CN113069108A

CN113069108A - User state monitoring method and device, electronic equipment and storage medium

Info

Publication number: CN113069108A
Application number: CN202110296722.6A
Authority: CN
Inventors: 田秀全; 王雨
Original assignee: Beijing Jingdong Tuoxian Technology Co Ltd
Current assignee: Beijing Jingdong Tuoxian Technology Co Ltd
Priority date: 2021-03-19
Filing date: 2021-03-19
Publication date: 2021-07-06

Abstract

The embodiment of the invention discloses a user state monitoring method, a user state monitoring device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring at least two target monitoring data for representing the user state of a target user; processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data; if the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, the user state of the target user is marked as a healthy state; and the theoretical monitoring statistic threshold is obtained by processing the sample index data in the sample data set based on a pre-constructed target data processing model. By the technical scheme of the embodiment of the invention, the user state is monitored by adopting multivariate data, so that the monitoring accuracy is improved, and the technical effect of user experience is also improved.

Description

User state monitoring method and device, electronic equipment and storage medium

Technical Field

The embodiment of the invention relates to the technical field of medical health, in particular to a user state monitoring method and device, electronic equipment and a storage medium.

Background

With the rise of internet hospitals, on-line medical treatment becomes a new development trend. Among the patients who visit the clinic, chronic patients such as hypertension, heart disease, diabetes, asthma, etc. need attention of physicians due to their long duration, the time sequence of disease development, etc.

Currently, the focus measures taken against the above-mentioned chronic diseases are: the equipment based on patient wears adopts the data of corresponding index to monitor corresponding index data, with when monitoring that certain index data is unsatisfactory, just remind the patient, for example, to the diabetes patient, when detecting that blood glucose exceeds the standard, can remind the user.

When the present invention is implemented based on the above-described embodiments, the inventors have found that the following problems occur:

when certain index data is detected to be not in accordance with the requirements, the patient is hospitalized, and the problem that the patient cannot be treated effectively in time due to the fact that the patient is not hospitalized timely is solved. Furthermore, because the patient is monitored by adopting the univariate data, inaccurate data monitoring exists, and when the patient is reminded based on inaccurate data, frequent reminding or lagging reminding exists, so that the technical problem of poor user experience is caused.

Disclosure of Invention

The invention provides a user state monitoring method and device, electronic equipment and a storage medium, which are used for monitoring a user state by adopting multivariate data, so that the monitoring accuracy is improved, and the technical effect of user experience is also improved.

In a first aspect, an embodiment of the present invention provides a user status monitoring method, where the method includes:

acquiring at least two target monitoring data for representing the user state of a target user;

processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data;

if the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, the user state of the target user is marked as a healthy state;

and the theoretical monitoring statistic threshold is obtained by processing the sample index data in the sample data set based on a pre-constructed target data processing model.

In a second aspect, an embodiment of the present invention further provides a user status monitoring device, where the device includes:

the target monitoring data acquisition module is used for acquiring at least two kinds of target monitoring data for representing the user state of a target user;

the target monitoring statistic determination module is used for processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data;

the health state marking module is used for marking the user state of the target user as a health state if the target monitoring statistic is within a theoretical monitoring statistic threshold range;

In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:

one or more processors;

a storage device for storing one or more programs,

when the one or more programs are executed by the one or more processors, the one or more processors implement the user status monitoring method according to any one of the embodiments of the present invention.

In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the user status monitoring method according to any one of the embodiments of the present invention.

According to the technical scheme of the embodiment of the invention, at least two kinds of target monitoring data are obtained, the target monitoring data are subjected to statistical analysis processing to comprehensively analyze the target monitoring data, the target monitoring statistic for measuring the user state is obtained, and if the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, the user state is marked as the healthy state, so that the problem that frequent reminding or delayed reminding of the user is caused due to inaccurate judgment when the user state is judged based on the change of single variable data in the prior art is solved, the user experience is poor, the user state is monitored by comprehensively considering the multivariate data, the monitoring accuracy is improved, the user is timely reminded, and the technical effects of timely diagnosis and user experience of the user are improved.

