Abnormal recognition system and method for heart sound signals of children with heart disease

Abstract

The present invention discloses a system and method for identifying abnormalities in pediatric heart sound signals. The system relates to the field of abnormality identification data processing technology and includes a pediatric heart sound signal data acquisition and processing module, a pediatric heart sound signal data analysis module, a comprehensive analysis module, and an optimization and adjustment module. The system uses a sensor to collect pediatric heart sound signals, pre-processes them, evaluates their synchronization and quality, and comprehensively analyzes them in combination with artifact detection rates. This optimizes the process for identifying abnormalities in pediatric heart sound signals, thereby improving the accuracy of the system for identifying abnormalities in pediatric heart sound signals. This addresses the problem of insufficient accuracy in existing pediatric heart sound signal anomaly identification systems.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormality recognition data processing, and in particular to an abnormality recognition system and method for heart sound signals of children with heart disease. Background Art

[0002] With advances in medical technology and improved living standards, the proportion of children in the population has increased in some countries and regions, and children's health issues have received greater attention. However, the imbalance of medical resources in different regions has led to differences in the diagnosis and treatment of childhood heart disease. By identifying abnormalities in heart sound signals, heart disease can be detected early and treatment outcomes can be improved. Early identification helps preventive measures and reduce the morbidity and mortality of heart disease. Accurate identification of heart sound signals helps rationally allocate medical resources and improve the efficiency of medical services. The use of traditional stethoscopes remains an important means of identifying heart sound signals. The development of modern electronic stethoscopes and electrocardiograms has improved the accuracy of heart sound signal analysis. The application of artificial intelligence and machine learning technologies in heart sound signal analysis can help improve the accuracy and efficiency of diagnosis.

[0003] Existing abnormality recognition systems for heart sound signals of children with heart disease convert heart sounds into digital signals using an electronic stethoscope or electrocardiograph; remove noise and improve the signal-to-noise ratio through low-pass, high-pass, band-pass and notch filters; and classify the extracted features using machine learning algorithms to identify normal and abnormal heart sound signals.

[0004] For example, a method for detecting abnormal heart sound signals disclosed in a patent application with publication number CN118468184A includes: a. unifying the sampling rate of heart sound signal samples, using linear interpolation to expand the sample size of abnormal heart sound signals, and storing the data path and label correspondence in a CSV file to construct a deep learning data set; b. initializing model parameters, defining a FocalLoss loss function to measure the model classification performance; c. adding discrete convolution mapping and convolution embedding modules to the basic neural network model to extract frequency domain features of heart sound signals, thereby improving the accuracy of the model in detecting abnormal heart sounds; d. using a Butterworth filter to reduce the noise of the data set in step a, and extracting Mel spectrogram features; e. training the extracted features in the heart sound classification model, and adjusting the model parameters according to the validation set to obtain the optimal model; f. using the optimal model obtained in step e to predict the heart sound signal, and visualize the prediction results.

[0005] For example, the abnormal heart sound recognition method and device based on a multi-scale attention neural network, as disclosed in the patent application publication number CN112036467B, includes: preprocessing the collected original heart sound signals and using the preprocessed heart sound signals as training samples; labeling the training samples for heart sound quality; training an abnormal heart sound recognition model based on the training samples and their labeled content; inputting the heart sound data to be detected into the trained abnormal heart sound recognition model to obtain a heart sound quality prediction result, and identifying abnormal heart sounds based on the heart sound quality prediction result.

[0006] However, in the process of implementing the technical solutions of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] When the system collects data synchronously, the timestamps of different sensors may be misaligned due to delays such as differences in signal transmission paths, resulting in insufficient synchronization assessment. Children's activities may lead to poor sensor contact, such as gaps between flexible electrodes and the skin, causing signal baseline drift or instantaneous dropout. Motion artifacts such as muscle vibration noise and systolic murmur partially overlap in the frequency domain, making them difficult to separate effectively. This leads to insufficient accuracy of the abnormal recognition system for heart sound signals of children with heart disease. Summary of the Invention

[0008] The present invention solves the problem of insufficient accuracy of abnormality recognition systems for heart sound signals of children with heart disease in the prior art by providing an abnormality recognition system and method for heart sound signals of children with heart disease, thereby improving the accuracy of abnormality recognition systems for heart sound signals of children with heart disease.

[0009] The present invention provides a system and method for identifying abnormalities in heart sound signals of children with heart disease, comprising: a heart sound signal data acquisition and processing module for children with heart disease, a heart sound signal data analysis module for children with heart disease, a comprehensive analysis module and an optimization and adjustment module: wherein the heart sound signal data acquisition and processing module for children with heart disease is used to acquire original heart sound signal data of children with heart disease through sensor equipment, and pre-process the heart sound signal data of children with heart disease to obtain heart sound signal data of children with heart disease; the heart sound signal data analysis module for children with heart disease is used to analyze the heart sound signal data of children with heart disease to obtain a synchronization evaluation value of the heart sound signal of children with heart disease and a quality evaluation value of the heart sound signal of children with heart disease; the comprehensive analysis module is used to obtain an accuracy evaluation value of the heart sound signal of children with heart disease through a comprehensive analysis of the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the artifact detection rate; and the optimization and adjustment module is used to adjust the abnormal identification of heart sound signals of children with heart disease according to the heart sound signal data analysis module for children with heart disease and the comprehensive analysis module.

[0010] Furthermore, the specific steps for obtaining the heart sound signal data of children with heart disease are as follows: collecting the original heart sound signal data of children with heart disease through sensor equipment; cleaning and denoising the original heart sound signal data of children with heart disease to obtain the heart sound signal data of children with heart disease; the heart sound signal data of children with heart disease includes the synchronization data of the heart sound signal of children with heart disease and the quality data of the heart sound signal of children with heart disease.

[0011] Furthermore, the specific steps for obtaining the synchronization evaluation value of the heart sound signal of children with heart disease are as follows: the synchronization data of the heart sound signal of children with heart disease include the heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, the maximum baseline drift amplitude and the minimum baseline drift amplitude; the heart sound signal transmission delay threshold, the heart sound signal data sampling rate standard value, the heart sound signal noise power threshold, the baseline drift amplitude standard value, the heart sound signal transmission delay weight factor, the heart sound signal data sampling rate weight factor, the heart sound signal noise power weight factor and the baseline drift coefficient weight factor are obtained from the heart sound signal database of children with heart disease; the heart sound signal transmission delay threshold and the heart sound signal transmission delay are subjected to a proportion analysis, and the weight factor of the heart sound signal transmission delay is used to correct the result of the proportion analysis to obtain the first component of the synchronization evaluation value of the heart sound signal of children with heart disease; the heart sound signal data sampling rate standard value and the heart sound signal data sampling rate and the heart sound signal data sampling rate standard value are subjected to a proportion analysis, and the weight factor of the heart sound signal data sampling rate is used to correct the result of the proportion analysis to obtain to the second component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform a ratio analysis on the heart sound signal noise power threshold and the heart sound signal noise power, use the weight factor of the heart sound signal noise power to correct the result of the ratio analysis, and record it as the third component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform a ratio analysis on the deviation of the maximum value of the baseline drift amplitude and the minimum value of the baseline drift amplitude and the standard value of the baseline drift amplitude, perform averaging on the result of the ratio analysis, and use the weight factor of the baseline drift coefficient to correct the result of the averaging, and record it as the fourth component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform coupling analysis on the first component of the synchronization evaluation value of the heart sound signal of children with heart disease, the second component of the synchronization evaluation value of the heart sound signal of children with heart disease and the third component of the synchronization evaluation value of the heart sound signal of children with heart disease, perform ratio processing on the result of the coupling analysis and the fourth component of the synchronization evaluation value of the heart sound signal of children with heart disease, and obtain the synchronization evaluation value of the heart sound signal of children with heart disease; the synchronization evaluation value of the heart sound signal of children with heart disease represents the quantitative data of the synchronization between the heart sound signal and the electrocardiogram signal.

