Blood pressure measurement method and device based on data processing

Abstract

The present application provides a blood pressure measurement method and device based on data processing, which relate to the field of medical technology. The blood pressure values ​​of the target user at various times in the history of 1 day with the highest similarity of the historical blood pressure portrait description feature can be determined from a preset user database. Three moments are removed from the target user's various moments in the history of 1 day, and according to the time series relationship, the blood pressure values ​​corresponding to each moment after removing the three moments from each moment in the history of 1 day are inserted into the blood pressure values ​​of the user collected at the three moments to obtain the blood pressure values ​​of the user at various times in 1 day. For each moment in 1 day, the body behavior posture data within the previous half an hour at that moment is determined from the preset database. According to the blood pressure values ​​of the user at various moments in 1 day and the adjustment factors corresponding to each moment, the first blood pressure value of the user is determined.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a blood pressure measurement method and device based on data processing. Background Art

[0002] At present, the prevalence of hypertension is increasing. People use blood pressure monitoring equipment to measure blood pressure in order to achieve the purpose of early prevention. At present, the way to determine blood pressure includes using blood pressure measuring equipment to collect blood pressure values ​​of the human body multiple times, and taking the average of the multiple blood pressure values ​​collected from the human body as the measured blood pressure value.

[0003] However, human blood pressure changes continuously over 24 hours. Using the average value of multiple blood pressure values ​​collected from the human body as the blood pressure measurement value has low reliability and accuracy. Summary of the invention

[0004] The present application provides a blood pressure measurement method and device based on data processing, which are used to solve the problems in the prior art.

[0005] In a first aspect, the present application provides a blood pressure measurement method based on data processing, which is applied to a server and includes:

[0006] At the target time of each day, the blood pressure value of the user collected by the blood pressure measuring device at the three previous moments of the day, the body behavior posture data of the user collected by the smart watch at each moment of the day, and the user attribute information of the user are obtained, wherein the interval between any two adjacent moments of the three moments is greater than the set time threshold, and the user attribute information includes at least gender information, age information, height information, and weight information;

[0007] Based on the user's blood pressure values ​​collected at three moments, the user's body posture data collected at each moment of the day, and the user's attribute information, a blood pressure profile description feature of the user is constructed;

[0008] Determine the similarity between the blood pressure profile description feature of the user and each historical blood pressure profile description feature in a preset user database;

[0009] From a preset user database, determine the blood pressure values ​​of the target user at various times within a historical day with the most similar historical blood pressure portrait description features;

[0010] Remove 3 moments from the target user's historical moments within one day, and according to the time series relationship, insert the blood pressure values ​​corresponding to the moments after removing the 3 moments from the historical moments within one day into the blood pressure values ​​collected at the 3 moments to obtain the blood pressure values ​​of the user at each moment within one day;

[0011] For each moment in 1 day, determine the body behavior posture data in the previous half hour at that moment from a preset database;

[0012] Determine the adjustment factor corresponding to the blood pressure value at that moment based on the body behavior and posture data in the previous half hour;

[0013] The user's first blood pressure value is determined according to the user's blood pressure values ​​at various times within a day and the adjustment factor corresponding to each time.

[0014] In a possible implementation, after determining the first blood pressure value of the user, the method provided by the present application includes:

[0015] Count the number of each value and the total number of each blood pressure value at each time point in a day;

[0016] Determine the ratio of the number of each value to the total number of blood pressure values ​​at each moment according to the number of each value and the total number of blood pressure values ​​at each value;

[0017] Determine the average body behavior posture data of the user on the day according to the body behavior posture data of the user collected at various times on the day;

[0018] Inputting the ratio of the number of each value to the total number of blood pressure values ​​of each value, the average body behavior posture data of the user on the day, and the user attribute information into a pre-trained blood pressure determination model to obtain a second blood pressure value, wherein the blood pressure determination model is trained by inputting multiple training samples into the first neural network, and each training sample includes the ratio of the number of each value to the total number of blood pressure values ​​at each moment in history, the average body behavior posture data of the user on the day in history, the user attribute information in history, and the corresponding second blood pressure value marked in history by an expert;

[0019] The first blood pressure value is corrected according to the second blood pressure value.

[0020] In a possible implementation, correcting the first blood pressure value according to the second blood pressure value includes:

[0021] According to the formula , correct the first blood pressure value, where, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction.

