Blood pressure detection method, device, equipment and medium

Abstract

The invention provides a blood pressure detection method, device and equipment and a medium, and belongs to the technical field of medical detection. The blood pressure detection method comprises the following steps: performing similarity comparison on blood pressure data corresponding to the same detection part in a plurality of detection parts of a detection object at different moments, and performing similarity comparison on blood pressure data of at least two different detection parts in the plurality of detection parts of the detection object at the same moment; whether the blood pressure of the detected object is abnormal at the current moment is determined. According to the blood pressure detection method, the blood pressure abnormity judgment result of the detected object is obtained by comparing the blood pressure of the same part of the detected object at different moments and comparing the blood pressure of different parts at the same moment, a preset standard does not need to be depended on, the detection result accuracy is high, and the detection efficiency is high. And the phenomenon of blood pressure abnormal missing detection or wrong detection is not easy to occur.

Description

Technical Field

[0001] The present application relates to the field of medical detection technology, and in particular to a blood pressure detection method, device, equipment and medium. Background Art

[0002] In modern society, due to the combined effects of risk factors such as unreasonable dietary structure and work and rest time, lack of exercise, smoking and drinking, the incidence of blood pressure-related diseases such as chronic cardiovascular and hypertension continues to rise, while the age of patients is gradually decreasing. Blood pressure-related diseases pose an increasing threat to human health, and its prevention and treatment have always been a concern of people. Therefore, accurate detection and timely detection of abnormal changes in blood pressure are crucial for the prevention and treatment of diseases.

[0003] The existing blood pressure detection method usually includes: presetting the judgment standard of normal blood pressure, then collecting the blood pressure data of the test object, and judging whether the test object has abnormal blood pressure according to the pre-set judgment conditions. For example, an existing specific implementation method of blood pressure detection includes: using a pre-trained time series model to predict the blood pressure trend after the current moment based on the blood pressure data in a period of time before the current moment, so as to confirm whether the test object will have abnormal blood pressure. The pre-trained time series model is obtained by comparing the classification result of the classifier or with the blood pressure threshold. The category of the classifier and the blood pressure threshold are both pre-set judgment conditions. However, different people may have different physical conditions, past medical history and other factors, and the corresponding blood pressure characteristics are also different. Therefore, the existing blood pressure detection method is prone to missed or wrong detection of abnormal blood pressure, that is, the accuracy of abnormal blood pressure detection is low, which is not conducive to the prevention of blood pressure-related diseases. Summary of the invention

[0004] In order to solve the existing technical problems, the present application provides a blood pressure detection method, device, equipment and computer-readable storage medium with high abnormality detection accuracy.

[0005] According to a first aspect of an embodiment of the present application, a blood pressure detection method is provided, comprising: Compare the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment, and determine the first similarity between the current blood pressure data and the previous blood pressure data of the same detection part; and comparing the current blood pressure data corresponding to different detection parts of the detection object at the current moment, and determining a second similarity between the current blood pressure data of the different detection parts; It is determined whether the detection object has abnormal blood pressure at the current moment according to the first similarity and the second similarity.

[0006] According to a second aspect of an embodiment of the present application, a blood pressure detection device is provided, comprising: A first determination module, configured to compare current blood pressure data corresponding to the same detection site of the detection subject at the current moment with previous blood pressure data corresponding to the previous moment, and determine a first similarity between the current blood pressure data and the previous blood pressure data of the same detection site; A second determination module, configured to compare the current blood pressure data corresponding to different detection parts of the detection object at the current moment, and determine a second similarity between the current blood pressure data of the different detection parts; A third determination module is configured to determine whether the detection object has abnormal blood pressure at the current moment according to the first similarity and the second similarity.

[0007] According to a third aspect of an embodiment of the present application, a blood pressure detection device is provided, including a memory and a processor, wherein the memory stores a computer-readable program, and the processor implements the blood pressure detection method when executing the computer-readable program.

[0008] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the blood pressure detection method is implemented.

[0009] As can be seen from the above, the blood pressure detection method provided in the embodiment of the present application obtains a first similarity characterizing the blood pressure changes at different moments of the same detection part by performing a similarity comparison on the blood pressure data corresponding to the same detection part of the multiple detection parts of the detection object at different moments, and obtains a second similarity characterizing the blood pressure changes at the same moment of different detection parts by performing a similarity comparison on the blood pressure data at the same moment of at least two different detection parts of the multiple detection parts of the detection object, thereby determining whether the detection object has abnormal blood pressure at the current moment through the first similarity and the second similarity. Therefore, the blood pressure detection method provided in the embodiment of the present application is always based on comparing the blood pressure of the same part of the detection object at different moments, and obtaining the abnormal blood pressure judgment result of the detection object based on comparing the blood pressure of different parts at the same moment, without relying on preset standards, and the detection result has high accuracy, and it is not easy to miss or misdetect abnormal blood pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings: Figure 1 A schematic diagram of a blood pressure detection method according to some embodiments of the present application; Figure 2 A schematic diagram of a process for determining a first similarity in a blood pressure detection method provided in some embodiments of the present application; Figure 3 A schematic diagram of a flow chart for determining a second similarity in a blood pressure detection method according to some embodiments of the present application; Figure 4 A schematic diagram of a calculation method for obtaining blood pressure data of the first detection site and the second detection site at each moment according to an embodiment of the present application; Figure 5 A schematic diagram of blood pressure time series data corresponding to each detection site in a blood pressure detection method provided according to an embodiment of the present application; Figure 6 A structural block diagram of a blood pressure detection device provided according to some embodiments of the present application; Figure 7 A schematic diagram of a specific implementation structure of a blood pressure detection device provided according to some embodiments of the present application; Figure 8 A structural block diagram of a blood pressure detection device provided according to some embodiments of the present application; Fig. 9 This is a schematic diagram of the appearance of a blood pressure detection device provided according to some embodiments of the present application. DETAILED DESCRIPTION

[0011] The technical solution of the present application is further elaborated in detail below in conjunction with the accompanying drawings and specific embodiments of the specification.