[0012] Furthermore, the specific process of obtaining the heart sound signal quality evaluation value of children with heart disease is as follows: the heart sound signal quality data of children with heart disease include heart sound signal transmission delay, heart sound signal noise power, heart sound signal peak amplitude, heart sound signal valley amplitude and systolic murmur power; obtain the heart sound signal transmission delay threshold, heart sound signal noise power threshold, systolic murmur power threshold, heart sound signal fluctuation amplitude standard value, heart sound signal transmission delay weight factor, heart sound signal noise power weight factor, motion artifact amplitude coefficient weight factor and systolic murmur power weight factor from the heart sound signal database of children with heart disease; perform a proportion analysis on the heart sound signal transmission delay and the heart sound signal transmission delay threshold, use the heart sound signal transmission delay weight factor to correct the result of the proportion analysis, and record it as the first component of the heart sound signal quality evaluation value of children with heart disease; perform a proportion analysis on the heart sound signal noise power and the heart sound signal noise power threshold, use the heart sound signal noise power weight factor to correct the result of the proportion analysis, and record it as the first component of the heart sound signal quality evaluation value of children with heart disease the second component of the heart sound signal quality assessment value; performing a ratio analysis on the deviation between the heart sound signal peak amplitude and the heart sound signal valley amplitude and the standard value of the heart sound signal fluctuation amplitude, performing averaging on the result of the ratio analysis, and using the weight factor of the motion artifact amplitude coefficient to correct the result of the averaging, which is recorded as the third component of the heart sound signal quality assessment value for children with heart disease; performing a ratio analysis on the systolic murmur power and the systolic murmur power threshold, and using the weight factor of the systolic murmur power to correct the result of the ratio analysis, which is recorded as the fourth component of the heart sound signal quality assessment value for children with heart disease; performing a coupling analysis on the first component of the heart sound signal quality assessment value for children with heart disease, the second component of the heart sound signal quality assessment value for children with heart disease, the third component of the heart sound signal quality assessment value for children with heart disease, and the fourth component of the heart sound signal quality assessment value for children with heart disease, and obtaining the heart sound signal quality assessment value for children with heart disease; the heart sound signal quality assessment value for children with heart disease represents the accuracy of the heart sound signal during the detection and analysis process, and is quantitative data of the heart sound signal quality.

[0013] Furthermore, the specific steps of comprehensively analyzing and obtaining the accuracy evaluation value of the heart sound signal of children with heart disease are as follows: obtaining the preset artifact detection rate standard value, the weight factor of the synchronization evaluation value of the heart sound signal of children with heart disease, the weight factor of the artifact detection rate and the weight factor of the heart sound signal quality evaluation value of children with heart disease from the heart sound signal database of children with heart disease; averaging the synchronization evaluation value of the heart sound signal of children with heart disease, and using the weight factor of the synchronization evaluation value of the heart sound signal of children with heart disease to correct the result of the averaging, which is recorded as the first component of the accuracy evaluation value of the heart sound signal of children with heart disease; performing a proportion analysis on the deviation between the artifact detection rate and the artifact detection rate standard value and the artifact detection rate standard value, averaging the result of the proportion analysis, and using the artifact detection rate standard value to correct the result of the proportion analysis. The weight factor of the shadow detection rate is used to correct the result of the averaging processing, which is recorded as the second component of the accuracy evaluation value of the heart sound signal of children with heart disease; the quality evaluation value of the heart sound signal of children with heart disease is averaged, and the weight factor of the quality evaluation value of the heart sound signal of children with heart disease is used to correct the result of the averaging processing, which is recorded as the third component of the accuracy evaluation value of the heart sound signal of children with heart disease; the coupling results of the first component of the accuracy evaluation value of the heart sound signal of children with heart disease, the second component of the accuracy evaluation value of the heart sound signal of children with heart disease and the third component of the accuracy evaluation value of the heart sound signal of children with heart disease are analyzed in proportion, and the accuracy evaluation value of the heart sound signal of children with heart disease is obtained through comprehensive analysis; the accuracy evaluation value of the heart sound signal of children with heart disease represents the quantitative data of the accuracy of the heart sound signal.

[0014] Furthermore, the specific steps for adjusting the abnormal identification of heart sound signals of children with heart disease are as follows: obtaining the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, the second threshold value of the quality evaluation value of the heart sound signal of children with heart disease and the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of children with heart disease from the heart sound signal database of children with heart disease; comparing and analyzing the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the accuracy evaluation value of the heart sound signal of children with heart disease with the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, the second threshold value of the quality evaluation value of the heart sound signal of children with heart disease and the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of children with heart disease, respectively, to obtain an optimization and adjustment method for abnormal identification of heart sound signals of children with heart disease; the optimization and adjustment method for abnormal identification of heart sound signals of children with heart disease includes a synchronization optimization method of heart sound signals of children with heart disease, a quality optimization method of heart sound signals of children with heart disease and an accuracy optimization method of heart sound signals of children with heart disease.

[0015] Furthermore, the specific steps of the method for optimizing the synchronization of heart sound signals of children with heart disease are as follows: if the synchronization evaluation value of the heart sound signal of children with heart disease is greater than or equal to the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, then the method for optimizing the synchronization of heart sound signals of children with heart disease is not required; if the synchronization evaluation value of the heart sound signal of children with heart disease is lower than the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, then by reducing the wireless relay nodes, a direct connection topology is adopted to shorten the transmission path, and a PTP precision clock synchronization chip is deployed to ensure sampling rate synchronization, and a Kalman filter is used to compensate for delay, and a blind source separation algorithm is used to separate and suppress noise.

[0016] Furthermore, the specific steps of the method for optimizing the quality of heart sound signals for children with heart disease are as follows: if the quality evaluation value of the heart sound signal for children with heart disease is lower than or equal to the second threshold value of the quality evaluation value of the heart sound signal for children with heart disease, then the method for optimizing the quality of heart sound signals for children with heart disease is not required; if the quality evaluation value of the heart sound signal for children with heart disease is greater than the second threshold value of the quality evaluation value of the heart sound signal for children with heart disease, then the data of the acceleration sensor and the heart sound sensor are combined through multi-sensor fusion technology, and a two-level buffering mechanism is implemented.

[0017] Furthermore, the specific steps of the method for optimizing the accuracy of the heart sound signal of pediatric heart disease are as follows: if the accuracy evaluation value of the heart sound signal of pediatric heart disease is greater than or equal to the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of pediatric heart disease, then the method for optimizing the accuracy of the heart sound signal of pediatric heart disease is not required; if the accuracy evaluation value of the heart sound signal of pediatric heart disease is lower than the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of pediatric heart disease, then caches of different time lengths are set on the device side, node and cloud side respectively. When a fluctuation in the heart sound signal is detected, data can be immediately obtained from the local cache. When errors occur in the data on the device side or node side, correct data can be obtained from the cloud cache for replacement.