Drawings

In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, a brief description is given below of the drawings used in describing the embodiments. It should be clear that the described figures are only views of some of the embodiments of the invention to be described, not all, and that for a person skilled in the art, other figures can be derived from these figures without inventive effort.

Fig. 1 is a schematic flowchart of a user status monitoring method according to an embodiment of the present invention;

fig. 2 is a schematic flowchart of a user status monitoring method according to a second embodiment of the present invention;

fig. 3 is a schematic flowchart of a user status monitoring method according to a third embodiment of the present invention;

fig. 4 is a schematic flowchart of a user status monitoring method according to a fourth embodiment of the present invention;

fig. 5 is a schematic structural diagram of a user status monitoring apparatus according to a fifth embodiment of the present invention;

fig. 6 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a schematic flow chart of a user status monitoring method according to an embodiment of the present invention, where the present embodiment is applicable to determining a user status according to a processing result by analyzing and processing collected multiple kinds of index data, and the method may be executed by a user status monitoring device, where the device may be implemented in a form of software and/or hardware, where the hardware may be an electronic device, and the electronic device may be a mobile terminal, a PC terminal, and the like.

As shown in fig. 1, the method of this embodiment specifically includes the following steps:

s110, at least two kinds of target monitoring data used for representing the user state of the target user are obtained.

The target user can be a patient with chronic diseases, and the target user can monitor the physical health condition through the method of the embodiment. The user state may be a state used to characterize the health of the target user, such as: the health state refers to the state that the index data collected by the chronic patient is processed to determine that the chronic patient is not abnormal, and correspondingly, the non-health state refers to the state that the index data collected by the chronic patient is processed to determine that the chronic patient is about to be abnormal or possibly abnormal. The target monitoring data may be physiological parameter information required for health monitoring of a patient with chronic disease and used for characterizing body functions, such as blood pressure parameters, blood glucose parameters, heart rate parameters, etc.

Specifically, at least two kinds of physiological parameter information of the target user can be collected in real time through the information collection device, or at least two kinds of physiological parameter information of the target user can be collected periodically through the information collection device, for example: the collection is carried out once every 1 minute or once every 5 minutes, and the like, and the collection frequency can be set according to the requirements of target users. In order to improve the monitoring accuracy and timely obtain early warning when the health condition of the user goes wrong, a data acquisition mode for acquiring monitoring data in real time or acquiring the monitoring data periodically at a high frequency can be selected. Of course, if the target user wants to save the power consumption of the information acquisition device and improve the service life of the information acquisition device, a data acquisition mode for acquiring the monitoring data in a low-frequency cycle may be selected.

It should be noted that the advantage of selecting at least two target monitoring data is that at least two target monitoring data can be considered comprehensively, so as to improve the monitoring accuracy and avoid the problem that when any monitoring data is monitored to be not satisfactory but the health of the target user is not in a state, the target user is reminded.

And S120, processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data.

The target data processing model is an algorithm model obtained by carrying out statistical analysis according to sample index data in the sample data set, and is used for processing at least two kinds of target monitoring data to obtain corresponding processing results, and further determining whether the target monitoring data is in a reasonable health range according to the processing results. The target monitoring statistics is data obtained by processing target monitoring data through a target data processing model, and can be understood as statistical data for measuring the state of a target user.

Specifically, the target monitoring data is input into a predetermined target data processing model, and the target monitoring data is processed to obtain target monitoring statistics corresponding to the target monitoring data. And for each group of collected target monitoring data, target monitoring statistics corresponding to each group of target monitoring data can be obtained through processing of the target data processing model.

And S130, if the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, marking the user state of the target user as a healthy state.

The theoretical monitoring statistic threshold is obtained by processing sample index data in the sample data set based on a pre-constructed target data processing model, and the processing mode may be a statistical processing mode, for example: the Principal Component Analysis (PCA), Independent Component Analysis (ICA), Kernel Entropy Component Analysis (KECA), and the like can be used. The data in the sample data set may be monitoring data of a target user in a past period of time, or may be monitoring data of a plurality of users in a past period of time.