[0012] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit the implementation of this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0013] See also Figure 1 , which is a flow chart of a blood pressure detection method provided according to some embodiments of the present application. In some embodiments, the blood pressure detection method provided by the present application specifically includes S02, S04 and S06, and the specific description of each step is as follows.

[0014] S02: Compare the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment, and determine the first similarity between the current blood pressure data and the previous blood pressure data of the same detection part.

[0015] The detection object refers to the object for blood pressure detection using the blood pressure detection device implemented according to the blood pressure detection method provided in the embodiment of the present application, which may be a person or other organism. The detection part is the part of the detection object used to collect blood pressure data. For example, when the blood pressure detection device is a bracelet, the corresponding detection part may be any part of the bracelet that contacts the wrist.

[0016] Blood pressure data is data used to characterize the blood pressure value of the test object, that is, the value of the blood pressure data can be directly the blood pressure value of the test object, or it can be a value determined by the relationship between the blood pressure value of the test object. Based on the value of the blood pressure data, the blood pressure value of the test object can be determined. For example, the blood pressure data can be directly collected by the blood pressure sensor, or it can be data obtained by processing and converting the data collected by the blood pressure sensor.

[0017] The blood pressure detection method provided in the embodiment of the present application detects the blood pressure of the detection object based on the blood pressure data collected by the blood pressure sensor, at a certain detection frequency (different from the collection frequency of the blood pressure sensor), and at regular intervals, determines whether the detection object has abnormal blood pressure based on the blood pressure collection data collected at the current moment and / or before the current moment. The detection frequency is the judgment frequency of abnormal blood pressure. The current moment and each previous moment before the current moment are the division points of the detection frequency, that is, the moment in this application refers to the separation point of the detection frequency, and the time between two adjacent moments is the detection period corresponding to the detection frequency (the inverse of the detection frequency).

[0018] The current blood pressure data refers to the blood pressure data corresponding to the current moment, and the previous blood pressure data refers to the blood pressure data corresponding to the previous moment before the current moment. The previous moment can be the previous moment before the current moment and adjacent to the current moment, or it can be multiple moments before the current moment, or it can be all moments before the current moment.

[0019] The first similarity is used to characterize the similarity degree or similarity category between the current blood pressure data and the previous blood pressure data of the same detection site. The current blood pressure data and the previous blood pressure data can be directly compared to obtain the first similarity, or the current blood pressure data and the previous blood pressure data of the same detection site can be processed to a certain extent respectively, and then the processed data can be compared. For example, the first similarity can be obtained by comparing the feature extraction data of the current blood pressure data with the feature extraction data of the previous blood pressure data. In addition, in other embodiments, the current blood pressure data of the same detection site can be compared with the previous blood pressure data, or the current blood pressure data corresponding to the same detection site can be compared with the multiple blood pressure data before the current moment for the first similarity to obtain the first similarity, that is, the first similarity can be determined based on the comparison results of multiple first similarity comparisons.

[0020] The higher the similarity represented by the first similarity, the lower the change in the current blood pressure at the same detection site compared to the previous blood pressure, and the more it indicates that the current blood pressure of the detection subject is normal, and vice versa.

[0021] The similarity category of the first similarity may include but is not limited to similarity and dissimilarity. If similar, it means that the current blood pressure at the same detection site has not changed much compared with the previous blood pressure, which means that the current blood pressure of the detection object may be stable compared with the previous blood pressure, that is, it may be normal; if dissimilar, it means that the current blood pressure at the same detection site has changed significantly compared with the previous blood pressure, and the current blood pressure of the detection object may be unstable compared with the previous blood pressure, that is, it may be abnormal.

[0022] S04: Compare the current blood pressure data corresponding to different detection parts of the detection object at the current moment to determine the second similarity between the current blood pressure data of different detection parts.

[0023] Different detection sites include two or more detection sites, and the different detection sites here may include the same detection site in S02, or may be at least two other detection sites other than the same detection site. Therefore, based on the blood pressure detection method provided in the embodiment of the present application, the detection sites corresponding to the detection object are two or more, that is, there are multiple detection sites. One of the multiple detection sites is used to compare the blood pressure changes at different times, and at least two of the multiple detection sites are used to compare the blood pressure changes at different detection sites.

[0024] The second similarity is used to characterize the similarity degree or similarity category between the current blood pressure data of different detection parts. The current blood pressure data of different detection parts can be directly compared to obtain the second similarity, or the current blood pressure data of different detection parts can be processed to a certain extent respectively, and then the processed data can be compared. For example, the second similarity can be obtained by extracting data based on the features of the current blood pressure data of different detection parts and comparing them. In addition, in other embodiments, the comparison of the current blood pressure data of different detection parts can also be performed by performing a second similarity comparison between the current blood pressure data corresponding to three or more detection parts respectively to obtain the second similarity, that is, the second similarity can be determined based on the comparison results of multiple second similarity comparisons.