[0018] The present invention provides a method for identifying abnormalities in heart sound signals of children with heart disease, comprising: collecting original data of heart sound signals of children with heart disease through a sensor device, and preprocessing the heart sound signal data of children with heart disease to obtain heart sound signal data of children with heart disease; analyzing the heart sound signal data of children with heart disease to obtain a synchronization evaluation value of the heart sound signal of children with heart disease and a quality evaluation value of the heart sound signal of children with heart disease; comprehensively analyzing the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and an artifact detection rate to obtain an accuracy evaluation value of the heart sound signal of children with heart disease; and adjusting the abnormal identification of heart sound signals of children with heart disease according to a heart sound signal data analysis module of children with heart disease and a comprehensive analysis module.

[0019] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0020] 1. The heart sound signals of children with heart disease are collected through sensors, and the synchronization and quality of the heart sound signals of children with heart disease are evaluated after preprocessing. The heart sound signals of children with heart disease are comprehensively analyzed in combination with the artifact detection rate, thereby optimizing the abnormal recognition process of heart sound signals of children with heart disease, thereby improving the accuracy of the abnormal recognition system of heart sound signals of children with heart disease, and solving the problem of insufficient accuracy of the abnormal recognition system of heart sound signals of children with heart disease in the existing technology.

[0021] 2. Collect heart sound signals from children with heart disease through sensors, evaluate the synchronization and quality of heart sound signals from children with heart disease after preprocessing, thereby improving the clarity and quality of heart sound signals, ensuring the accuracy of subsequent analysis, and thus improving diagnostic accuracy and realizing remote monitoring.

[0022] 3. By combining the artifact detection rate to comprehensively analyze the heart sound signals of children with heart disease, the abnormal identification process of heart sound signals of children with heart disease can be optimized, thereby effectively detecting and removing artifacts, improving the quality of heart sound signals, reducing misdiagnosis and missed diagnosis, and thus improving efficiency and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of the structure of a system for identifying abnormal heart sound signals of children with heart disease provided in an embodiment of the present application;

[0024] Figure 2 A schematic diagram of the structure of a method for identifying abnormalities in heart sound signals of children with heart disease provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application solve the problem of insufficient accuracy of abnormality recognition systems for pediatric heart sound signals in the prior art by providing an abnormality recognition system and method for pediatric heart sound signals. The system collects pediatric heart sound signals through sensors, evaluates the synchronization and quality of pediatric heart sound signals after preprocessing, and comprehensively analyzes pediatric heart sound signals in combination with the artifact detection rate, thereby optimizing the abnormality recognition process for pediatric heart sound signals and thereby improving the accuracy of the abnormality recognition system for pediatric heart sound signals.

[0026] The technical solution in the embodiment of the present application is to solve the problem of insufficient accuracy of the abnormal recognition system for heart sound signals of children with heart disease. The overall idea is as follows:

[0027] The heart sound signals of children with heart disease are collected through sensors, and the synchronization and quality of the heart sound signals of children with heart disease are evaluated after preprocessing. The heart sound signals of children with heart disease are comprehensively analyzed in combination with the artifact detection rate, thereby optimizing the abnormal recognition process of heart sound signals of children with heart disease, thereby improving the accuracy of the abnormal recognition system of heart sound signals of children with heart disease.

[0028] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0029] like Figure 1 As shown, it is a structural schematic diagram of an abnormality identification system for a heart sound signal of a child with heart disease provided by an embodiment of the present application. The abnormality identification system for a heart sound signal of a child with heart disease provided by an embodiment of the present application includes: a heart sound signal data acquisition and processing module for a heart sound signal of a child with heart disease, a heart sound signal data analysis module for a child with heart disease, a comprehensive analysis module and an optimization and adjustment module: wherein the heart sound signal data acquisition and processing module for a heart sound signal of a child with heart disease is used to collect the original heart sound signal data of a heart sound signal of a child with heart disease through a sensor device, and pre-process the heart sound signal data of a heart sound signal of a child with heart disease to obtain the heart sound signal data of a child with heart disease; the heart sound signal data analysis module for a heart sound signal of a child with heart disease is used to analyze the heart sound signal data of a child with heart disease to obtain a synchronization evaluation value of the heart sound signal of a child with heart disease and a quality evaluation value of the heart sound signal of a child with heart disease; the comprehensive analysis module is used to obtain an accuracy evaluation value of the heart sound signal of a child with heart disease through a comprehensive analysis of the synchronization evaluation value of the heart sound signal of a child with heart disease, the quality evaluation value of the heart sound signal of a child with heart disease and the artifact detection rate; the optimization and adjustment module is used to adjust the abnormality identification of the heart sound signal of a child with heart disease according to the heart sound signal data analysis module for a child with heart disease and the comprehensive analysis module.

[0030] In this embodiment, the sensor device includes but is not limited to an accelerometer, a magnetic sensor, an ultrasonic sensor, and the like.

[0031] Furthermore, the specific steps for obtaining the heart sound signal data of children with heart disease are as follows: collecting the original heart sound signal data of children with heart disease through sensor equipment; cleaning and denoising the original heart sound signal data of children with heart disease to obtain the heart sound signal data of children with heart disease; the heart sound signal data of children with heart disease includes the synchronization data of the heart sound signal of children with heart disease and the quality data of the heart sound signal of children with heart disease.

[0032] Furthermore, the specific steps for obtaining the synchronization evaluation value of the heart sound signal of children with heart disease are as follows: the synchronization data of the heart sound signal of children with heart disease include the heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, the maximum baseline drift amplitude and the minimum baseline drift amplitude; the heart sound signal transmission delay threshold, the heart sound signal data sampling rate standard value, the heart sound signal noise power threshold, the baseline drift amplitude standard value, the heart sound signal transmission delay weight factor, the heart sound signal data sampling rate weight factor, the heart sound signal noise power weight factor and the baseline drift coefficient weight factor are obtained from the heart sound signal database of children with heart disease; the heart sound signal transmission delay threshold and the heart sound signal transmission delay are subjected to a proportion analysis, and the weight factor of the heart sound signal transmission delay is used to correct the result of the proportion analysis to obtain the first component of the synchronization evaluation value of the heart sound signal of children with heart disease; the heart sound signal data sampling rate standard value and the heart sound signal data sampling rate and the heart sound signal data sampling rate standard value are subjected to a proportion analysis, and the weight factor of the heart sound signal data sampling rate is used to correct the result of the proportion analysis to obtain to the second component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform a ratio analysis on the heart sound signal noise power threshold and the heart sound signal noise power, use the weight factor of the heart sound signal noise power to correct the result of the ratio analysis, and record it as the third component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform a ratio analysis on the deviation of the maximum value of the baseline drift amplitude and the minimum value of the baseline drift amplitude and the standard value of the baseline drift amplitude, perform averaging on the result of the ratio analysis, and use the weight factor of the baseline drift coefficient to correct the result of the averaging, and record it as the fourth component of the synchronization evaluation value of the heart sound signal of children with heart disease; perform coupling analysis on the first component of the synchronization evaluation value of the heart sound signal of children with heart disease, the second component of the synchronization evaluation value of the heart sound signal of children with heart disease and the third component of the synchronization evaluation value of the heart sound signal of children with heart disease, perform ratio processing on the result of the coupling analysis and the fourth component of the synchronization evaluation value of the heart sound signal of children with heart disease, and obtain the synchronization evaluation value of the heart sound signal of children with heart disease; the synchronization evaluation value of the heart sound signal of children with heart disease represents the quantitative data of the synchronization between the heart sound signal and the electrocardiogram signal.

[0033] In this embodiment, the specific method for analyzing and obtaining the synchronization evaluation value of the heart sound signal of a child with heart disease is as follows:

[0034]

[0035] The preset pediatric heart sound signal synchronization detection points are numbered in sequence, S0 represents the number of the pediatric heart sound signal synchronization detection point under the L0th pediatric heart sound signal synchronization detection segment, S0=1,2,…,S, S represents the total number of pediatric heart sound signal synchronization detection points.

