Blood pressure monitoring method, device and equipment
By using the blood pressure monitoring model trained by CNN neural network, combined with the multi-dimensional health index data set, we find out the mapping relationship between blood oxygen data and blood pressure data, and solve the accuracy and stability problems caused by relying on a single parameter in the existing technology of blood pressure monitoring. , achieving more accurate blood pressure monitoring.
Patent Information
- Application Number
- CN202411468172.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-21
- Publication Date
- 2025-02-11
AI Technical Summary
Most existing blood pressure monitoring technologies rely on a single physiological parameter, such as pulse wave conduction time, and their accuracy and stability are affected by a variety of factors, resulting in inaccurate blood pressure monitoring results.
By obtaining the multi-dimensional health index data set, the blood pressure monitoring model is trained using CNN neural network to find the mapping relationship between blood oxygen data and blood pressure data. Relying on this mapping relationship, more accurate user blood pressure data can be obtained through blood oxygen data.
It is achieved to avoid the problem of inaccurate blood pressure monitoring results caused by environmental interference in the prior art due to the stability of a single parameter, and provide more accurate and stable blood pressure monitoring results.
Smart Images
Figure CN119405286A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a blood pressure monitoring method, device and equipment. Background Art
[0002] As people pay more and more attention to health management, blood pressure, as an important indicator for assessing cardiovascular health, is particularly important for continuous and accurate monitoring. Traditional blood pressure measurement methods, such as cuff-type sphygmomanometers, are accurate but cumbersome to operate and cannot achieve continuous monitoring. In recent years, with the development of wearable devices and Internet of Things technology, non-invasive and continuous blood pressure monitoring has become possible.
[0003] However, most existing blood pressure monitoring technologies rely on a single physiological parameter, such as pulse wave transit time, whose accuracy and stability are affected by many factors, such as individual differences, environmental changes, exercise status, etc. In addition, for patients with hypertension or cardiovascular disease who require long-term monitoring, frequent hospital visits and manual measurements not only increase the burden on patients, but may also miss important health information due to inappropriate measurement timing.
[0004] In view of this, the inventor proposes a blood pressure monitoring method, device and equipment. Summary of the invention
[0005] The embodiments of the present application provide a blood pressure monitoring method, device and equipment to solve the technical problem that the existing blood pressure monitoring technology mostly relies on a single parameter such as pulse physiological parameters, and the stability is too easily disturbed by other environments, resulting in inaccurate blood pressure monitoring results.
[0006] A first aspect of an embodiment of the present application provides a blood pressure monitoring method, the method comprising:
[0007] Obtain multi-dimensional health indicator datasets;
[0008] Adjusting the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model;
[0009] Obtain user blood oxygen monitoring data;
[0010] The blood pressure monitoring model is used to analyze and process the user's blood oxygen monitoring data to obtain the user's blood pressure monitoring result;
[0011] The blood pressure monitoring result of the user is displayed.
[0012] Furthermore, the initial CNN neural network model is adjusted according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model, including:
[0013] Extracting blood oxygen data and blood pressure data corresponding to the blood oxygen data from a comprehensive health monitoring data set of a user to obtain a blood oxygen and blood pressure data set;
[0014] Performing data preprocessing on the blood oxygen and blood pressure data set to obtain a blood oxygen and blood pressure data set after data preprocessing;
[0015] Dividing the preprocessed blood oxygen and blood pressure data set into a training set, a validation set, and a test set;
[0016] The blood pressure monitoring model is trained using a CNN neural network according to the training set, validation set and test set until the model converges to obtain a trained blood pressure monitoring model.