[0025] The higher the similarity represented by the second similarity, the lower the degree of variation between the current blood pressures at the different detection sites, and the more it indicates that the current blood pressure of the detection subject is normal, and vice versa.

[0026] The similarity categories of the second similarity level may include but are not limited to similarity and dissimilarity. If they are similar, it means that the current blood pressures at different detection sites do not vary much, indicating that the current blood pressures at different sites of the detection object are stable, that is, they may be normal; if they are dissimilar, it means that the current blood pressures at different detection sites vary greatly, that is, the current blood pressure of the detection object may be abnormal.

[0027] S06: Determine whether the detected object has abnormal blood pressure at the current moment according to the first similarity and the second similarity.

[0028] Based on the above analysis, it can be known that the first similarity is used to characterize the degree of change in blood pressure of the same part of the test object at different times, and the second similarity is used to characterize the degree of change in blood pressure between different parts of the test object at the same time. Therefore, based on the first similarity and the second similarity corresponding to the current moment, the blood pressure state of the test object at the current moment can be determined, such as determining whether the blood pressure of the test object at the current moment is abnormal.

[0029] The blood pressure detection method provided in the embodiment of the present application compares the blood pressure data corresponding to the same detection part of the multiple detection parts of the detection object at different times to obtain a first similarity that characterizes the blood pressure changes of the same detection part at different times, and compares the blood pressure data of at least two different detection parts of the multiple detection parts of the detection object at the same time to obtain a second similarity that characterizes the blood pressure changes of different detection parts at the same time, thereby determining whether the detection object has abnormal blood pressure at the current moment through the first similarity and the second similarity. Therefore, the blood pressure detection method provided in the embodiment of the present application is always based on the comparison of the blood pressure of the same part of the detection object at different times, and based on the comparison of the blood pressure of different parts at the same time, to obtain the abnormal blood pressure judgment result of the detection object, without relying on the preset standard, and the detection result has high accuracy, and it is not easy to miss or misdetect abnormal blood pressure.

[0030] See also Figure 2 As shown, it is a flow chart of determining the first similarity in the blood pressure detection method provided in some embodiments of the present application. Specifically, in some embodiments, S02 specifically includes S022 and S024, and the descriptions of each step are as follows.

[0031] S022: Inputting the first current blood pressure time series data corresponding to the same detection site at the current moment into the first feature extraction network model for feature extraction to obtain first feature data and second feature data.

[0032] The first current blood pressure time series data is the time series data corresponding to the current blood pressure data of the same detection site as the last data. It is one-dimensional data arranged in time series by the current blood pressure data corresponding to the same detection site and the previous blood pressure data corresponding to multiple moments before the current moment.

[0033] The first feature extraction network model is a neural network model that is trained based on the blood pressure data set to learn the feature representation of the blood pressure data. The trained neural network model can extract features from the input blood pressure data to obtain blood pressure feature data.

[0034] Specifically, in some embodiments, the first feature extraction network model can be specifically a long short-term memory network (LSTM) model after training the blood pressure data set. The long short-term memory network model has a feature extraction layer corresponding to the number of blood pressure time series data inputted thereto, each feature extraction layer being an LSTM feature extraction unit layer, the output of the previous LSTM feature extraction unit layer being used as the input of the next LSTM feature extraction unit layer, and each LSTM feature extraction unit layer being also used to input the blood pressure data at the corresponding moment in the blood pressure time series data. For example, one input of the last LSTM feature extraction unit layer is the output of the second-to-last LSTM feature extraction unit layer, and the other input is the current blood pressure data in the first current blood pressure time series data. The long short-term memory network model can record the output of the previous moment as the input of the current moment, and add the blood pressure data inputted at the current moment, that is, the current moment of the long short-term memory network model includes 2 inputs, so that the model can use historical data, has time series, and can accurately extract the effective blood pressure feature data in the blood pressure time series data, thereby avoiding the adverse effect of noise information on the accuracy of blood pressure detection.

[0035] The first feature data is the feature data output by the first feature extraction network model at the moment before the current moment, and the second feature data is the feature data output by the first feature extraction network model at the current moment. Taking the long short-term memory network model as an example, the first feature data is the feature data output by the second-to-last LSTM feature extraction unit layer, and the second feature data is the feature data output by the last LSTM feature extraction unit layer. The feature data output by each LSTM feature extraction unit layer contains the information of the blood pressure data before the corresponding moment.

[0036] In other embodiments, the first feature extraction network model may also be a Gated Recurrent Unit (GRU) or other recurrent network model that can extract features from blood pressure time series data.

[0037] S024: Input the first feature data and the second feature data into the first twin network model to obtain a first similarity output by the first twin network model.

[0038] The first characteristic data is used to characterize the blood pressure at the same detection site before the current moment, and the second characteristic data is used to characterize the blood pressure corresponding to the same detection site at the current moment. The first similarity between the first characteristic data and the second characteristic data is used to characterize the change in blood pressure at the same detection site at the current moment relative to the blood pressure before the current moment.