[0022] In a possible implementation, determining the first blood pressure value of the user according to the blood pressure values ​​of the user at various times within a day and the adjustment factor corresponding to each time includes:

[0023] According to the formula , determine a first blood pressure value of the user, wherein, is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

[0024] In a possible implementation, determining the adjustment factor corresponding to the blood pressure value at this moment based on the body behavior posture data in the previous half hour includes:

[0025] Based on the body behavior posture data within the previous half hour and the pre-trained adjustment factor model, the adjustment factor corresponding to the blood pressure value at this moment is determined, wherein the adjustment factor model is trained by inputting multiple training samples into the second neural network, and each training sample includes the body behavior posture data within the previous half hour and the corresponding historical standard adjustment factor.

[0026] In a second aspect, the present application also provides a blood pressure measurement device based on data processing, which is applied to a server. The device provided by the present application includes:

[0027] a data acquisition unit, configured to, at a target time each day, collect the blood pressure value of the user at three moments before receiving the blood pressure measurement device on the same day, receive the body behavior posture data of the user collected by the smart watch at each moment on the same day, and obtain the user attribute information of the user, wherein the interval between any two adjacent moments in the three moments is greater than a set time threshold, and the user attribute information includes at least gender information, age information, height information, and weight information;

[0028] A feature construction unit, used to construct a blood pressure portrait description feature of the user based on the blood pressure values ​​of the user collected at three moments, the body posture data of the user collected at each moment of the day, and the user attribute information;

[0029] A similarity determination unit, used to determine the similarity between the description features of the user's blood pressure portrait and the description features of each historical blood pressure portrait in a preset user database;

[0030] A blood pressure value search unit, used to determine the blood pressure values ​​of the target user at various times within one day in history with the most similar historical blood pressure portrait description feature from a preset user database;

[0031] The blood pressure data insertion unit is used to remove 3 moments from each moment in the history of the target user within one day, and according to the time sequence relationship, insert the blood pressure values ​​corresponding to each moment after removing the 3 moments from each moment in the history of one day into the blood pressure values ​​of the user collected at the 3 moments, so as to obtain the blood pressure values ​​of the user at each moment in one day;

[0032] A posture data determination unit, for determining, at each moment in a day, from a preset database, body behavior posture data within the previous half hour at that moment;

[0033] An adjustment factor determination unit, used to determine the adjustment factor corresponding to the blood pressure value at that moment according to the body behavior posture data in the previous half hour;

[0034] The first blood pressure value determination unit is used to determine the first blood pressure value of the user according to the blood pressure values ​​of the user at various moments within a day and the adjustment factor corresponding to each moment.

[0035] In a possible implementation, the device provided by the present application further includes:

[0036] A data statistics unit is used to count the number of each value and the total number of each blood pressure value at each time of the user's blood pressure value within one day;

[0037] a ratio determination unit, for determining a ratio of the number of each value to the total number of blood pressure values ​​of each value according to the number of each value in the blood pressure values ​​at each moment and the total number of blood pressure values ​​of each value;

[0038] The average posture data determination unit is used to determine the average body behavior posture data of the user on the day based on the body behavior posture data of the user collected at various times on the day.

[0039] a second blood pressure value determination unit, for inputting a ratio of the number of each value to the total number of blood pressure values ​​of each value, average body behavior posture data of the user on the day, and user attribute information into a pre-trained blood pressure determination model to obtain a second blood pressure value, wherein the blood pressure determination model is trained by inputting a plurality of training samples into the first neural network, each training sample including a ratio of the number of each value to the total number of blood pressure values ​​at each moment in history, average body behavior posture data of the user on the day in history, user attribute information in history, and a corresponding second blood pressure value marked in history by an expert;

[0040] The blood pressure value correction unit is used to correct the first blood pressure value according to the second blood pressure value.

[0041] In a possible implementation, the blood pressure value correction unit is specifically configured to calculate the blood pressure value according to the formula , correct the first blood pressure value, where, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction.

[0042] In a possible implementation manner, the first blood pressure value determination unit is specifically configured to determine the blood pressure value according to the formula , determine a first blood pressure value of the user, wherein, is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

[0043] In one possible implementation, the adjustment factor determination unit is specifically used to determine the adjustment factor corresponding to the blood pressure value at that moment based on the body behavior posture data within the previous half hour and a pre-trained adjustment factor model, wherein the adjustment factor model is trained based on multiple training samples input into the second neural network, and each training sample includes the body behavior posture data within the previous half hour and the corresponding historical standard adjustment factor.