[0036] The preset pediatric heart disease heart sound signal synchronization time is divided into pediatric heart disease heart sound signal synchronization detection segments of the same time length, L0 represents the number of the pediatric heart disease heart sound signal synchronization detection segment, L0=1,2,…,L, L represents the total number of the pediatric heart disease heart sound signal synchronization detection segments.

[0037] The synchronization evaluation value of the heart sound signal of a child with heart disease is represented by the S0th synchronization detection point of the heart sound signal of a child with heart disease.

[0038] Q0 represents the heart sound signal transmission delay threshold, which is a preset heart sound signal transmission delay threshold obtained from a pediatric heart disease heart sound signal database, and can be the average heart sound signal transmission delay value under a preset historical pediatric heart sound signal synchronization detection point from the pediatric heart disease heart sound signal database.

[0039] The transmission delay of the heart sound signal at the synchronization detection point S0 of the heart sound signal of a child with heart disease is represented. A signal with a precise time stamp is sent from the sensor end. After the data processing unit receives the signal, the arrival time of the signal is measured. The calculated time difference is the transmission delay of the heart sound signal.

[0040] C0 represents the standard value of the heart sound signal data sampling rate, which is a preset standard value of the heart sound signal data sampling rate obtained from the pediatric heart disease heart sound signal database, and can be the average value of the heart sound signal data sampling rate at the preset historical pediatric heart disease heart sound signal synchronization detection point from the pediatric heart disease heart sound signal database.

[0041] represents the sampling rate of the heart sound signal data at the S0th pediatric heart disease heart sound signal synchronization detection point.

[0042] K1 represents a constant term, which is a number set to avoid meaningless points in the data.

[0043] Z0 represents the heart sound signal noise power threshold, which is a preset heart sound signal noise power threshold obtained from a pediatric heart disease heart sound signal database, and can be the average heart sound signal noise power value at a preset historical pediatric heart sound signal synchronization detection point from the pediatric heart disease heart sound signal database.

[0044] represents the noise power of the heart sound signal at the S0th synchronization detection point of the heart sound signal of a child with heart disease.

[0045] It represents the baseline drift coefficient of the L0th pediatric heart disease heart sound signal synchronization detection segment.

[0046] K2 represents a constant term, which is a number set to avoid meaningless points in the data.

[0047] It represents the maximum value of the baseline drift amplitude in the synchronization detection segment of the heart sound signal of a child with heart disease L0.

[0048] It represents the minimum value of the baseline drift amplitude in the synchronization detection segment of the heart sound signal of the child with heart disease L0.

[0049] F0 represents the baseline drift amplitude standard value, which is a preset baseline drift amplitude standard value obtained from the pediatric heart disease heart sound signal database, and can be the average baseline drift amplitude under the synchronization detection point of the preset historical pediatric heart disease heart sound signal from the pediatric heart disease heart sound signal database.

[0050] ρ1 is the preset heart sound signal transmission delay weight factor obtained from the pediatric heart disease heart sound signal database.

[0051] ρ2 is the preset heart sound signal data sampling rate weighting factor obtained from the pediatric heart disease heart sound signal database.

[0052] ρ3 is the preset heart sound signal noise power weighting factor obtained from the pediatric heart disease heart sound signal database.

[0053] ρ4 is a preset baseline drift coefficient weighting factor obtained from a pediatric heart sound signal database.

[0054] By obtaining a mapping table of weight factors from a database, the corresponding weight factors, such as the weight factor for the heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, and the baseline wander coefficient, are quickly extracted based on the current heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, and the baseline wander coefficient. For example, this mapping table defines a clear set of association rules that converts the specific values ​​of the heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, and the baseline wander coefficient into their corresponding weight factors. This mechanism effectively achieves dynamic acquisition of weight factors, whether achieving a one-to-one precise match or a many-to-one relationship where multiple parameters are aggregated into a single weight.

[0055] The higher the sampling rate of the heart sound signal data, the larger the amount of data generated. During the transmission process, more data means that it takes longer to transmit the heart sound signal of the same length, thereby increasing the transmission delay. The transmission delay of the heart sound signal increases; the higher the sampling rate of the heart sound signal data, the higher the bandwidth required, which will introduce more noise and increase the noise power of the heart sound signal; the higher the noise power of the heart sound signal, the more it will mask the subtle changes in the heart sound signal, so that the baseline drift will cause the baseline of the signal to be unstable, and the difference between the maximum baseline drift amplitude and the minimum baseline drift amplitude will increase.

[0056] There is a negative correlation between the transmission delay of heart sound signal and the synchronization evaluation value of heart sound signal in children with heart disease. The transmission delay will affect the real-time monitoring and feedback of the signal. The greater the delay, the less synchronized the signal will be. The higher the transmission delay of heart sound signal, the smaller the synchronization evaluation value of heart sound signal in children with heart disease. There is a negative correlation between the absolute value of the difference between the sampling rate of heart sound signal data and the standard value of the sampling rate of heart sound signal data and the synchronization evaluation value of heart sound signal in children with heart disease. The higher the sampling rate, the more data volume will be, resulting in increased transmission delay. The lower the sampling rate, the more incomplete the sampling. The sampling rate of heart sound signal data and the standard value of the sampling rate of heart sound signal data are negatively correlated. The larger the absolute value of the accuracy difference, the smaller the synchronization evaluation value of the heart sound signal in children with heart disease; there is a negative correlation between the noise power of the heart sound signal and the synchronization evaluation value of the heart sound signal in children with heart disease. The lower the noise power of the heart sound signal, the less interference to the heart sound signal and the better the clarity of the signal. The higher the noise power of the heart sound signal, the smaller the synchronization evaluation value of the heart sound signal in children with heart disease. There is a negative correlation between the baseline drift coefficient and the synchronization evaluation value of the heart sound signal in children with heart disease. The smaller the baseline drift coefficient, the more stable the baseline of the signal. The larger the baseline drift coefficient, the smaller the synchronization evaluation value of the heart sound signal in children with heart disease.

[0057] Furthermore, the specific process of obtaining the heart sound signal quality evaluation value of children with heart disease is as follows: the heart sound signal quality data of children with heart disease include heart sound signal transmission delay, heart sound signal noise power, heart sound signal peak amplitude, heart sound signal valley amplitude and systolic murmur power; obtain the heart sound signal transmission delay threshold, heart sound signal noise power threshold, systolic murmur power threshold, heart sound signal fluctuation amplitude standard value, heart sound signal transmission delay weight factor, heart sound signal noise power weight factor, motion artifact amplitude coefficient weight factor and systolic murmur power weight factor from the heart sound signal database of children with heart disease; perform a proportion analysis on the heart sound signal transmission delay and the heart sound signal transmission delay threshold, use the heart sound signal transmission delay weight factor to correct the result of the proportion analysis, and record it as the first component of the heart sound signal quality evaluation value of children with heart disease; perform a proportion analysis on the heart sound signal noise power and the heart sound signal noise power threshold, use the heart sound signal noise power weight factor to correct the result of the proportion analysis, and record it as the first component of the heart sound signal quality evaluation value of children with heart disease the second component of the heart sound signal quality assessment value; performing a ratio analysis on the deviation between the heart sound signal peak amplitude and the heart sound signal valley amplitude and the standard value of the heart sound signal fluctuation amplitude, performing averaging on the result of the ratio analysis, and using the weight factor of the motion artifact amplitude coefficient to correct the result of the averaging, which is recorded as the third component of the heart sound signal quality assessment value for children with heart disease; performing a ratio analysis on the systolic murmur power and the systolic murmur power threshold, and using the weight factor of the systolic murmur power to correct the result of the ratio analysis, which is recorded as the fourth component of the heart sound signal quality assessment value for children with heart disease; performing a coupling analysis on the first component of the heart sound signal quality assessment value for children with heart disease, the second component of the heart sound signal quality assessment value for children with heart disease, the third component of the heart sound signal quality assessment value for children with heart disease, and the fourth component of the heart sound signal quality assessment value for children with heart disease, and obtaining the heart sound signal quality assessment value for children with heart disease; the heart sound signal quality assessment value for children with heart disease represents the accuracy of the heart sound signal during the detection and analysis process, and is quantitative data of the heart sound signal quality.