[0039] In this embodiment, the first network model used to compare the first feature data and the second feature data is a first twin network model. In other embodiments, the first network model can also be a neural network model that can compare the similarity between two input data. The first twin network model is used to compare the Euclidean distance between the first feature data and the second feature data, thereby outputting a first similarity that characterizes the similarity between the first feature data and the second feature data. The value of the first similarity is 1 or 0. The value of the first similarity is 1, which means that the first feature data and the second feature data are similar, indicating that the blood pressure corresponding to the same detection site at the current moment does not change much relative to the blood pressure before the current moment, that is, the blood pressure at different moments is relatively stable; the first similarity is 0, which means that the first feature data and the second feature data are not similar, indicating that the blood pressure corresponding to the same detection site at the current moment changes greatly relative to the blood pressure before the current moment, that is, the blood pressure is relatively unstable.

[0040] See also Figure 3 As shown, it is a flow chart of determining the second similarity in the blood pressure detection method provided in some embodiments of the present application. Specifically, in some embodiments, S04 specifically includes the description between each step of S042 and S044 as follows.

[0041] S042: Input the current blood pressure data corresponding to the two different detection positions at the current moment into the second feature extraction network model for feature extraction to obtain third feature data and fourth feature data.

[0042] The second feature extraction network model may include, but is not limited to, two convolutional neural network (CNN) models that are trained based on blood pressure data sets to learn feature representations of blood pressure data, and the two convolutional neural network models are used to extract features of current blood pressure data corresponding to two different detection sites, respectively, to output third feature data and fourth feature data, respectively. The current blood pressure data of the first detection site of the two different detection sites is input into the first convolutional neural network model in the second feature extraction network model, so that the first convolutional neural network model outputs the third feature data. The current blood pressure data of the second detection site of the two different detection sites is input into the second convolutional neural network model in the second feature extraction network model, so that the second convolutional neural network model outputs the fourth feature data. In other embodiments, the second feature extraction network model may also be other neural network models that can extract spatial features of input data.

[0043] S044: Input the third feature data and the fourth feature data into the second twin network model to obtain a second similarity output by the second twin network model.

[0044] The third characteristic data is used to characterize the blood pressure at the first detection site before the current moment, and the fourth characteristic data is used to characterize the blood pressure corresponding to the second detection site at the current moment. The second similarity between the third characteristic data and the fourth characteristic data is used to characterize the change between the current blood pressure at the first detection site and the current blood pressure at the second detection site.

[0045] In this embodiment, the second network model used to compare the third feature data and the fourth feature data is a second twin network model. In other embodiments, the second network model can also be a neural network model that can compare the similarity between two input data. The second twin network model is used to compare the Euclidean distance between the third feature data and the fourth feature data, thereby outputting a second similarity that characterizes the similarity between the third feature data and the fourth feature data. The value of the second similarity is 1 or 0. The value of the second similarity is 1, indicating that the third feature data and the fourth feature data are similar, indicating that the current blood pressure of the first detection site does not change much relative to the current blood pressure of the second detection site, that is, indicating that the blood pressure of different detection sites is relatively stable; the second similarity is 0, indicating that the third feature data and the fourth feature data are not similar, indicating that the current blood pressure of the first detection site changes greatly relative to the current blood pressure of the second detection site, that is, indicating that the blood pressure of different detection sites is relatively unstable.

[0046] In some embodiments, before performing S02 and S04, the blood pressure detection method provided in the embodiment of the present application also includes: obtaining blood pressure collection data collected in real time at each detection part according to a preset collection frequency; calculating the blood pressure data corresponding to each detection part at each time according to the blood pressure collection data, and obtaining the current blood pressure time series data corresponding to each detection part; the blood pressure data corresponding to each time is calculated according to the blood pressure collection data in the corresponding calculation time period, and the calculation time periods corresponding to two adjacent times have a preset overlap rate. It should be noted here that the time in this application, except for special statements, refers to the calculation time of the blood pressure data, that is, the time corresponding to the detection frequency.

[0047] The blood pressure detection method provided in some embodiments of the present application uses two blood pressure sensors to synchronously collect blood pressure at the first detection site and the second detection site at the same collection frequency, and obtains the blood pressure collection data corresponding to the first detection site and the second detection site respectively. Figure 4 As shown, it is a schematic diagram of the calculation method for obtaining blood pressure data at each moment of the first detection part and the second detection part according to the embodiment of the present application. The blood pressure collection data obtained at each collection time point of the first detection part include a0, a1, a2, a3, ..., and the blood pressure collection data obtained at each collection time point of the second detection part include b0, b1, b2, b3, .... For example, the above collection frequency is 1s, and the calculation time period corresponding to each moment is 6s. Then, a blood pressure collection data of each detection part can be obtained every 1s. After the valid blood pressure collection data is collected, the blood pressure data of the first moment is calculated after 6s. Taking the first detection part as an example, the blood pressure data P11 corresponding to the first moment is the first calculation time period corresponding to the first moment after the valid blood pressure collection data a0 is collected. The blood pressure collection data in the first calculation time period are a0, a1, a2, ..., a6. According to the blood pressure collection data a0, a1, a2, ..., a6, the corresponding calculation is performed to obtain the blood pressure data P11 corresponding to the first detection part at the first moment.

[0048] In the embodiment of the present application, in order to improve the timing logic of the first feature extraction network model and / or the second feature extraction network model, so as to improve the accuracy of blood pressure detection without causing the current computing power to be unable to meet the implementation of the blood pressure detection method, the calculation time periods corresponding to the two adjacent moments of the same adjacent detection part are set to have a certain overlap rate. For example, in some embodiments, the above overlap rate can be 25%. If the time when the valid blood pressure collection data a0 is collected is defined as the start time, the second calculation time period corresponding to the second moment of the first detection part should be the time period from 4.5s to 10.5s. The corresponding blood pressure collection data in the second calculation time period are a5, a6, ..., a10. According to the blood pressure collection data a5, a6, ..., a10, corresponding calculations are performed to obtain the blood pressure data P12 corresponding to the first detection part at the second moment.