[0044] The present application provides a blood pressure measurement method and device based on data processing, which can construct a blood pressure portrait description feature of the user based on the user's blood pressure values ​​collected at three moments, the user's body posture data collected at each moment of the day, and the user's attribute information.

[0045] Determine the similarity between the user's blood pressure portrait description feature and each historical blood pressure portrait description feature in a preset user database; from the preset user database, determine the blood pressure values ​​of the target user with the historical blood pressure portrait description feature with the highest similarity at various times within one day in history. It can be understood that the more similar the user's blood pressure portrait description feature is to the historical blood pressure portrait description feature, the more similar their blood pressure values ​​at various times within one day are, and then the blood pressure values ​​of the target user with the historical blood pressure portrait description feature with the highest similarity at various times within one day in history can be used as the blood pressure values ​​of the user to be measured at various times within one day.

[0046] Remove 3 moments from the target user's historical moments within a day, and based on the time series relationship, insert the blood pressure values ​​corresponding to each moment after removing the 3 moments from the historical moments within a day into the blood pressure values ​​of the user collected at the 3 moments to obtain the blood pressure values ​​of the user at each moment within a day with high reliability.

[0047] For each moment in a day, the body behavior posture data in the previous half hour at that moment is determined from a preset database. It is understandable that the body behavior posture data in the previous half hour (such as slow walking, running, push-ups, or squatting exercises, etc.) have different effects on the blood pressure at the current moment, so the adjustment factor corresponding to the blood pressure value at that moment can be determined based on the body behavior posture data in the previous half hour; the user's first blood pressure value is determined based on the user's blood pressure values ​​at various moments in a day and the adjustment factors corresponding to each moment, and the obtained user's first blood pressure value has high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0049] Figure 1 A flow chart of a blood pressure measurement method based on data processing provided in an embodiment of the present application;

[0050] Figure 2 This is a functional module block diagram of a blood pressure measurement device based on data processing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments made by ordinary technicians in this field under the enlightenment of the embodiments belong to the scope of protection of the present application.

[0052] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. 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 necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0053] See also Figure 1 The embodiment of the present application provides a blood pressure measurement method based on data processing, which is applied to a server and includes:

[0054] S101: At a target time each day, the blood pressure value of the user collected by the blood pressure measuring device at three times before the day is received, the body behavior posture data of the user collected by the smart watch at each time of the day is received, and the user attribute information of the user is obtained.

[0055] At 21:00 every evening (i.e. the target time), the blood pressure measurement device collects the user's blood pressure values ​​at 8:00 in the morning, 13:00 in the afternoon, and 20:00 in the evening (i.e. the previous three times of the day) on the same day.

[0056] Among them, the interval between any two adjacent moments in the three moments is greater than the set time threshold (such as 4 hours), and the user attribute information includes at least gender information, age information, height information, and weight information. Exemplarily, the smart watch includes an accelerometer and a gyroscope, and the user's body behavior posture data can be collected by the accelerometer and the gyroscope, and the user's body behavior posture data can describe the user's body motion state. When the user's body behavior posture data changes regularly, the user's body motion state is a standard motion state, such as a slow walking state, a fast walking state, a running state, a squatting motion state, or a combination of at least two of the above motion states. When the user's body behavior posture data changes regularly, the user's body motion state is a non-standard motion state. It can be understood that for different users' body behavior posture data, regardless of whether the user's body behavior posture data changes regularly, the degree of influence on the user's blood pressure is different.

[0057] S102: Construct a blood pressure profile description feature of the user based on the blood pressure values ​​of the user collected at three moments, the body posture data of the user collected at each moment of the day, and the user attribute information.

[0058] Exemplarily, the user's blood pressure values ​​collected at three moments, the user's body posture data collected at various moments of the day, and the user's attribute information may be spliced ​​together to construct a blood pressure portrait description feature of the user.

[0059] S103: Determine the similarity between the blood pressure portrait description feature of the user and each historical blood pressure portrait description feature in a preset user database.