[0017] Further, the blood oxygen and blood pressure data set is preprocessed to obtain the blood oxygen and blood pressure data set after data preprocessing, including:
[0018] Cleaning the blood oxygen and blood pressure data set to obtain a cleaned blood oxygen and blood pressure data set;
[0019] Dividing the time series data of the blood oxygen data in the cleaned blood oxygen and blood pressure data set into windows of fixed length to obtain windowed time series blood oxygen data;
[0020] Normalizing the windowed time series blood oxygen data to ensure that the windowed time series blood oxygen data of different samples have the same scale;
[0021] Aligning the cleaned blood oxygen and blood pressure data set blood pressure data with the windowed time series blood oxygen data to obtain blood pressure data of the aligned blood oxygen data;
[0022] The windowed time series blood oxygen data and the blood pressure data of the aligned blood oxygen data are combined to obtain a blood oxygen and blood pressure data set after data preprocessing.
[0023] Furthermore, the blood oxygen and blood pressure data set is cleaned to obtain a cleaned blood oxygen and blood pressure data set, including:
[0024] Identify missing values and outliers in the blood oxygen and blood pressure dataset;
[0025] Interpolation method is used to correct the missing data values and the abnormal values are directly eliminated;
[0026] The corrected data replaces the missing data values in the original data set to obtain a cleaned blood oxygen and blood pressure data set.
[0027] Furthermore, the CNN neural network is used to train the blood pressure monitoring model according to the training set, the validation set and the test set until the model converges, thereby obtaining a trained blood pressure monitoring model, including:
[0028] The blood oxygen data window in the training set is used as input, and forward propagation is performed through the CNN neural network to calculate the predicted blood pressure value;
[0029] Use the loss function to calculate the loss value between the predicted blood pressure value and the true blood pressure value;
[0030] According to the loss value, the gradient of the model parameters is calculated by a back propagation algorithm;
[0031] Using an optimizer to update model parameters according to the gradient;
[0032] Repeat the above-mentioned blood oxygen data window in the training set as input, perform forward propagation through the CNN neural network, calculate the predicted blood pressure value, and use the optimizer to update the model parameters according to the gradient until the gradient update model parameters converge to obtain a trained blood pressure monitoring model.
[0033] Furthermore, a loss function is used to calculate a loss value between the predicted blood pressure value and the actual blood pressure value, including: the loss function adopts a mean absolute error method.
[0034] Furthermore, an optimizer is used to update model parameters according to the gradient, including: the optimizer uses a stochastic gradient descent method as an optimization function.
[0035] A second aspect of an embodiment of the present application provides a blood pressure monitoring device, the device comprising:
[0036] A first acquisition unit, used to acquire a multi-dimensional health indicator data set;
[0037] A second acquisition unit, used to acquire the user's blood oxygen monitoring data;
[0038] A first processing unit, configured to adjust the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model;
[0039] The second processing unit is used to analyze and process the user's blood oxygen data using a blood pressure monitoring model to obtain a blood pressure monitoring result of the user;
[0040] The first display unit is used to display the blood pressure monitoring result of the user.
[0041] A third aspect of an embodiment of the present application provides an electronic device, comprising a processor, an input device, an output device and a memory, wherein the processor, input device, output device and memory are interconnected, wherein the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to execute the blood pressure monitoring method as described in any one of the first aspects.
[0042] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the blood pressure monitoring method as described in any one of the first aspects.