[0058] In this embodiment, the specific method for analyzing and obtaining the quality assessment value of the heart sound signal of a child with heart disease is as follows:

[0059]

[0060] The preset pediatric heart disease heart sound signal quality detection points are numbered in sequence, N0 represents the number of the pediatric heart disease heart sound signal quality detection point under the T0th pediatric heart disease heart sound signal quality detection segment, N0=1,2,...,N, N represents the total number of pediatric heart disease heart sound signal quality detection points.

[0061] The preset pediatric heart disease heart sound signal quality time is divided into pediatric heart disease heart sound signal quality detection segments of equal length. T0 represents the number of the pediatric heart disease heart sound signal quality detection segment. T0=1, 2, ..., T, T represents the total number of the pediatric heart disease heart sound signal quality detection segments.

[0062] It represents the quality assessment value of the pediatric heart sound signal of the N0th pediatric heart sound signal quality detection point.

[0063] It represents the heart sound signal transmission delay at the N0th heart sound signal quality detection point for children with heart disease.

[0064] Y0 represents the heart sound signal transmission delay threshold, which is a preset heart sound signal transmission delay threshold obtained from a pediatric heart disease heart sound signal database, and can be the average value of the heart sound signal transmission delay under historical pediatric heart disease heart sound signal quality detection points preset in the pediatric heart disease heart sound signal database.

[0065] It represents the noise power of the heart sound signal at the N0th detection point of the heart sound signal quality of children with heart disease.

[0066] P0 represents the heart sound signal noise power threshold, which is a preset heart sound signal noise power threshold obtained from a pediatric heart disease heart sound signal database, and can be the average heart sound signal noise power value at a preset historical pediatric heart sound signal quality detection point in the pediatric heart disease heart sound signal database.

[0067] It represents the motion artifact amplitude coefficient of the T0th pediatric heart sound signal quality detection segment.

[0068] It represents the systolic murmur power at the N0th pediatric heart sound signal quality detection point.

[0069] M0 represents the systolic murmur power threshold, which is a preset systolic murmur power threshold obtained from a pediatric heart disease heart sound signal database, and may be an average systolic murmur power value under a preset historical pediatric heart disease heart sound signal quality detection point from the pediatric heart disease heart sound signal database.

[0070] It represents the peak amplitude of the heart sound signal in the T0th pediatric heart disease heart sound signal quality detection segment.

[0071] It represents the valley amplitude of the heart sound signal in the T0th pediatric heart disease heart sound signal quality detection segment.

[0072] X0 represents the standard value of the heart sound signal fluctuation amplitude, which is a preset standard value of the heart sound signal fluctuation amplitude obtained from the pediatric heart disease heart sound signal database, and can be the average value of the heart sound signal fluctuation amplitude under the historical pediatric heart disease heart sound signal quality detection segment preset from the pediatric heart disease heart sound signal database.

[0073] Q1 is a preset heart sound signal transmission delay weighting factor obtained from a pediatric heart disease heart sound signal database.

[0074] Q2 is a preset heart sound signal noise power weighting factor obtained from a pediatric heart disease heart sound signal database.

[0075] Q3 is a preset motion artifact amplitude coefficient weighting factor obtained from a pediatric heart sound signal database.

[0076] Q4 is the preset systolic murmur power weighting factor obtained from the pediatric heart sound signal database.

[0077] By obtaining a mapping table of weight factors from a database, the corresponding weight factors, such as the weight factor for the heart sound signal transmission delay, the heart sound signal noise power, the motion artifact amplitude coefficient, and the systolic murmur power, are quickly extracted based on the current heart sound signal transmission delay, the heart sound signal noise power, the motion artifact amplitude coefficient, and the systolic murmur power. For example, this mapping table defines a clear set of association rules that converts the specific values ​​of the heart sound signal transmission delay, the heart sound signal noise power, the motion artifact amplitude coefficient, and the systolic murmur power into their corresponding weight factors. Under this mechanism, whether achieving a one-to-one exact match or a many-to-one relationship where multiple parameters are aggregated into a single weight, the dynamic acquisition of weight factors can be effectively achieved.

[0078] The greater the transmission delay of the heart sound signal, the longer the time difference between the signal generation and reception, which will introduce more noise and increase the noise power of the heart sound signal; the increase in the noise power of the heart sound signal will mask the subtle changes in the signal, the less accurate the motion artifact amplitude coefficient will be, and the greater the fluctuation between the peak amplitude and the valley amplitude of the heart sound signal will be; the larger the motion artifact coefficient is, the more it will interfere with the detection of systolic murmur, and increasing the motion artifact amplitude coefficient will lead to an increase in the power of the systolic murmur.

[0079] There is a positive correlation between the transmission delay of heart sound signals and the quality assessment value of heart sound signals in children with heart disease. The longer the time difference between the generation and reception of the signal, the lower the real-time performance of the signal, the greater the transmission delay of the heart sound signal, and the greater the quality assessment value of the heart sound signal in children with heart disease; there is a positive correlation between the noise power of the heart sound signal and the quality assessment value of the heart sound signal in children with heart disease. The increase in noise power will reduce the signal-to-noise ratio of the signal and affect the clarity of the signal. The greater the noise power of the heart sound signal, the greater the quality assessment value of the heart sound signal in children with heart disease; there is a positive correlation between the motion artifact amplitude coefficient and the quality assessment value of the heart sound signal in children with heart disease. The greater the signal is affected by motion, the lower the accuracy of the signal. The greater the motion artifact amplitude coefficient, the greater the quality assessment value of the heart sound signal in children with heart disease; there is a positive correlation between the systolic murmur power and the quality assessment value of the heart sound signal in children with heart disease. The increase in systolic murmur power will mask normal heart sounds and affect diagnosis. The greater the systolic murmur power, the greater the quality assessment value of the heart sound signal in children with heart disease.