[0060] Exemplarily, the Euclidean distance between the user's blood pressure portrait description feature and each historical blood pressure portrait description feature in a preset user database may be determined. The shorter the Euclidean distance, the higher the similarity.

[0061] S104: Determine, from a preset user database, the blood pressure values ​​of the target user at various times within a historical day with the highest similarity to the historical blood pressure portrait description feature.

[0062] S105: Remove 3 moments from the target user's various moments in the history of 1 day, and according to the time series relationship, insert the blood pressure values ​​corresponding to each moment after removing the 3 moments from the various moments in the history of 1 day into the blood pressure values ​​of the user collected at the 3 moments to obtain the blood pressure values ​​of the user at each moment in 1 day.

[0063] S106: For each moment in a day, determine the body behavior posture data in the previous half hour at that moment from a preset database.

[0064] S107: Determine the adjustment factor corresponding to the blood pressure value at this moment based on the body behavior posture data in the previous half hour.

[0065] For example, based on the body behavior posture data in the previous half hour and the pre-trained adjustment factor model, the adjustment factor corresponding to the blood pressure value at that moment is determined.

[0066] The adjustment factor model is trained by inputting multiple training samples into the second neural network, and each training sample includes body behavior posture data within half an hour before the history and the corresponding historical standard adjustment factor.

[0067] S108: Determine a first blood pressure value of the user according to the blood pressure values ​​of the user at various times within a day and the adjustment factor corresponding to each time.

[0068] For example, according to the formula , determine the user's first blood pressure value. is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

[0069] For example, It can be understood that the actual blood pressure value at time i, assuming that the blood pressure value measured at time 1 is 142 (high blood pressure), if the adjustment factor corresponding to time 1 is 0.966, then the actual blood pressure value corresponding to time 1 is (normal blood pressure); the blood pressure value measured at time 2 is 143 (high blood pressure), and the adjustment factor corresponding to time 2 is 0.959, so the actual blood pressure value corresponding to time 2 is (is normal blood pressure), and so on.

[0070] To summarize, the blood pressure measurement method based on data processing provided in the embodiment of the present application can construct a blood pressure portrait description feature of the user based on the blood pressure values ​​of the user collected at three moments, the body posture data of the user collected at each moment of the day, and the user attribute information.

[0071] Determine the similarity between the user's blood pressure portrait description feature and each historical blood pressure portrait description feature in a preset user database; from the preset user database, determine the blood pressure values ​​of the target user with the historical blood pressure portrait description feature with the highest similarity at various times within one day in history. It can be understood that the more similar the user's blood pressure portrait description feature is to the historical blood pressure portrait description feature, the more similar their blood pressure values ​​at various times within one day are, and then the blood pressure values ​​of the target user with the historical blood pressure portrait description feature with the highest similarity at various times within one day in history can be used as the blood pressure values ​​of the user to be measured at various times within one day.

[0072] Remove 3 moments from the target user's historical moments within a day, and based on the time series relationship, insert the blood pressure values ​​corresponding to each moment after removing the 3 moments from the historical moments within a day into the blood pressure values ​​of the user collected at the 3 moments to obtain the blood pressure values ​​of the user at each moment within a day with high reliability.

[0073] For each moment in a day, the body behavior posture data in the previous half hour at that moment is determined from a preset database. It is understandable that the body behavior posture data in the previous half hour (such as slow walking, running, push-ups, or squatting exercises, etc.) have different effects on the blood pressure at the current moment, so the adjustment factor corresponding to the blood pressure value at that moment can be determined based on the body behavior posture data in the previous half hour; the user's first blood pressure value is determined based on the user's blood pressure values ​​at various moments in a day and the adjustment factors corresponding to each moment, and the obtained user's first blood pressure value has high accuracy.

[0074] In a possible implementation manner, after S108, the method provided in the embodiment of the present application may further include:

[0075] Step 1: Count the number of each value and the total number of each blood pressure value at each time point in a day.

[0076] For example, there are X blood pressure values ​​of 135, Y blood pressure values ​​of 136, and the total number of blood pressure values ​​of each value is Z, ..., and so on.

[0077] Step 2: Determine the ratio of the number of each value to the total number of blood pressure values ​​at each moment according to the number of each value and the total number of blood pressure values ​​at each value.