[0043] Beneficial effects:
[0044] The present invention proposes a blood pressure monitoring method, which first obtains a multi-dimensional health indicator data set, uses a CNN neural network to train a blood pressure monitoring model, and uses the CNN neural network to find out the mapping relationship between blood oxygen data and blood pressure data. Based on this mapping relationship, relatively accurate user blood pressure data can be obtained through blood oxygen data, and then the user's blood oxygen data can be collected and analyzed by the blood pressure monitoring model to obtain the user's blood pressure monitoring result, so as to avoid the technical problem that most existing blood pressure monitoring technologies rely on a single parameter such as pulse physiological parameters, the stability of which is too easily disturbed by other environments, and the blood pressure monitoring result is inaccurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0046] Figure 1 A schematic diagram of the overall process of a blood pressure monitoring method is provided for an embodiment of the present application;
[0047] Figure 2 A schematic diagram of the training process of a blood pressure monitoring model is provided for an embodiment of the present application;
[0048] Figure 3 A schematic diagram of the data preprocessing process of a blood pressure monitoring model is provided for an embodiment of the present application;
[0049] Figure 4 A schematic diagram of a data cleaning process of a blood pressure monitoring model is provided for an embodiment of the present application;
[0050] Figure 5 A schematic diagram of a model training process of a blood pressure monitoring model is provided for an embodiment of the present application;
[0051] Figure 6 A structural schematic diagram of a blood pressure monitoring device is provided for an embodiment of the present application;
[0052] Figure 7 A schematic diagram of the structure of an electronic device is provided for an embodiment of the present application;
[0053] Reference numerals:
[0054] First acquisition unit-001, second acquisition unit-002, first processing unit-003, second processing unit-004, first display unit-005. DETAILED DESCRIPTION
[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0056] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0057] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.
[0058] In order to better understand a blood pressure monitoring method provided in an embodiment of the present application, the following first briefly introduces the scenario of applying the business product determination method.
[0059] Since most existing blood pressure monitoring technologies rely on a single physiological parameter, such as pulse wave transmission time, their accuracy and stability are affected by many factors, such as individual differences, environmental changes, exercise status, etc. Specifically, the application scenario of this embodiment is to use a sports bracelet worn on the user's wrist to monitor the user's blood pressure. Since most existing bracelets directly measure the pulse signal through an optical sensor, the blood pressure information of the current user is obtained from the pulse signal. However, the user's blood pressure measured in this way is inaccurate, because blood pressure changes over time, especially after exercise, emotional excitement or taking medicine, its fluctuation is more significant. Optical sensors can usually only provide relatively stable measurement data under static or low-dynamic conditions, and it is difficult to accurately capture the instantaneous changes in blood pressure. Therefore, it is impossible to directly measure relatively accurate user blood pressure data in this application scenario.
[0060] In summary, in order to solve the technical problems that the existing blood pressure monitoring technologies in the prior art mostly rely on a single parameter such as pulse physiological parameters, the stability is too easily disturbed by other environments, and the blood pressure monitoring results are inaccurate, such as Figure 1 As shown, this embodiment provides a blood pressure monitoring method, including:
[0061] S1. Obtain a multidimensional health indicator data set, wherein the multidimensional health indicator data set includes: heart rate, age, gender, weight, height, exercise, eating habits, sleep quality and drug use information. Since the multidimensional health indicator data set contains too much data and the processing task is too complicated, it is necessary to screen and extract the information therein, and the heart rate data, which is the factor that has the greatest impact on blood pressure, is used.
[0062] S2. Adjust the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model. This embodiment is preferred, such as Figure 2 As shown, including:
[0063] S201. Extract blood oxygen data and blood pressure data corresponding to the blood oxygen data from the user health monitoring comprehensive data set to obtain a blood oxygen and blood pressure data set. This step provides the most original training materials for the subsequent blood pressure monitoring model and provides basic conditions for training.
[0064] S202, preprocessing the blood oxygen and blood pressure data set to obtain a preprocessed blood oxygen and blood pressure data set, as a preferred solution of this embodiment, such as Figure 3 As shown, the specific steps include:
[0065] S2021. Clean the blood oxygen and blood pressure data set to obtain a cleaned blood oxygen and blood pressure data set. Specifically, Figure 4 As shown, including:
[0066] S20211. Identify missing values and outliers in the blood oxygen and blood pressure dataset.
[0067] S20212. Use interpolation method to correct the missing data values and directly eliminate the abnormal values.
[0068] S20213. Replace the missing data values in the original data set with the corrected data to obtain a cleaned blood oxygen and blood pressure data set. Through the above data cleaning steps, a complete blood oxygen and blood pressure data set can be obtained without missing or abnormal values, thereby increasing the accuracy of subsequent model training.