Specifically, when the target monitoring statistic corresponding to the target monitoring data is within the theoretical monitoring statistic threshold range, the target user can be considered to be in a healthy state at the current moment, and the user state of the target user is marked as a healthy state; when the target monitoring statistic corresponding to the target monitoring data is out of the range of the theoretical monitoring statistic threshold, the target user can be considered to be in an unhealthy state at the current moment, the user state of the target user is marked as the unhealthy state, and early warning prompt information can be sent to the target user, so that the target user can see a doctor in time.

If the target user is bound with a fixed doctor user, the user state and the monitoring data can be sent to the terminal equipment of the doctor user corresponding to the target user in real time or periodically, so that the doctor user can analyze the monitoring data of the target user and further know the physical health state of the target user, and the doctor user can remind the target user to see a doctor when the target user is in a non-health state.

It should be noted that the method of S120 and/or S130 in this embodiment may be integrated in a chip or a processor of the information collecting device, and the target monitoring data is processed and compared to determine the user status of the target user, so that the user status of the target user can be analyzed even in a network outage state, and an analysis interruption is avoided. The method of S120 and/or S130 of this embodiment may also be integrated on a terminal or a server, and the target monitoring data is collected by the information collecting device and uploaded to the terminal or the server, so as to complete statistics and comparative analysis on the target monitoring data on the terminal or the server, which has the advantage of improving the efficiency of data calculation and analysis.

It is further noted that the theoretical monitoring statistic threshold may be a value determined by statistical analysis based on the detected data of the target user over a past period of time. When the target monitoring statistic does not exceed the theoretical monitoring statistic threshold, marking the user state of the target user as a healthy state; when the target monitoring statistic exceeds the theoretical monitoring statistic threshold, the user state of the target user is marked as a non-healthy state, and the setting mode of the theoretical monitoring statistic threshold has the advantage of high matching degree with the target user.

It should be noted that the theoretical monitoring statistic threshold value may also be a plurality of values determined by statistical analysis based on the detected data of a plurality of target users in the past period of time, and a slightly larger theoretical monitoring statistic threshold value range may be determined according to the plurality of values. When the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, marking the user state of the target user as a healthy state; and when the target monitoring statistic is out of the range of the theoretical monitoring statistic threshold, marking the user state of the target user as a non-healthy state. The setting mode of the theoretical monitoring statistic threshold has the advantages of large statistical data amount and more accurate target data processing model.

Example two

Fig. 2 is a schematic flow chart of a user status monitoring method according to a second embodiment of the present invention, and the embodiment refers to the technical solution of this embodiment for determining the theoretical monitoring statistic threshold value based on the foregoing embodiments. The same or corresponding terms as those in the above embodiments are not explained in detail herein.

As shown in fig. 2, the method of this embodiment specifically includes the following steps:

and S210, determining a theoretical monitoring statistic threshold value.

To improve the accuracy of the theoretical monitoring statistic threshold, the theoretical monitoring statistic threshold may be determined using the sample data set and a pre-constructed target data processing model. The method of determining a theoretical monitoring statistic threshold may comprise the steps of:

the method comprises the steps of firstly, obtaining a sample data set, and carrying out standardization processing on multiple index sample data in the sample data set to obtain a standard sample data set.

The sample data set comprises a plurality of index sample data used for representing the user state, which can include index sample data of a target user, and can also include index sample data of other users with chronic diseases. It should be noted that the data in the sample data set is acquired based on the index data of the target user and/or other users with chronic diseases in the healthy state.

It should be further noted that the acquisition of the multiple kinds of index data of the chronic disease patient in the healthy state is to determine the theoretical monitoring statistic of the target user in the healthy state, and if the multiple kinds of index data are doped with the index data in the unhealthy state, the calculation of the theoretical monitoring statistic may be inaccurate, which affects the determination of the theoretical monitoring statistic threshold.

Acquiring various index data which indicate that the user state is a healthy state within a preset time length, and taking the various index data as a group of sample data in a sample data set.

The multiple kinds of index data which are collected within the preset time and represent that the user state is the health state can be used for constructing a sample data set, and the preset time can be a long period of time, for example: for 30 days, various index data can be acquired within a preset time length, for example, the index data is sampled 2 times per day within the preset time length, or the index data is sampled 10 times per day within the preset time length.