[0078] For example, when the number of blood pressure values ​​is X, the number of blood pressure values ​​is Y, and the total number of blood pressure values ​​of each value is Z, the ratio of the blood pressure value of 135 to the total number of blood pressure values ​​of each value is , the ratio of blood pressure value 136 to the total number of blood pressure values ​​of each value is .

[0079] Step 2: Determine the average body behavior posture data of the user on the day based on the body behavior posture data of the user collected at various times on the day.

[0080] Step 4: Input the ratio of the number of each value to the total number of blood pressure values ​​of each value, the user's average body behavior posture data on that day, and the user attribute information into the pre-trained blood pressure determination model to obtain a second blood pressure value.

[0081] Among them, the blood pressure determination model is trained based on multiple training samples input into the first neural network, each training sample includes the ratio of the number of each historical value to the total number of blood pressure values ​​at each moment, the historical average body behavior posture data of the user on the day, the historical user attribute information, and the corresponding second blood pressure value marked by the expert in history.

[0082] Step 5: Correct the first blood pressure value according to the second blood pressure value.

[0083] For example, according to the formula , correct the first blood pressure value, where, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction. For example, the first correction factor may be 0.55, and the second correction factor may be 0.45.

[0084] Based on the above steps 1 to 5, the first blood pressure value can be obtained more accurately.

[0085] See also Figure 2 , the embodiment of the present application also provides a blood pressure measuring device based on data processing, which is applied to a server. It should be noted that the basic principle and technical effect of the blood pressure measuring device based on data processing provided in the embodiment of the present application are the same as those in the above-mentioned embodiment. For the sake of brief description, for parts not mentioned in the embodiment of the present application, please refer to the corresponding content in the above-mentioned embodiment. The device provided in the embodiment of the present application includes a data acquisition unit, a feature construction unit, a similarity determination unit, a blood pressure value search unit, a blood pressure data insertion unit, a posture data determination unit, an adjustment factor determination unit and a first blood pressure value determination unit, wherein,

[0086] The data acquisition unit is used to collect the user's blood pressure value at the three moments before the receiving blood pressure measurement device on the same day, receive the user's body behavior posture data collected by the smart watch at each moment on the same day, and obtain the user's user attribute information at the target time of each day, wherein the interval between any two adjacent moments in the three moments is greater than the set duration threshold, and the user attribute information includes at least gender information, age information, height information, and weight information.

[0087] The feature construction unit is used to construct the user's blood pressure portrait description features based on the user's blood pressure values ​​collected at three moments, the user's body posture data collected at each moment of the day, and the user's attribute information.

[0088] The similarity determination unit is used to determine the similarity between the user's blood pressure portrait description features and each historical blood pressure portrait description feature in a preset user database.

[0089] The blood pressure value search unit is used to determine the blood pressure values ​​of the target user at various times within one day in history with the highest similarity to the historical blood pressure portrait description feature from a preset user database.

[0090] The blood pressure data insertion unit is used to remove 3 moments from the target user's various moments in the history of 1 day, and according to the time series relationship, insert the blood pressure values ​​corresponding to each moment after removing the 3 moments from the various moments in the history of 1 day into the blood pressure values ​​of the user collected at the 3 moments to obtain the blood pressure values ​​of the user at each moment in 1 day.

[0091] The posture data determination unit is used to determine the body behavior posture data within the previous half hour at each moment in 1 day from a preset database.

[0092] The adjustment factor determination unit is used to determine the adjustment factor corresponding to the blood pressure value at that moment based on the body behavior posture data in the previous half hour.

[0093] The first blood pressure value determination unit is used to determine the first blood pressure value of the user according to the blood pressure values ​​of the user at various moments within a day and the adjustment factor corresponding to each moment.

[0094] In a possible implementation, the device provided by the present application further includes:

[0095] A data statistics unit is used to count the number of each value and the total number of each blood pressure value at each time of the user's blood pressure value within one day;

[0096] a ratio determination unit, for determining a ratio of the number of each value to the total number of blood pressure values ​​of each value according to the number of each value in the blood pressure values ​​at each moment and the total number of blood pressure values ​​of each value;

[0097] An average posture data determination unit, used to determine the average body behavior posture data of the user on the day according to the body behavior posture data of the user collected at various times on the day;

[0098] a second blood pressure value determination unit, for inputting a ratio of the number of each value to the total number of blood pressure values ​​of each value, average body behavior posture data of the user on the day, and user attribute information into a pre-trained blood pressure determination model to obtain a second blood pressure value, wherein the blood pressure determination model is trained by inputting a plurality of training samples into the first neural network, each training sample including a ratio of the number of each value to the total number of blood pressure values ​​at each moment in history, average body behavior posture data of the user on the day in history, user attribute information in history, and a corresponding second blood pressure value marked in history by an expert;

[0099] The blood pressure value correction unit is used to correct the first blood pressure value according to the second blood pressure value.