[0069] S2022. Divide the time series data of the blood oxygen data in the cleaned blood oxygen and blood pressure data set into windows of fixed length to obtain windowed time series blood oxygen data. By dividing the long time series into windows of fixed length, the subsequent processing steps can be simplified so that each window can be processed independently or in parallel, thereby improving the overall processing efficiency. For the CNN neural network, the input data is required to have a fixed dimension or length, which can ensure that the input blood oxygen data meets the requirements of the input features.
[0070] S2023. Normalize the windowed time series blood oxygen data to ensure that the windowed time series blood oxygen data of different samples have the same scale. Normalization ensures that the windowed time series blood oxygen data of different samples are compared on the same scale, thereby reducing the deviation caused by different data ranges.
[0071] S2024. Align the blood pressure data in the cleaned blood oxygen and blood pressure data set with the windowed time series blood oxygen data to obtain blood pressure data of the aligned blood oxygen data. The alignment operation ensures the temporal consistency of the blood oxygen data and the blood pressure data, providing an accurate data basis for subsequent data analysis and model training. Moreover, the aligned data enables the two physiological indicators of blood oxygen and blood pressure to be jointly analyzed, which helps to reveal the potential relationship or mutual influence between them.
[0072] S2025. Merge the windowed time series blood oxygen data and the blood pressure data of the aligned blood oxygen data to obtain a blood oxygen and blood pressure data set after data preprocessing. By merging the data, the number of data sets that need to be processed in subsequent processing steps is reduced to simplify the processing flow.
[0073] In summary, steps S2021-S2025 together constitute an efficient and accurate data preprocessing process, which provides a solid foundation for subsequent data analysis and application and is highly practical.
[0074] S203, dividing the preprocessed blood oxygen and blood pressure data set into a training set, a validation set and a test set;
[0075] S204, training the blood pressure monitoring model using a CNN neural network according to the training set, the validation set and the test set until the model converges, thereby obtaining a trained blood pressure monitoring model, such as Figure 5 Specifically, the following steps are included:
[0076] S2041. The blood oxygen data window in the training set is used as input, and forward propagation is performed through the CNN neural network to calculate the predicted blood pressure value. Specifically, the CNN neural network is good at processing data with spatial or temporal correlation. Therefore, through the forward propagation of the CNN neural network, the network can capture the features in the blood oxygen data and predict the blood pressure value based on these features, thereby realizing the automatic prediction from blood oxygen data to blood pressure value and improving the prediction efficiency and accuracy.
[0077] S2042. Use a loss function to calculate the loss value between the predicted blood pressure value and the actual blood pressure value. The loss function adopts the mean absolute error method. This embodiment is preferred. The loss function provides an indicator to quantify the prediction error, which helps to evaluate the prediction performance of the model. At the same time, the mean absolute error is used as a loss function, which can guide the model to optimize in the direction of reducing the prediction error during the training process.
[0078] S2043. According to the loss value, the gradient of the model parameters is calculated by the back propagation algorithm. Specifically, by calculating the gradient of the loss value with respect to the model parameters, a direction is provided for parameter updating. The back propagation algorithm can be used to efficiently calculate the gradient to optimize the model parameters.
[0079] S2044. Use an optimizer to update the model parameters according to the gradient. The optimizer uses stochastic gradient descent as an optimization function. This embodiment is preferred. According to the gradient information, the optimizer can adjust the model parameters to reduce the loss value, thereby improving the prediction performance of the model. As a commonly used optimization function, stochastic gradient descent has the advantages of simple implementation and strong adaptability, and can effectively cope with the training of large-scale data sets.
[0080] S2045. Repeat the above-mentioned blood oxygen data window in the training set as input, perform forward propagation through the CNN neural network, calculate the predicted blood pressure value, and use the optimizer to update the model parameters according to the gradient until the gradient update model parameters converge to obtain a trained blood pressure monitoring model. Through multiple iterative training, the model parameters gradually converge to a stable state. At this time, the prediction performance of the model reaches the optimal state, thereby improving the generalization ability and prediction accuracy of the blood pressure monitoring model.
