then the feature maps of the three convolution branches are weightedly fused to obtain a fused feature.

5. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 3, characterized in that: In step S2, when the pre-trained ECG signal prediction model is trained using the acute heart failure sample dataset, the parameters of the multimodal signal feature embedding module and the parallel multi-scale convolution module are frozen, and the parameters of the attention module and the linear output layer are adjusted using the acute heart failure sample dataset.

6. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 1, characterized in that: The ECG signal data of patients with acute heart failure are continuously acquired, and the acquired data are used to continue online training of the ECG signal prediction model for acute heart failure.

7. The method for predicting electrocardiogram signals of patients with acute heart failure according to claim 1, characterized in that: Also includes: The predicted ECG signal of the acute heart failure patient within a preset time period in the future is input into an R peak detector to detect the moment when the future R peak appears.

8. An electrocardiogram signal prediction system for patients with acute heart failure, characterized in that: The steps include: The data processing module is used to obtain the historical ECG signals of patients with acute heart failure and perform preprocessing to generate multimodal signals including feature-enhanced ECG signals, dynamic features, and power spectral density, thereby constructing an acute heart failure sample data set; A transfer learning module is used to obtain a pre-trained ECG signal prediction model, the ECG signal prediction model input is a multimodal signal, and the output is an ECG signal within a preset time period in the future; Based on the transfer learning technology, the pre-trained ECG signal prediction model is trained using the acute heart failure sample data set to obtain the acute heart failure ECG signal prediction model; The signal prediction module is used to obtain the ECG signals of patients with acute heart failure in real time and perform preprocessing to obtain real-time multimodal signals, and input them into the acute heart failure ECG signal prediction model to predict the ECG signals of patients with acute heart failure within a preset time period in the future.

9. An electronic device, characterized in that: include: Memory on which computer programs or instructions are stored; A processor, configured to load and execute the computer program or instructions to implement the method for predicting electrocardiogram signals for patients with acute heart failure as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: include: When the computer program or instruction is executed by a processor, the method for predicting electrocardiogram signals for patients with acute heart failure as described in any one of claims 1 to 7 is implemented.

Images