[0049] Similarly, according to the above calculation method, the blood pressure data P13 at the third moment, the blood pressure data P14 at the fourth moment, ..., of the first detection site can be obtained, and the blood pressure data P21 at the first moment, the blood pressure data P22 at the second moment, the blood pressure data P23 at the third moment, the blood pressure data P24 at the fourth moment, ..., of the second detection site can be obtained. Specifically, in some embodiments, the blood pressure data at the corresponding moment can be obtained by calculating the average value of the blood pressure collection data in each calculation time period, so as to use the average value of each blood pressure collection data in each calculation time period as the blood pressure data at the corresponding time.

[0050] See also Figure 5 As shown, it is a schematic diagram of blood pressure time series data corresponding to each detection part in the blood pressure detection method provided in the embodiment of the present application. In order to facilitate the observation of the relationship between the blood pressure data at two adjacent moments in time series, one of the two adjacent blood pressure data is deliberately set as a solid line frame and the other is set as a dotted line frame. In addition, Figure 5 The blood pressure change curves of the first detection site and the second detection site are respectively illustrated in the figure. The horizontal extension width of the blood pressure data represents its corresponding calculation time period. Obviously, for the blood pressure data of two adjacent moments, since the calculation time periods of the two are partially overlapped, the blood pressure data of the latter moment includes a part of the blood pressure collection data corresponding to the previous moment, which is beneficial to improve the detection accuracy. At the same time, starting from the first moment, according to the detection frequency, the blood pressure data is calculated once every certain period of time (4.5s in the above example). If the certain period of time here is the first preset time, the duration of the calculation time period corresponding to each moment is the second preset time, and the first preset time is less than the second preset time, as an optional implementation method, the first preset time is 75% of the second preset time.

[0051] In some embodiments, the blood pressure data corresponding to each moment are arranged in time series, and the blood pressure time series data corresponding to each non-detected part are obtained respectively. Among them, the blood pressure time series data corresponding to the first detection part at the current moment is the first current blood pressure time series data P1, and the blood pressure time series data corresponding to the second detection part at the current moment is the second current blood pressure time series data P2. The last data in the blood pressure time series data corresponding to each detection part at each moment is the blood pressure data of the detection part at the corresponding moment.

[0052] If the first detection site is the same detection site mentioned above, the input data of the first feature extraction network model at the current moment is the first current blood pressure time series data P1, and the input data of the second feature extraction network model at the current moment is the last blood pressure data between the first current blood pressure time series data P1 and the second current blood pressure time series data P2, that is, the current blood pressure data of the first detection site and the current blood pressure data of the second detection site.

[0053] Furthermore, in some embodiments, S042 specifically includes: converting the current blood pressure data corresponding to the two different detection sites at the current moment into image data respectively, and inputting the two image data into the corresponding convolutional neural network model for feature extraction respectively.

[0054] In the blood pressure detection method provided in the embodiment of the present application, the convolutional neural network model used in the second feature extraction network model is a one-dimensional convolutional neural network model. However, the one-dimensional convolutional neural network model has limited ability to extract spatial features, and is more suitable for extracting features in image data. In order to more accurately extract the third feature data and the fourth feature data, before the current blood pressure data of the first detection site and the second detection site are respectively input into the first convolutional neural network model and the second convolutional neural network model in the second feature extraction network model, the current blood pressure data corresponding to the first detection site and the second detection site are first converted into image data, and then the image data is input into the corresponding convolutional neural network model for image feature extraction. In this embodiment, the first convolutional neural network model and the second convolutional neural network model are models trained based on the blood pressure image data set after the blood pressure data set is converted.

[0055] In some embodiments, an optional implementation method of converting the blood pressure data corresponding to each detection part at each time, that is, the blood pressure time series data corresponding to each part, into image data is: converting each blood pressure data into a Gram angular field map. The converted Gram angular field map is input into the corresponding convolutional neural network model to obtain the corresponding feature data.

[0056] Specifically, the main steps of converting the blood pressure time series data of each detection part into the Gram angle field map include the following four steps.

[0057] Step 1: Normalize the blood pressure time series data to scale the scale of each data in the blood pressure time series data to the range of [-1,1].

[0058] Step 2: Convert the normalized data into polar coordinate data. Specifically, the corresponding polar coordinates are converted according to the sequence values ​​of the normalized data to obtain the polar coordinate data.

[0059] Step 3: Convert the polar coordinate data into matrix data. The matrix data may be GASF or GADF matrix data.

[0060] Step 4: Convert each element in the matrix data into a pixel value in a corresponding grayscale range to obtain a corresponding Gram angle field map. The corresponding grayscale range may be, but is not limited to, a grayscale range of 0-256. Specifically, the matrix data may be converted into an image using a heat map, that is, the values ​​at different positions of the matrix data represent the color of the image.

[0061] In some embodiments, S06 specifically includes: determining whether the first similarity and the second similarity both meet corresponding preset conditions; if not, determining whether the detected object has abnormal blood pressure at the current moment.