[0081] S3. Obtain the user's blood oxygen monitoring data. Specifically, in this embodiment, the application scenario is to collect the user's blood oxygen monitoring data through a blood oxygen monitoring device. This embodiment preferably uses an optical sensor to monitor the user's blood oxygen to obtain the user's blood oxygen monitoring data.
[0082] S4. Use the blood pressure monitoring model to analyze and process the user's blood oxygen monitoring data to obtain the user's blood pressure monitoring results. As a preferred solution of this embodiment, after obtaining the user's blood pressure monitoring results, upload them to the cloud server via wireless transmission, so that the user's blood pressure data is stored in the cloud and saved in the historical record.
[0083] S5. Display the blood pressure monitoring result of the user, that is, display it through the first display unit of the blood pressure monitoring. The user can see his or her blood pressure data indicators in real time through the first display unit. In addition, even if the user replaces the blood pressure monitoring device, the blood pressure data previously stored in the cloud server can be obtained from the cloud server.
[0084] That is, this embodiment provides a blood pressure monitoring method, which first obtains a multi-dimensional health indicator data set, uses a CNN neural network to train a blood pressure monitoring model, and uses the CNN neural network to find out the mapping relationship between blood oxygen data and blood pressure data. Based on this mapping relationship, relatively accurate user blood pressure data can be obtained through blood oxygen data. Then, by collecting the user's blood oxygen data, the user's blood pressure monitoring result can be obtained through analysis and processing by the blood pressure monitoring model, so as to avoid the technical problem that most existing blood pressure monitoring technologies rely on a single parameter such as pulse physiological parameters, the stability of which is too easily disturbed by other environments, and the blood pressure monitoring results are inaccurate.
[0085] For the above embodiments, please refer to Figure 7 , Figure 7 A schematic diagram of the structure of a blood pressure monitoring device provided in an embodiment of the present application, as shown in the figure, includes a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, the processor is configured to call the program instructions, and the program includes instructions for executing the following steps;
[0086] Obtain multi-dimensional health indicator datasets;
[0087] Adjusting the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model;
[0088] Obtain user blood oxygen monitoring data;
[0089] The blood pressure monitoring model is used to analyze and process the user's blood oxygen monitoring data to obtain the user's blood pressure monitoring result;
[0090] The blood pressure monitoring result of the user is displayed.
[0091] In this example, a multi-dimensional health indicator data set is obtained, and a CNN neural network is used to train a blood pressure monitoring model. The mapping relationship between blood oxygen data and blood pressure data is found through the CNN neural network. Based on this mapping relationship, more accurate user blood pressure data can be obtained through blood oxygen data. Then, by collecting the user's blood oxygen data, the blood pressure monitoring result of the user can be obtained through analysis and processing by the blood pressure monitoring model, so as to avoid the technical problem that most existing blood pressure monitoring technologies rely on a single parameter such as pulse physiological parameters, and the stability is too easily disturbed by other environments, resulting in inaccurate blood pressure monitoring results.
[0092] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that in order to realize the above functions, the terminal includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0093] The embodiment of the present application can divide the electronic device into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.
[0094] In line with the above, see Figure 6 , Figure 6 The present invention provides a schematic diagram of the structure of a blood pressure monitoring device. Figure 6 As shown, the device comprises:
[0095] The first acquisition unit 001 is used to acquire a multi-dimensional health indicator data set;
[0096] The second acquisition unit 002 is used to acquire the user's blood oxygen monitoring data;
[0097] The first processing unit 003 is used to adjust the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model;
[0098] The second processing unit 004 is used to analyze and process the user's blood oxygen data using a blood pressure monitoring model to obtain a blood pressure monitoring result of the user;
[0099] The first display unit 005 is used to display the blood pressure monitoring result of the user.