[0080] Furthermore, the specific steps of comprehensively analyzing and obtaining the accuracy evaluation value of the heart sound signal of children with heart disease are as follows: obtaining the preset artifact detection rate standard value, the weight factor of the synchronization evaluation value of the heart sound signal of children with heart disease, the weight factor of the artifact detection rate and the weight factor of the heart sound signal quality evaluation value of children with heart disease from the heart sound signal database of children with heart disease; averaging the synchronization evaluation value of the heart sound signal of children with heart disease, and using the weight factor of the synchronization evaluation value of the heart sound signal of children with heart disease to correct the result of the averaging, which is recorded as the first component of the accuracy evaluation value of the heart sound signal of children with heart disease; performing a proportion analysis on the deviation between the artifact detection rate and the artifact detection rate standard value and the artifact detection rate standard value, averaging the result of the proportion analysis, and using the artifact detection rate standard value to correct the result of the proportion analysis. The weight factor of the shadow detection rate is used to correct the result of the averaging processing, which is recorded as the second component of the accuracy evaluation value of the heart sound signal of children with heart disease; the quality evaluation value of the heart sound signal of children with heart disease is averaged, and the weight factor of the quality evaluation value of the heart sound signal of children with heart disease is used to correct the result of the averaging processing, which is recorded as the third component of the accuracy evaluation value of the heart sound signal of children with heart disease; the coupling results of the first component of the accuracy evaluation value of the heart sound signal of children with heart disease, the second component of the accuracy evaluation value of the heart sound signal of children with heart disease and the third component of the accuracy evaluation value of the heart sound signal of children with heart disease are analyzed in proportion, and the accuracy evaluation value of the heart sound signal of children with heart disease is obtained through comprehensive analysis; the accuracy evaluation value of the heart sound signal of children with heart disease represents the quantitative data of the accuracy of the heart sound signal.

[0081] In this embodiment, the specific method for analyzing and obtaining the accuracy evaluation value of the heart sound signal of a child with heart disease is as follows:

[0082]

[0083] The preset pediatric heart sound signal accuracy detection points are numbered in sequence, H0 represents the number of the pediatric heart sound signal accuracy detection point, H0=1,2,…,H,H represents the total number of pediatric heart sound signal accuracy detection points.

[0084] Indicates the accuracy assessment value of heart sound signals in children with heart disease.

[0085] The synchronization evaluation value of the heart sound signal of a child with heart disease is represented by the S0th synchronization detection point of the heart sound signal of a child with heart disease.

[0086] It represents the artifact detection rate at the H0th detection point of the heart sound signal accuracy of children with heart disease.

[0087] A0 represents the standard value of the artifact detection rate, which is a preset standard value of the artifact detection rate obtained from the pediatric heart disease heart sound signal database, and can be the average value of the artifact detection rate under the preset historical pediatric heart disease heart sound signal accuracy detection points from the pediatric heart disease heart sound signal database.

[0088] It represents the quality assessment value of the pediatric heart sound signal of the N0th pediatric heart sound signal quality detection point.

[0089] K3 represents a constant term, which is a number set to avoid meaningless points in the data.

[0090] is a weighting factor of a preset pediatric heart disease heart sound signal synchronization evaluation value obtained from a pediatric heart disease heart sound signal database.

[0091] is the weighting factor for the preset artifact detection rate obtained from the pediatric heart sound signal database.

[0092] is a weighting factor of a preset pediatric heart disease heart sound signal quality assessment value obtained from a pediatric heart disease heart sound signal database.

[0093] By obtaining a mapping table of weight factors from a database, the corresponding weight factors can be quickly extracted based on the current pediatric heart sound signal synchronization assessment value, artifact detection rate, and pediatric heart sound signal quality assessment value. For example, this mapping table defines a clear set of association rules that converts the specific values ​​of the pediatric heart sound signal synchronization assessment value, artifact detection rate, and pediatric heart sound signal quality assessment value into their corresponding weight factors. This mechanism effectively achieves dynamic acquisition of weight factors, whether achieving a one-to-one exact match or a many-to-one relationship where multiple parameters are aggregated into a single weight.

[0094] The higher the synchronization assessment value of the heart sound signal of children with heart disease, the more accurate the signal acquisition and processing can be, ensuring that the signal acquisition and transmission will not introduce time delay or distortion, and the lower the quality assessment value of the heart sound signal of children with heart disease; the higher the artifact detection rate is, the more likely it is that useful heart sound information will be mistakenly deleted, and the higher the quality assessment value of the heart sound signal of children with heart disease.

[0095] There is a positive correlation between the synchronization evaluation value of the heart sound signal of children with heart disease and the accuracy evaluation value of the heart sound signal of children with heart disease. The higher the synchronization evaluation value of the heart sound signal of children with heart disease, the more accurate the time relationship between the heart sound signal and the actual heart activity. The higher the synchronization evaluation value of the heart sound signal of children with heart disease, the higher the accuracy evaluation value of the heart sound signal of children with heart disease; there is a negative correlation between the absolute value of the difference between the artifact detection rate and the standard value of the artifact detection rate and the accuracy evaluation value of the heart sound signal of children with heart disease. The larger the absolute value of the difference between the artifact detection rate and the standard value of the artifact detection rate, the more useful heart sound information will be mistakenly deleted. The larger the absolute value of the difference between the artifact detection rate and the standard value of the artifact detection rate, the smaller the accuracy evaluation value of the heart sound signal of children with heart disease; there is a negative correlation between the quality evaluation value of the heart sound signal of children with heart disease and the accuracy evaluation value of the heart sound signal of children with heart disease. The higher the quality evaluation value of the heart sound signal of children with heart disease, the worse the overall quality of the signal. The larger the quality evaluation value of the heart sound signal of children with heart disease, the smaller the accuracy evaluation value of the heart sound signal of children with heart disease.

[0096] Furthermore, the specific steps for adjusting the abnormal identification of heart sound signals of children with heart disease are as follows: obtaining the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, the second threshold value of the quality evaluation value of the heart sound signal of children with heart disease and the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of children with heart disease from the heart sound signal database of children with heart disease; comparing and analyzing the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the accuracy evaluation value of the heart sound signal of children with heart disease with the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, the second threshold value of the quality evaluation value of the heart sound signal of children with heart disease and the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of children with heart disease, respectively, to obtain an optimization and adjustment method for abnormal identification of heart sound signals of children with heart disease; the optimization and adjustment method for abnormal identification of heart sound signals of children with heart disease includes a synchronization optimization method of heart sound signals of children with heart disease, a quality optimization method of heart sound signals of children with heart disease and an accuracy optimization method of heart sound signals of children with heart disease.

[0097] Furthermore, the specific steps of the method for optimizing the synchronization of heart sound signals of children with heart disease are as follows: if the synchronization evaluation value of the heart sound signal of children with heart disease is greater than or equal to the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, then the method for optimizing the synchronization of heart sound signals of children with heart disease is not required; if the synchronization evaluation value of the heart sound signal of children with heart disease is lower than the first threshold value of the synchronization evaluation value of the heart sound signal of children with heart disease, then by reducing the wireless relay nodes, a direct connection topology is adopted to shorten the transmission path, and a PTP precision clock synchronization chip is deployed to ensure sampling rate synchronization, and a Kalman filter is used to compensate for delay, and a blind source separation algorithm is used to separate and suppress noise.

[0098] In this embodiment, adjustments are first made at the hardware level. By reducing wireless relay nodes or adopting a direct connection topology to shorten the transmission path and reduce latency, a PTP precision clock synchronization chip is deployed to ensure high synchronization of timestamps between all sensors and data processing units. At the software level, a Kalman filter is used to adjust and compensate for signal misalignment caused by hardware delays in real time. For sampling rate synchronization, an atomic clock-level reference clock source, such as a rubidium clock or a cesium clock, is deployed to provide a high-precision clock signal. A clock tree distribution architecture is used to distribute the master clock signal to the ADCs of multiple sensors to ensure sampling rate consistency. A sinc interpolation algorithm is used to resample signals at different sampling rates, and a blind source separation algorithm (such as independent component analysis (ICA)) is used to jointly process the signals of multiple sensors to separate heart sound signals and noise signals and suppress noise.