[0062] Whether the first similarity and the second similarity meet the corresponding preset conditions, specifically, whether the values ​​of the first similarity and the second similarity are 1 respectively. If so, it means that both the first similarity and the second similarity meet the corresponding preset conditions, indicating that the blood pressure changes at the same detection site of the detection object at different times are relatively stable, and the blood pressure at different detection sites is also relatively stable, and the blood pressure of the detection object at the current moment is normal; if not, it means that at least one of the first similarity and the second similarity does not meet the corresponding preset conditions, which means that the detection object has abnormal blood pressure at the current moment.

[0063] In some embodiments, after S06, it further includes: when it is determined that the blood pressure of the detected object is abnormal at the current moment, issuing an early warning prompt.

[0064] The warning prompt can be a warning by displaying corresponding images and / or texts, or by sound and / or light. At every moment, once it is determined that the test subject has abnormal blood pressure, a warning prompt will be issued, which can promptly remind the test subject to pay attention to the abnormal blood pressure and take corresponding countermeasures in time, effectively improving the health index of the test subject.

[0065] In other embodiments, after S06, it further includes: when it is determined that the detection object has abnormal blood pressure at the current moment, if the number of moments when the detection object is determined to have abnormal blood pressure in the preset moments before the current moment meets the preset number, an early warning prompt is issued.

[0066] The blood pressure detection method provided by the present application is always based on the blood pressure changes of the detection object at different times and different detection positions to determine whether the blood pressure is abnormal. When the blood pressure changes at different times of the same detection part are relatively large or the blood pressure changes at different detection parts at the same time are relatively large, it is determined that the detection object has abnormal blood pressure at the current moment. If an early warning is given at each moment when the blood pressure abnormality is determined, frequent early warnings may occur, which may cause anxiety to the detection object. Based on this, the blood pressure detection method provided by the embodiment of the present application performs an early warning prompt for abnormal blood pressure in the following manner: when the detection object is determined to have abnormal blood pressure at the current moment, an early warning prompt is not directly given, but first determines whether the number of moments when the detection object has abnormal blood pressure reaches a certain number within the most recent preset moments before the current moment, or first determines whether the number of moments when the detection object has abnormal blood pressure reaches a certain number within the most recent set number of moments including the current moment, and then an early warning prompt is given to avoid the phenomenon of frequent early warnings. For example, when it is determined that the detection object has abnormal blood pressure at the current moment, if there are 6 moments in the nearly 10 moments including the current moment that the detection object has abnormal blood pressure, an early warning prompt is given.

[0067] See also Figure 6 As shown, it is a structural block diagram of a blood pressure detection device provided according to some embodiments of the present application. In some embodiments, the blood pressure detection device provided by the present application includes a first determination module 100, a second determination module 200 and a third determination module 300. Specifically, the first determination module 100 is used to compare the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment, and determine the first similarity between the current blood pressure data of the same detection part and the previous blood pressure data. The second determination module 200 is used to compare the current blood pressure data corresponding to different detection parts of the detection object at the current moment, and determine the second similarity between the current blood pressure data of different detection parts. The third determination module 300 is used to determine whether the detection object has abnormal blood pressure at the current moment based on the first similarity and the second similarity. The blood pressure detection device provided in the embodiment of the present application and the blood pressure detection method provided in the embodiment of the present application can achieve the same detection effect, which will not be described here.

[0068] See also Figure 7As shown, it is a schematic diagram of a specific implementation structure of a blood pressure detection device provided according to some embodiments of the present application. In some embodiments, the first determination module 100 specifically includes a first feature extraction module 101 and a first comparison module 102. The first feature extraction module 101 is used to input the first current blood pressure time series data P1 corresponding to the same detection part at the current moment into the first feature extraction network model for feature extraction to obtain first feature data and second feature data. The first feature extraction module 101 includes a first feature extraction network model, and the first feature extraction network model can be but is not limited to a long short-term memory network (LSTM) model trained based on a blood pressure data set. The first feature data and the second feature data have been described above, and will not be repeated here. The first comparison module 102 is used to input the first feature data and the second feature data into the first twin network model to obtain the first similarity output by the first twin network model. That is, the first comparison module 102 specifically includes the first twin network model.

[0069] Please continue reading Figure 7 As shown, in some embodiments, the second determination module 200 further includes a second feature extraction module 201 and a second comparison module 202. The second feature extraction module 201 is used to input the current blood pressure data corresponding to the two different detection parts at the current moment into the second feature extraction network model for feature extraction to obtain third feature data and fourth feature data. Specifically, the two different detection parts here include the same detection part mentioned above, and the blood pressure data corresponding to the two different detection parts at the current moment are respectively the last data in the first current blood pressure time series data P1 and the second current blood pressure time series data P2, that is, the current blood pressure data corresponding to the two different detection parts at the current moment. Therefore, in this embodiment, the second feature extraction module 201 is specifically used to input the last data in the first current blood pressure time series data P1 and the second current blood pressure time series data P2 into the corresponding convolutional neural network (CNN) model in the second feature extraction network model for feature extraction to obtain the third feature data and the fourth feature data respectively.

[0070] The second comparison module 202 is used to input the third feature data and the fourth feature data into the second twin network model to obtain the second similarity output by the second twin network model. That is, the second comparison module 202 specifically includes the second twin network model.

[0071] In some embodiments, the blood pressure detection device provided by the present application further includes an acquisition module. The acquisition module is used to: obtain the blood pressure data collected in real time at each detection site according to a preset acquisition frequency; calculate the blood pressure data corresponding to each detection site at each time according to the blood pressure data, and obtain the current blood pressure time series data corresponding to each detection site; the blood pressure data corresponding to each time is calculated according to the blood pressure data collected in the corresponding calculation time period, and the calculation time periods corresponding to two adjacent times have a preset overlap rate.