[0100] The first processing unit 003 is further used for:
[0101] Extracting blood oxygen data and blood pressure data corresponding to the blood oxygen data from a comprehensive health monitoring data set of a user to obtain a blood oxygen and blood pressure data set;
[0102] Performing data preprocessing on the blood oxygen and blood pressure data set to obtain a blood oxygen and blood pressure data set after data preprocessing;
[0103] Dividing the preprocessed blood oxygen and blood pressure data set into a training set, a validation set, and a test set;
[0104] The blood pressure monitoring model is trained using a CNN neural network according to the training set, validation set and test set until the model converges to obtain a trained blood pressure monitoring model.
[0105] In a possible implementation, the first processing unit 003 is further configured to:
[0106] Cleaning the blood oxygen and blood pressure data set to obtain a cleaned blood oxygen and blood pressure data set;
[0107] Dividing the time series data of the blood oxygen data in the cleaned blood oxygen and blood pressure data set into windows of fixed length to obtain windowed time series blood oxygen data;
[0108] Normalizing the windowed time series blood oxygen data to ensure that the windowed time series blood oxygen data of different samples have the same scale;
[0109] Aligning the cleaned blood oxygen and blood pressure data set blood pressure data with the windowed time series blood oxygen data to obtain blood pressure data of the aligned blood oxygen data;
[0110] The windowed time series blood oxygen data and the blood pressure data of the aligned blood oxygen data are combined to obtain a blood oxygen and blood pressure data set after data preprocessing.
[0111] In a possible implementation, the first processing unit 003 is further configured to:
[0112] Identify missing values and outliers in the blood oxygen and blood pressure dataset;
[0113] Interpolation method is used to correct the missing data values and the abnormal values are directly eliminated;
[0114] The corrected data replaces the missing data values in the original data set to obtain a cleaned blood oxygen and blood pressure data set.
[0115] In a possible implementation, the first processing unit 003 is further configured to:
[0116] The blood oxygen data window in the training set is used as input, and forward propagation is performed through the CNN neural network to calculate the predicted blood pressure value;
[0117] Use the loss function to calculate the loss value between the predicted blood pressure value and the true blood pressure value;
[0118] According to the loss value, the gradient of the model parameters is calculated by a back propagation algorithm;
[0119] Using an optimizer to update model parameters according to the gradient;
[0120] Repeat the above-mentioned blood oxygen data window in the training set as input, perform forward propagation through the CNN neural network, calculate the predicted blood pressure value, and use the optimizer to update the model parameters according to the gradient until the gradient update model parameters converge to obtain a trained blood pressure monitoring model.
[0121] In a possible implementation, the second processing unit 004 is used to:
[0122] The blood pressure monitoring model is used to analyze and process the user's blood oxygen monitoring data to obtain the user's blood pressure monitoring results.
[0123] In a possible implementation, the first display unit 005 is used to display the blood pressure monitoring result of the user.
[0124] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any blood pressure monitoring method recorded in the above method embodiments.
[0125] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute part or all of the steps of any blood pressure monitoring method recorded in the above method embodiments.
[0126] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0127] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0128] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.
[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0130] In addition, the functional units in the various embodiments of the application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software program modules.
[0131] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, disk or optical disk and other media that can store program codes.
[0132] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which can include: a flash drive, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.
[0133] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of the present application. At the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
Claims
1. A blood pressure monitoring method, applied to a blood pressure monitoring device, characterized in that: include: Obtain multi-dimensional health indicator datasets; Adjusting the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model; Obtain user blood oxygen monitoring data; The blood pressure monitoring model is used to analyze and process the user's blood oxygen monitoring data to obtain the user's blood pressure monitoring result; The user's blood pressure monitoring results are displayed.