[0100] In a possible implementation, the blood pressure value correction unit is specifically configured to calculate the blood pressure value according to the formula , correct the first blood pressure value, where, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction.

[0101] In a possible implementation manner, the first blood pressure value determination unit is specifically configured to determine the blood pressure value according to the formula , determine a first blood pressure value of the user, wherein, is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

[0102] In one possible implementation, the adjustment factor determination unit is specifically used to determine the adjustment factor corresponding to the blood pressure value at that moment based on the body behavior posture data within the previous half hour and a pre-trained adjustment factor model, wherein the adjustment factor model is trained based on multiple training samples input into the second neural network, and each training sample includes the body behavior posture data within the previous half hour and the corresponding historical standard adjustment factor.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A blood pressure measurement method based on data processing, characterized in that: Applied to a server, the method comprises: At a target time of each day, receiving the blood pressure value of the user collected by the blood pressure measuring device at three previous times of the day, receiving the body behavior posture data of the user collected by the smart watch at each time of the day, and obtaining the user attribute information of the user, wherein the interval between any two adjacent moments of the three moments is greater than a set time threshold, and the user attribute information includes at least gender information, age information, height information, and weight information; Constructing a blood pressure portrait description feature of the user according to the blood pressure values ​​of the user collected at the three moments, the body posture data of the user collected at each moment of the day, and the user attribute information; Determine the similarity between the blood pressure portrait description feature of the user and each historical blood pressure portrait description feature in a preset user database; Determine, from the preset user database, the blood pressure values ​​of the target user having the historical blood pressure portrait description feature with the highest similarity at various times within one day in history; The three moments are removed from the target user's historical moments within one day, and according to the time sequence relationship, the blood pressure values ​​corresponding to the moments after removing the three moments from the historical moments within one day are inserted into the blood pressure values ​​of the user collected at the three moments to obtain the blood pressure values ​​of the user at each moment within one day; For each moment in the day, determine the body behavior posture data in the previous half hour at the moment from a preset database; Determine the adjustment factor corresponding to the blood pressure value at that moment according to the body behavior posture data in the previous half hour; The first blood pressure value of the user is determined according to the blood pressure values ​​of the user at various moments within a day and the adjustment factor corresponding to each moment.

2. The method according to claim 1, characterized in that After determining the first blood pressure value of the user, the method includes: Counting the number of each value and the total number of each blood pressure value at each time point in a day for the user; Determine a ratio of the number of each value to the total number of the blood pressure values ​​at each moment according to the number of each value and the total number of the blood pressure values ​​at each value; Determine the average body behavior posture data of the user on the day according to the body behavior posture data of the user collected at various times on the day; Inputting the ratio of the number of each value to the total number of blood pressure values ​​of each value, the average body behavior posture data of the user on the day, and the user attribute information into a pre-trained blood pressure determination model to obtain a second blood pressure value, wherein the blood pressure determination model is trained by inputting multiple training samples into a first neural network, and each of the training samples includes the ratio of the number of each value in history to the total number of blood pressure values ​​at each moment, the average body behavior posture data of the user on the day in history, the user attribute information in history, and the corresponding second blood pressure value marked in history by an expert; The first blood pressure value is corrected according to the second blood pressure value.

3. The method according to claim 2, characterized in that The step of correcting the first blood pressure value according to the second blood pressure value includes: According to the formula , correct the first blood pressure value, wherein, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction.

4. The method according to claim 1, characterized in that: The determining the first blood pressure value of the user according to the blood pressure values ​​of the user at various moments within a day and the adjustment factor corresponding to each moment includes: According to the formula , determine a first blood pressure value of the user, wherein, is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

5. The method according to claim 1, characterized in that Determining the adjustment factor corresponding to the blood pressure value at this moment based on the body behavior posture data in the previous half hour includes: Based on the body behavior posture data within the previous half hour and the pre-trained adjustment factor model, the adjustment factor corresponding to the blood pressure value at this moment is determined, wherein the adjustment factor model is trained by inputting multiple training samples into the second neural network, and each training sample includes the body behavior posture data within the previous half hour and the corresponding historical standard adjustment factor.