[0072] Furthermore, the above-mentioned different detection sites include a first detection site (the same detection site mentioned above) and a second detection site. The acquisition module is specifically used to acquire the blood pressure acquisition data collected in real time at a certain acquisition frequency for the first detection site and the second detection site respectively, and then calculate the currently acquired blood pressure acquisition data at each moment according to the above-mentioned detection frequency to obtain the blood pressure data corresponding to each moment, and the blood pressure data corresponding to each moment are arranged in time series to obtain the blood pressure time series data corresponding to the first detection site and the second detection site respectively. Among them, the blood pressure time series data corresponding to the first detection site at the current moment is the first current blood pressure time series data P1, and the blood pressure time series data corresponding to the second detection site at the current moment is the second current blood pressure time series data P2.

[0073] Furthermore, in some embodiments, the second feature extraction module 201 also includes a data conversion module, which is used to convert the current blood pressure corresponding to the two different detection sites at the current moment into image data, and input the two image data into the corresponding convolutional neural network model for feature extraction.

[0074] In some embodiments, the third determination module 300 is specifically used to determine whether the first similarity and the second similarity both meet the corresponding preset conditions, and when it is determined that the first similarity and the second similarity both meet the corresponding preset conditions, it is determined that the detection object is normal at the current moment, otherwise, it is determined that the detection object has abnormal blood pressure at the current moment. For further information, please continue to refer to Figure 7 As shown, in some embodiments, the third determination module 300 includes an OR calculation determination module 301, which is used to calculate the OR value of the first similarity output by the first comparison module 102 and the second similarity output by the second comparison module 202, and determine whether the blood pressure of the detection object is abnormal at the current moment according to the OR value. If the values ​​of the first similarity and the second similarity are 1 respectively, and the OR value obtained by the OR calculation determination module 301 is 1, it is determined that the blood pressure of the detection object at the current moment is normal; if one of the values ​​of the first similarity and the second similarity is 0, and the OR value obtained by the OR calculation determination module 301 is 0, it is determined that the blood pressure of the detection object at the current moment is abnormal.

[0075] The blood pressure detection device provided by the present application further includes an early warning module. In some embodiments, the early warning module is used to issue an early warning prompt when it is determined that the detection object has abnormal blood pressure at the current moment. In other embodiments, the early warning module is also specifically used to: when it is determined that the detection object has abnormal blood pressure at the current moment, if the number of moments when the detection object has abnormal blood pressure in the preset moments before the current moment meets the preset number, then an early warning prompt is issued.

[0076] See also Figure 8 As shown, it is a structural block diagram of a blood pressure detection device provided according to some embodiments of the present application. The blood pressure detection device provided in the embodiments of the present application includes a memory 1 and a processor 2. The memory 1 stores a computer program that can be executed by the processor 2. When the computer program is executed by the processor 2, the blood pressure detection method provided in any embodiment of the present application is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0077] In some embodiments, the blood pressure detection method provided by the present application further includes at least two sensors connected to the processor 2, and the at least two sensors are used to sense the blood pressure of different detection parts of the detection object to collect blood pressure collection data corresponding to the different detection parts, and send the blood pressure collection data to the processor 2 for corresponding processing. Figure 8 As shown, at least two sensors include a first sensor 3 and a second sensor 4. The first sensor 3 is used to collect the blood pressure of the first detection part of the detection object in real time according to the first acquisition frequency, obtain the first blood pressure acquisition data and send it to the processor 2. The second sensor 4 is used to collect the blood pressure of the second detection part of the detection object in real time according to the second acquisition frequency, obtain the second blood pressure acquisition data and send it to the processor 2. The first acquisition frequency and the second acquisition frequency can be the same, and the first sensor 3 and the second sensor 4 synchronously collect the blood pressure of the corresponding detection part according to the same acquisition frequency to obtain the first blood pressure acquisition data of the first detection part and the second blood pressure acquisition data of the second detection part. The same detection part in the blood pressure detection method can be the first detection part collected by the first sensor 3, and the different detection parts in the blood pressure detection method include the first detection part and the second detection part collected by the second sensor 4. In order to improve the detection accuracy, the distance between the first detection part and the second detection part meets a certain distance condition. For example, the distance between the first detection part and the second detection part is greater than the set distance.

[0078] In some embodiments, the blood pressure detection device further includes an early warning module connected to the processor 2, and the early warning module is used to make corresponding early warning prompts based on the early warning prompt information sent by the processor 2. Figure 8As shown, in some embodiments, the early warning module specifically includes a display screen 5 and / or a sound generator 6. The display screen 5 is used to display early warning prompts in the form of images and / or text. The generator 6 is used to make sound early warning prompts according to the early warning prompt information. Further, the display screen 5 is also used to display the blood pressure information of the detected object. The blood pressure information of the detected object includes its corresponding blood pressure collection data, blood pressure data corresponding to each moment and / or blood pressure change information, etc. In other embodiments, the early warning module may also include an indicator light for emitting light prompts according to the early warning prompt information sent by the processor 2.