2. The blood pressure monitoring method according to claim 1, characterized in that: The initial CNN neural network model is adjusted according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model, including: Extracting blood oxygen data and blood pressure data corresponding to the blood oxygen data from a comprehensive health monitoring data set of a user to obtain a blood oxygen and blood pressure data set; Performing data preprocessing on the blood oxygen and blood pressure data set to obtain a blood oxygen and blood pressure data set after data preprocessing; Dividing the preprocessed blood oxygen and blood pressure data set into a training set, a validation set, and a test set; The blood pressure monitoring model is trained using a CNN neural network according to the training set, validation set and test set until the model converges to obtain a trained blood pressure monitoring model.
3. The blood pressure monitoring method according to claim 2, characterized in that: Performing data preprocessing on the blood oxygen and blood pressure data set to obtain a blood oxygen and blood pressure data set after data preprocessing, including: Cleaning the blood oxygen and blood pressure data set to obtain a cleaned blood oxygen and blood pressure data set; Dividing the time series data of the blood oxygen data in the cleaned blood oxygen and blood pressure data set into windows of fixed length to obtain windowed time series blood oxygen data; Normalizing the windowed time series blood oxygen data to ensure that the windowed time series blood oxygen data of different samples have the same scale; Aligning the cleaned blood oxygen and blood pressure data set blood pressure data with the windowed time series blood oxygen data to obtain blood pressure data aligned with the blood oxygen data; The windowed time series blood oxygen data and the blood pressure data of the aligned blood oxygen data are combined to obtain a blood oxygen and blood pressure data set after data preprocessing.
4. The blood pressure monitoring method according to claim 3, characterized in that: The blood oxygen and blood pressure data set is cleaned to obtain the cleaned blood oxygen and blood pressure data set, including: Identify missing values and outliers in the blood oxygen and blood pressure dataset; Interpolation method is used to correct the missing data values and the abnormal values are directly eliminated; The corrected data replaces the missing data values in the original data set to obtain a cleaned blood oxygen and blood pressure data set.
5. The blood pressure monitoring method according to claim 2, characterized in that: The blood pressure monitoring model is trained using a CNN neural network according to the training set, validation set, and test set until the model converges to obtain a trained blood pressure monitoring model, including: The blood oxygen data window in the training set is used as input, and forward propagation is performed through the CNN neural network to calculate the predicted blood pressure value; Use the loss function to calculate the loss value between the predicted blood pressure value and the true blood pressure value; According to the loss value, the gradient of the model parameters is calculated by a back propagation algorithm; Using an optimizer to update model parameters according to the gradient; Repeat the above-mentioned blood oxygen data window in the training set as input, perform forward propagation through the CNN neural network, calculate the predicted blood pressure value, and use the optimizer to update the model parameters according to the gradient until the gradient update model parameters converge to obtain a trained blood pressure monitoring model.
6. The blood pressure monitoring method according to claim 5, characterized in that: The loss function is used to calculate the loss value between the predicted blood pressure value and the actual blood pressure value, including: the loss function adopts the mean absolute error method.
7. The blood pressure monitoring method according to claim 5, characterized in that: Adopting an optimizer to update the model parameters according to the gradient includes: the optimizer adopts a stochastic gradient descent method as an optimization function.
8. A blood pressure monitoring device, characterized in that: The device comprises: A first acquisition unit, used to acquire a multi-dimensional health indicator data set; A second acquisition unit, used to acquire the user's blood oxygen monitoring data; A first processing unit, configured to adjust the initial CNN neural network model according to the multi-dimensional health indicator data set to obtain a blood pressure monitoring model; The second processing unit is used to analyze and process the user's blood oxygen data using a blood pressure monitoring model to obtain a blood pressure monitoring result of the user; The first display unit is used to display the blood pressure monitoring result of the user.
9. An electronic device, characterized in that: It includes a processor, an input device, an output device and a memory, which are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the blood pressure monitoring method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the blood pressure monitoring method according to any one of claims 1 to 7.