[0099] Furthermore, the specific steps of the method for optimizing the quality of heart sound signals for children with heart disease are as follows: if the quality evaluation value of the heart sound signal for children with heart disease is lower than or equal to the second threshold value of the quality evaluation value of the heart sound signal for children with heart disease, then the method for optimizing the quality of heart sound signals for children with heart disease is not required; if the quality evaluation value of the heart sound signal for children with heart disease is greater than the second threshold value of the quality evaluation value of the heart sound signal for children with heart disease, then the data of the acceleration sensor and the heart sound sensor are combined through multi-sensor fusion technology, and a two-level buffering mechanism is implemented.

[0100] In this embodiment, multi-sensor fusion technology is first used to combine data from an accelerometer and a heart sound sensor. The accelerometer provides information on motion status, while the heart sound sensor captures heart sound signals. To reduce data transmission delays, a two-level buffering mechanism is implemented. The first-level buffer is used to temporarily store the collected data, primarily serving as a temporary storage area during the data collection phase. When sensors (such as accelerometers and heart sound sensors) collect data at a preset frequency threshold, the data is first stored in the first-level buffer. The first-level buffer decouples the data collection process from subsequent processing (such as data fusion and LSTM analysis), allowing data collection to proceed continuously without waiting for the completion of the processing process. The second-level buffer is used to provide smooth output during the data transmission process. Its main function is to receive data from the first-level buffer or other data sources and transmit it to the final destination (such as data storage) at a controllable rate. The buffering mechanism can smooth the data flow and avoid transmission bottlenecks caused by bursty data volumes. At the same time, the use of a deep denoising autoencoder and a dynamic gain control strategy reduces noise power, improving the real-time and clarity of the signal.

[0101] Furthermore, the specific steps of the method for optimizing the accuracy of the heart sound signal of pediatric heart disease are as follows: if the accuracy evaluation value of the heart sound signal of pediatric heart disease is greater than or equal to the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of pediatric heart disease, then the method for optimizing the accuracy of the heart sound signal of pediatric heart disease is not required; if the accuracy evaluation value of the heart sound signal of pediatric heart disease is lower than the comprehensive threshold value of the accuracy evaluation value of the heart sound signal of pediatric heart disease, then caches of different time lengths are set on the device side, node and cloud side respectively. When a fluctuation in the heart sound signal is detected, data can be immediately obtained from the local cache. When errors occur in the data on the device side or node side, correct data can be obtained from the cloud cache for replacement.

[0102] In this embodiment, a local cache with a preset time is set on the device side to store the most recently received data. When the device detects signal fluctuations, it can immediately obtain data from the local cache to ensure data continuity, thereby effectively resisting instantaneous jitter. A node cache with a preset time is set at each node (such as a server or relay station), which helps to align during multi-channel data transmission and ensure that the data of each channel remains consistent in time. A cloud cache with a preset time is set on the cloud for long-term error correction. When errors occur in the data on the device side or node side, the correct data can be obtained from the cloud cache for replacement.

[0103] An embodiment of the present application provides a method for identifying abnormalities in heart sound signals of children with heart disease, including: collecting original data of heart sound signals of children with heart disease through a sensor device, and preprocessing the heart sound signal data of children with heart disease to obtain heart sound signal data of children with heart disease; analyzing the heart sound signal data of children with heart disease to obtain a synchronization evaluation value of the heart sound signal of children with heart disease and a quality evaluation value of the heart sound signal of children with heart disease; comprehensively analyzing the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the artifact detection rate to obtain an accuracy evaluation value of the heart sound signal of children with heart disease; and adjusting the abnormal identification of heart sound signals of children with heart disease according to the heart sound signal data analysis module of children with heart disease and the comprehensive analysis module.

[0104] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0105] Obviously, those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations fall within the scope of the claims and their equivalents, the present invention is intended to include such modifications and variations.

Claims

1. A system for identifying abnormal heart sound signals of children with heart disease, characterized in that: It includes a children's heart sound signal data acquisition and processing module, a children's heart sound signal data analysis module, a comprehensive analysis module, and an optimization and adjustment module: The pediatric heart sound signal data acquisition and processing module is used to collect original data of pediatric heart sound signals through sensor equipment, and pre-process the data of pediatric heart sound signals to obtain pediatric heart sound signal data; The pediatric heart sound signal data analysis module is used to analyze the pediatric heart sound signal data to obtain a pediatric heart sound signal synchronization evaluation value and a pediatric heart sound signal quality evaluation value; The specific steps of obtaining the synchronization evaluation value of the heart sound signal of a child with heart disease are as follows: The synchronization data of the heart sound signal of the child with heart disease includes the heart sound signal transmission delay, the heart sound signal data sampling rate, the heart sound signal noise power, the maximum value of the baseline drift amplitude and the minimum value of the baseline drift amplitude; Obtaining from a pediatric heart disease heart sound signal database a heart sound signal transmission delay threshold, a heart sound signal data sampling rate standard value, a heart sound signal noise power threshold, a baseline drift amplitude standard value, a heart sound signal transmission delay weighting factor, a heart sound signal data sampling rate weighting factor, a heart sound signal noise power weighting factor, and a baseline drift coefficient weighting factor; The heart sound signal transmission delay threshold and the heart sound signal transmission delay are analyzed in proportion, and the weight factor of the heart sound signal transmission delay is used to correct the results of the proportion analysis to obtain the first component of the heart sound signal synchronization evaluation value for children with heart disease. The deviations between the standard value of the heart sound signal data sampling rate and the standard value of the heart sound signal data sampling rate and the heart sound signal data sampling rate are analyzed in proportion. The weight factor of the heart sound signal data sampling rate is used to correct the results of the proportion analysis to obtain the second component of the heart sound signal synchronization assessment value for children with heart disease. The heart sound signal noise power threshold and the heart sound signal noise power are analyzed for a certain percentage. The weight factor of the heart sound signal noise power is used to modify the result of the analysis and record it as the third component of the heart sound signal synchronization assessment value for children with heart disease. The deviations of the maximum and minimum baseline drift amplitudes were compared with the standard value of the baseline drift amplitude. The results of the ratio analysis were averaged and corrected using the weight factor of the baseline drift coefficient. The results were recorded as the fourth component of the synchronization assessment value of the heart sound signal in children with heart disease. The first component of the synchronization evaluation value of the heart sound signal of the child with heart disease, the second component of the synchronization evaluation value of the heart sound signal of the child with heart disease and the third component of the synchronization evaluation value of the heart sound signal of the child with heart disease are coupled and analyzed, and the result of the coupling analysis is proportionally processed with the fourth component of the synchronization evaluation value of the heart sound signal of the child with heart disease to obtain the synchronization evaluation value of the heart sound signal of the child with heart disease; The synchronicity evaluation value of the heart sound signal in children with heart disease represents the quantitative data of the synchronicity between the heart sound signal and the electrocardiogram signal; The specific process of obtaining the quality assessment value of the heart sound signal of a child with heart disease is as follows: The heart sound signal quality data of children with heart disease includes heart sound signal transmission delay, heart sound signal noise power, heart sound signal peak amplitude, heart sound signal valley amplitude and systolic murmur power; The heart sound signal transmission delay threshold, heart sound signal noise power threshold, systolic murmur power threshold, heart sound signal fluctuation amplitude standard value, heart sound signal transmission delay weight factor, heart sound signal noise power weight factor, motion artifact amplitude coefficient weight factor, and systolic murmur power weight factor were obtained from the pediatric heart disease heart sound signal database; A ratio analysis is performed on the heart sound signal transmission delay and the heart sound signal transmission delay threshold. The weight factor of the heart sound signal transmission delay is used to correct the result of the ratio analysis and record it as the first component of the heart sound signal quality assessment value for pediatric heart disease. The heart sound signal noise power and the heart sound signal noise power threshold are analyzed for their proportions, and the weight factor of the heart sound signal noise power is used to correct the result of the proportion analysis, which is recorded as the second component of the heart sound signal quality assessment value for children with heart disease. The deviations of the heart sound signal peak amplitude and the heart sound signal valley amplitude with the standard value of the heart sound signal fluctuation amplitude were analyzed for a certain percentage. The results of the percentage analysis were averaged and corrected using the weight factor of the motion artifact amplitude coefficient. The results were recorded as the third component of the heart sound signal quality assessment value for children with heart disease. The systolic murmur power and the systolic murmur power threshold were analyzed for proportion, and the weight factor of the systolic murmur power was used to correct the results of the proportion analysis and record it as the fourth component of the heart sound signal quality assessment value for children with heart disease. Couple the first component of the heart sound signal quality assessment value for children with heart disease, the second component of the heart sound signal quality assessment value for children with heart disease, the third component of the heart sound signal quality assessment value for children with heart disease, and the fourth component of the heart sound signal quality assessment value for children with heart disease to obtain the heart sound signal quality assessment value for children with heart disease; The heart sound signal quality assessment value for children with heart disease indicates the accuracy of the heart sound signal during detection and analysis, and is a quantitative data of the heart sound signal quality; The comprehensive analysis module is used to comprehensively analyze the synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the artifact detection rate to obtain the accuracy evaluation value of the heart sound signal of children with heart disease; The optimization and adjustment module is used to adjust the abnormal recognition of the heart sound signals of children with heart disease based on the heart sound signal data analysis module of children with heart disease and the comprehensive analysis module.