Specifically, various index data indicating that the user state is a healthy state within a preset time period can be acquired. Furthermore, a group of sample data is composed of a plurality of index data of the same sampling time, and all sample data can constitute a sample data set. In order to make the index data more intuitive, the acquired data may be presented in a tabular form, for example, an index data information table is constructed by taking the sampling time as a row header and the index type as a column header. The index data may also be counted in other forms, and is not particularly limited in this embodiment.

To improve the efficiency of the data normalization process, the data may be stored in office software, e.g., Excel. And the head of the Excel list is an index type, and the data of the same index type are stored in sequence.

In order to enhance the availability of the data in the sample data set, the data in the sample data set may be filtered and cleaned, that is, operations such as outdated data removal processing, incomplete data completion or elimination processing, and data de-duplication processing may be performed. Furthermore, the filtered data can be stored locally and/or in a cloud so as to be convenient for subsequent statistical analysis.

And determining a data mean value and a variance corresponding to the index data of the current list head aiming at the index data corresponding to the same list head in the sample data set.

The list header may represent the index data type, and the index data corresponding to the same list header may represent the index data in the same index type.

Specifically, in the sample data set, the index data corresponding to each list header is obtained, that is, the index data of each index type is obtained. And calculating to obtain the data mean and variance corresponding to the index data of each list head, namely determining the mean and variance of the index data corresponding to each index type.

Illustratively, the index type corresponding to the current list header is heart rate, and the heart rate data in the sample data set is 85, 89, 83, 92, 95, 87, 85, 82, 88 and 93 (unit: bmp). The mean value of the data corresponding to the index data of the current list head is 87.9, and the variance is 17.1.

And aiming at each group of sample data, determining the data standard value of the current index data according to the current index data, the corresponding data mean value and the variance.

The data standard value may be a value obtained by normalizing the current index data, and the normalization may be a difference between the current index data and a corresponding data mean value divided by a corresponding variance.

Specifically, the list heads (index types) corresponding to each index data in each group of sample data are different, the difference value is obtained by subtracting the data mean value corresponding to the current index data and the corresponding list head, and the data standard value of the current index data is obtained by dividing the difference value by the variance corresponding to the list head. And processing other index data in the same way to obtain the data standard value of each index data.

Illustratively, the type of the index corresponding to the current index data is heart rate, and accordingly, the mean value of the data corresponding to the heart rate at the head of the list is 87.9, and the variance is 17.1. If the current heart rate index data is 88, the data standard value of the current heart rate index data is 0.0058 through the following formula.

And aiming at each group of sample data, obtaining standard sample data according to the data standard value corresponding to each list head in the current group of sample data.

Specifically, for each group of sample data, data composed of data standard values corresponding to each list header in the current group of sample data is used as the standard sample data of the current group. It can be understood that, for each set of sample data, the data standard values corresponding to the indicator types in the current set of sample data form a set, and the set is used as the standard sample data of the current set.

And obtaining a standard sample data set based on the standard sample data corresponding to each group of sample data.

Specifically, the standard sample data set may be obtained by combining the standard sample data corresponding to each group of sample data, and the size of the standard sample data set may be consistent with the size of the sample data set.

And secondly, processing the standard sample data set based on a pre-constructed target data processing model to obtain a sample score matrix corresponding to the standard sample data set.

Wherein the sample score matrix is a matrix for measuring the importance of each component.

Specifically, the sample score matrix may be determined by the following steps.

(1) Determining a window probability density function based on the kernel function and the kernel function parameters.

The independent variable in the kernel function is standard sample data in the standard sample data set, and taking the KECA algorithm as an example, the kernel function may be a Mercer kernel function or the like.

Specifically, the standard sample data set may be recorded as a matrix form, i.e. a standard sample data matrix D (x)₁，x₂，…，x_N) Wherein x is₁Indicates the standard sample data, x, corresponding to the 1 st list header₂Indicates the standard sample data, x, corresponding to the 2 nd list header_NAnd the standard sample data corresponding to the Nth list head is shown, and N represents the total number of the columns in the standard sample data set.