[0079] See also Fig. 9 As shown, it is a schematic diagram of the appearance of a blood pressure detection device provided according to some embodiments of the present application. In some embodiments, the blood pressure detection device provided in the embodiments of the present application is a blood pressure detection watch. The blood pressure detection bracelet includes a strap 11 and a case 12. The case 12 is provided with a display screen 5, and the case 12 is provided with the above-mentioned memory 1 and processor 2 ( Fig. 9 Not shown in the figure), one of the first sensor 3 and the second sensor 4 is arranged on the strap, and the other is arranged on the strap or the lower surface of the watch case. Specifically, the first sensor 3 is arranged on a side of the watch case 12 close to the skin of the test object, and is used to collect the blood pressure on the back of the subject's wrist. The second sensor 4 is arranged on the strap at a position opposite to the first sensor 3, and is used to collect the blood pressure on the front of the subject's wrist. The blood pressure detection watch provided in the embodiment of the present application collects blood pressure collection data of multiple detection parts of the test object in real time based on at least two sensors, and processes the blood pressure collection data accordingly based on the processor 2, accurately detects whether the blood pressure of the test object is abnormal, and issues corresponding early warning prompts, which is beneficial for the test object to effectively prevent blood pressure-related diseases.

[0080] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned blood pressure detection method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. Among them, the computer-readable storage medium is such as a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0081] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A blood pressure detection method, characterized in that: include: Comparing the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment, and determining a first similarity between the current blood pressure data and the previous blood pressure data of the same detection part; as well as, comparing the current blood pressure data corresponding to different detection parts of the detection object at the current moment, and determining a second similarity between the current blood pressure data of the different detection parts; It is determined whether the detection object has abnormal blood pressure at the current moment according to the first similarity and the second similarity.

2. The blood pressure detection method according to claim 1, characterized in that: The step of comparing the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment to determine the first similarity between the current blood pressure data and the previous blood pressure data of the same detection part includes: Input the first current blood pressure time series data corresponding to the same detection site at the current moment into the first feature extraction network model for feature extraction to obtain first feature data and second feature data; the first feature extraction network model is a long short-term memory network model that is trained based on the blood pressure data set to learn the feature representation of the blood pressure data, the first feature data is the feature data output by the first feature extraction network model at the moment before the current moment, and the second feature data is the feature data output by the first feature extraction network model at the current moment; The first feature data and the second feature data are input into a first twin network model to obtain the first similarity output by the first twin network model.

3. The blood pressure detection method according to claim 1, characterized in that: Before comparing the current blood pressure data corresponding to the same detection part of the detection object at the current moment with the previous blood pressure data corresponding to the previous moment, the method further includes: Acquire blood pressure data collected in real time at each detection site according to a preset collection frequency; The blood pressure data corresponding to each of the detection parts at each of the described moments are calculated based on the blood pressure collection data to obtain the current blood pressure time series data corresponding to each of the described moments; the blood pressure data corresponding to each of the described moments are calculated based on the blood pressure collection data within the corresponding calculation time period, and the calculation time periods corresponding to two adjacent described moments have a preset overlap rate.

4. The blood pressure detection method according to claim 2, characterized in that: The comparing the current blood pressure data corresponding to different detection parts of the detection object at the current moment to determine the second similarity between the current blood pressure data of the different detection parts includes: Inputting the current blood pressure data corresponding to the two different detection parts at the current moment into a second feature extraction network model for feature extraction to obtain third feature data and fourth feature data, wherein the second feature extraction network model includes two convolutional neural network models that are respectively trained based on the blood pressure data set to learn the feature representation of the blood pressure data, and the two convolutional neural network models are respectively used to extract features from the current blood pressure data corresponding to the two different detection parts to output the third feature data and the fourth feature data respectively; The third feature data and the fourth feature data are input into the second twin network model to obtain the second similarity output by the second twin network model.

5. The blood pressure detection method according to claim 4, characterized in that: The step of inputting the current blood pressure data corresponding to the two different detection positions at the current moment into the second feature extraction network model for feature extraction comprises: The current blood pressure data corresponding to the two different detection sites at the current moment are respectively converted into image data, and the two image data are respectively input into the corresponding convolutional neural network model for feature extraction.

6. The blood pressure detection method according to any one of claims 1 to 5, characterized in that: The determining, based on the first similarity and the second similarity, whether the detection object has abnormal blood pressure at the current moment includes: Determining whether the first similarity and the plurality of second similarities all satisfy corresponding preset conditions; If not, it is determined that the detection subject has abnormal blood pressure at the current moment.

7. The blood pressure detection method according to any one of claims 1 to 5, characterized in that: Also includes: When it is determined that the test subject has abnormal blood pressure at the current moment, issuing an early warning prompt; or, When it is determined that the detection object has abnormal blood pressure at the current moment, if the number of moments when the detection object is determined to have abnormal blood pressure in the preset moments before the current moment meets the preset number, an early warning prompt is issued.

8. A blood pressure detection device, characterized in that: include: A first determination module, configured to compare current blood pressure data corresponding to the same detection site of the detection subject at the current moment with previous blood pressure data corresponding to the previous moment, and determine a first similarity between the current blood pressure data and the previous blood pressure data of the same detection site; A second determination module, configured to compare the current blood pressure data corresponding to different detection parts of the detection object at the current moment, and determine a second similarity between the current blood pressure data of the different detection parts; A third determination module is configured to determine whether the detection object has abnormal blood pressure at the current moment according to the first similarity and the second similarity.

9. A blood pressure detection device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the blood pressure detection method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the blood pressure detection method according to any one of claims 1 to 7 is implemented.

Images