2. The abnormality recognition system for heart sound signals of children with heart disease according to claim 1, characterized in that: The specific steps of obtaining the heart sound signal data of children with heart disease are: Collect original data of heart sound signals of children with heart disease through sensor equipment; Cleaning and denoising the original data of the heart sound signal of children with heart disease to obtain the heart sound signal data of children with heart disease; The heart sound signal data of children with heart disease includes synchronization data of the heart sound signal of children with heart disease and quality data of the heart sound signal of children with heart disease.

3. The abnormality recognition system for heart sound signals of children with heart disease according to claim 1, characterized in that: The specific steps of comprehensively analyzing and obtaining the accuracy evaluation value of the heart sound signal of a child with heart disease are as follows: Preset standard values ​​for artifact detection rates, weighting factors for synchronicity assessment values ​​of pediatric heart sound signals, weighting factors for artifact detection rates, and weighting factors for quality assessment values ​​of pediatric heart sound signals obtained from a pediatric heart sound signal database; The synchronization evaluation value of the heart sound signal of the child with heart disease is averaged, and the weight factor of the synchronization evaluation value of the heart sound signal of the child with heart disease is used to correct the result of the averaged processing, which is recorded as the first component of the accuracy evaluation value of the heart sound signal of the child with heart disease; The deviation between the artifact detection rate and the standard value of the artifact detection rate was analyzed with respect to the standard value of the artifact detection rate. The results of the proportion analysis were averaged and the weight factor of the artifact detection rate was used to correct the averaged results. The results were recorded as the second component of the accuracy assessment value of the heart sound signal for children with heart disease. The pediatric heart sound signal quality assessment value is averaged, and the result of the averaged processing is corrected using the weight factor of the pediatric heart sound signal quality assessment value, which is recorded as the third component of the pediatric heart sound signal accuracy assessment value; The first component of the accuracy assessment value of the heart sound signal of children with heart disease is analyzed with the coupling results of the second component of the accuracy assessment value of the heart sound signal of children with heart disease and the third component of the accuracy assessment value of the heart sound signal of children with heart disease, and the accuracy assessment value of the heart sound signal of children with heart disease is obtained through comprehensive analysis; The heart sound signal accuracy assessment value for pediatric heart disease represents quantitative data of the heart sound signal accuracy.

4. The abnormality recognition system for heart sound signals of children with heart disease according to claim 1, characterized in that: The specific steps of adjusting the abnormal recognition of heart sound signals of children with heart disease are: Obtaining from a pediatric heart sound signal database a first threshold value for evaluating the synchronization of a pediatric heart sound signal, a second threshold value for evaluating the quality of a pediatric heart sound signal, and a comprehensive threshold value for evaluating the accuracy of a pediatric heart sound signal; The synchronization evaluation value of the heart sound signal of children with heart disease, the quality evaluation value of the heart sound signal of children with heart disease and the accuracy evaluation value of the heart sound signal of children with heart disease were compared and analyzed with the first threshold of the synchronization evaluation value of the heart sound signal of children with heart disease, the second threshold of the quality evaluation value of the heart sound signal of children with heart disease and the comprehensive threshold of the accuracy evaluation value of the heart sound signal of children with heart disease, and the optimization and adjustment method of abnormal recognition of the heart sound signal of children with heart disease was obtained; The method for optimizing and adjusting abnormal identification of heart sound signals of children with heart disease includes a method for optimizing synchronization of heart sound signals of children with heart disease, a method for optimizing quality of heart sound signals of children with heart disease, and a method for optimizing accuracy of heart sound signals of children with heart disease.

5. The abnormality recognition system for heart sound signals of children with heart disease according to claim 4, characterized in that: The specific steps of the method for optimizing the synchronization of heart sound signals in children with heart disease are as follows: If the pediatric heart disease heart sound signal synchronization evaluation value is greater than or equal to the pediatric heart disease heart sound signal synchronization evaluation value first threshold, then the pediatric heart disease heart sound signal synchronization optimization method does not need to be performed; If the synchronization evaluation value of the heart sound signal of a child with heart disease is lower than the first threshold value of the synchronization evaluation value of the heart sound signal of a child with heart disease, the transmission path is shortened by reducing the number of wireless relay nodes and adopting a direct connection topology, and a PTP precision clock synchronization chip is deployed to ensure sampling rate synchronization, and a Kalman filter is used to compensate for delays, and the blind source separation algorithm is used to separate and suppress noise.

6. The abnormality recognition system for heart sound signals of children with heart disease according to claim 4, characterized in that: The specific steps of the method for optimizing the quality of heart sound signals of children with heart disease are as follows: If the pediatric heart disease heart sound signal quality assessment value is lower than or equal to the pediatric heart disease heart sound signal quality assessment value second threshold, then there is no need to perform the pediatric heart disease heart sound signal quality optimization method; If the heart sound signal quality assessment value of a child with heart disease is greater than the second threshold value of the heart sound signal quality assessment value of a child with heart disease, the data of the acceleration sensor and the heart sound sensor are combined through multi-sensor fusion technology, and a two-level buffering mechanism is implemented.

7. The abnormality recognition system for heart sound signals of children with heart disease according to claim 4, characterized in that: The specific steps of the method for optimizing the accuracy of heart sound signals of children with heart disease are as follows: If the pediatric heart disease heart sound signal accuracy assessment value is greater than or equal to the pediatric heart disease heart sound signal accuracy assessment value comprehensive threshold, then the pediatric heart disease heart sound signal accuracy optimization method does not need to be performed; If the accuracy assessment value of the heart sound signal of pediatric heart disease is lower than the comprehensive threshold of the accuracy assessment value of the heart sound signal of pediatric heart disease, caches of different time lengths are set on the device side, node side and cloud side respectively. When a fluctuation in the heart sound signal is detected, data is immediately obtained from the local cache. When an error occurs in the data on the device side or node side, the correct data is obtained from the cloud cache for replacement.

Images