6. A blood pressure measuring device based on data processing, characterized in that: Applied to a server, the device comprises: a data acquisition unit, configured to, at a target time each day, collect the blood pressure value of the user at three moments before receiving the blood pressure measurement device on the day, receive the body behavior posture data of the user collected by the smart watch at each moment on the day, and obtain the user attribute information of the user, wherein the interval between any two adjacent moments of the three moments is greater than a set time threshold, and the user attribute information includes at least gender information, age information, height information, and weight information; a feature construction unit, configured to construct a blood pressure portrait description feature of the user according to the blood pressure values ​​of the user collected at the three moments, the body posture data of the user collected at each moment of the day, and the user attribute information; A similarity determination unit, used to determine the similarity between the blood pressure portrait description feature of the user and each historical blood pressure portrait description feature in a preset user database; A blood pressure value search unit, used to determine the blood pressure values ​​of the target user having the historical blood pressure portrait description feature with the highest similarity at various times within one day in history from the preset user database; a blood pressure data insertion unit, configured to remove the three moments from the target user's historical moments within one day, and insert the blood pressure values ​​corresponding to the moments after removing the three moments from the historical moments within one day into the blood pressure values ​​collected at the three moments, according to the time sequence relationship, to obtain the blood pressure values ​​of the user at each moment within one day; A posture data determination unit, for determining, for each moment in the day, from a preset database, the body behavior posture data within the previous half hour at the moment; An adjustment factor determination unit, used to determine the adjustment factor corresponding to the blood pressure value at that moment according to the body behavior posture data in the previous half hour; The first blood pressure value determination unit is used to determine the first blood pressure value of the user according to the blood pressure values ​​of the user at various moments within a day and the adjustment factor corresponding to each moment.

7. The device according to claim 6, characterized in that The device also includes: A data statistics unit, used to count the number of each value and the total number of each blood pressure value at each time in a day of the user; a ratio determination unit, configured to determine a ratio of the number of each value to the total number of the blood pressure values ​​at each moment according to the number of each value in the blood pressure values ​​at each moment and the total number of the blood pressure values ​​at each value; An average posture data determination unit, used to determine the average body behavior posture data of the user on the day according to the body behavior posture data of the user collected at various times on the day; A second blood pressure value determination unit, configured to input the ratio of the number of each value to the total number of blood pressure values ​​of each value, the average body behavior posture data of the user on the day, and the user attribute information into a pre-trained blood pressure determination model to obtain a second blood pressure value, wherein the blood pressure determination model is trained by inputting a plurality of training samples into the first neural network, each of the training samples including the ratio of the number of each value in history to the total number of blood pressure values ​​at each moment, the average body behavior posture data of the user on the day in history, the user attribute information in history, and the corresponding second blood pressure value marked in history by an expert; The blood pressure value correction unit is used to correct the first blood pressure value according to the second blood pressure value.

8. The device according to claim 7, characterized in that The blood pressure value correction unit is specifically used to calculate the , correct the first blood pressure value, wherein, is the first blood pressure value before correction, is the second blood pressure value, is a preset first correction factor, and the value range of the preset first correction factor is 0.4-0.6; is a preset second correction factor, and the value range of the preset second correction factor is 0.4-0.6, , is the first blood pressure value after correction.

9. The device according to claim 6, characterized in that The first blood pressure value determination unit is specifically configured to determine the blood pressure value according to the formula , determine a first blood pressure value of the user, wherein, is the user's blood pressure value at the i-th moment within one day, is the adjustment factor corresponding to the blood pressure value at the i-th moment, is the first blood pressure value, and n is the number of the user's blood pressure values ​​at various times within a day.

10. The device according to claim 6, characterized in that The adjustment factor determination unit is specifically used to determine the adjustment factor corresponding to the blood pressure value at this moment based on the body behavior posture data within the previous half hour and the pre-trained adjustment factor model, wherein the adjustment factor model is trained by inputting multiple training samples into the second neural network, and each of the training samples includes the body behavior posture data within the previous half hour of history and the corresponding historical standard adjustment factor.

Images