Renyi entropy in the KECA algorithm is defined as

H(p)＝-log∫p²(x)dx

Where H is Renyi entropy and p (x) is the probability density function of the standard sample data matrix D.

According to the formula, since the logarithmic function is a monotonic function, only ^ p can be analyzed²(x) dx is, will ^ p-²(x) dx is denoted as V (p). To estimateThe value of v (p), the Parzen window probability density estimation function can be introduced:

wherein k is_σ(x，x_t) Representing the Mercer kernel, σ represents a parameter of the kernel, x_tRepresenting any column of data in the standard sample data matrix D.

(2) And carrying out sample mean approximation processing on the standard sample data in the standard sample data set through the window probability density function to obtain a data vector matrix corresponding to the standard sample data set.

Wherein each component in the data vector matrix corresponds to an estimate of the entropy of the feature vector.

And processing each index data and the data quantity of the index data according to the window probability density function to obtain a data vector matrix to be converted.

Specifically, each index data and the data amount of the index data are processed according to the window probability density function, and a data vector matrix to be converted corresponding to each index data can be obtained and recorded as

And performing characteristic decomposition on the kernel matrix in the kernel function to obtain a characteristic decomposition vector corresponding to the kernel matrix.

The feature decomposition vector is composed of a diagonal feature matrix and a column feature matrix.

Specifically, the kernel matrix may be denoted as K, the size of K is N × N, the Renyi entropy is estimated by using the eigenvalue and the eigenvector of the kernel matrix, and the kernel matrix K may be subjected to eigen decomposition to obtain an eigen decomposition vector, that is, K may be decomposed into:

wherein Λ represents a characteristic value λ of the kernel matrix K₁，λ₂，…，λ_NA diagonal feature matrix composed of E representing a feature vector E₁，e₂，…，e_NA composed column feature matrix.

And processing the data vector matrix to be converted based on the characteristic decomposition vector to obtain the data vector matrix.

In particular, since v (p) ═ p²(x) dx, obtaining a data vector matrix by means of sample mean approximation

Further, the above formula can be represented by a kernel matrix K as:

wherein K is an NxN kernel matrix; 1 is an N × 1 unit vector, and 1T is a transpose of the unit vector. That is, the kernel matrix K is:

further, it can be derived

From the above formula can be seen

Each corresponding to an estimate of the entropy of a feature vector, different feature values contributing differently to the Renyi entropy than the feature vector.

(3) And performing dimensionality reduction transformation on the data vector matrix to obtain a sample score matrix of the standard sample data set.

Wherein the dimension of the sample score matrix is smaller than the dimension of the data vector matrix.

And determining a diagonal feature matrix to be processed, which is formed by preset number of target feature values, according to the feature values in the diagonal feature matrix.

The preset number should be less than or equal to the dimension of the diagonal feature matrix, and may be denoted as k.

Specifically, the first k eigenvalues with larger eigenvalues in the diagonal eigenvalue matrix Λ are selected as principal elements, and eigenvectors corresponding to the k eigenvalues are determined, and then the k eigenvectors are used as the principal axis of the KECA to open a subspace U_mDetermining subspace U_mDimension m of (2) is the number of target eigenvalues, and m eigenvalues can be taken as target eigenvalues. The diagonal feature matrix can be formed according to the target feature value

Λ_m＝diag(λ₁，λ₂，...λ_m)

Wherein, Λ_mIs represented by a characteristic value lambda₁，λ₂，...，λ_mAnd forming a diagonal feature matrix to be processed.

And performing dimension reduction transformation on the diagonal feature matrix to be processed to obtain a sample score matrix of the standard sample data set.

In particular, a non-linear mapping from the input space to the kernel feature space may be defined

The representation form can be

In this case, it is possible to have

Further, an N-dimensional matrix of data vectors may be passed through

Performing mapping process, i.e. mapping the N-dimensional data vector matrix to the subspace U formed by k KECA main axes_mUpper (m)<N), a sample score matrix T can be obtained_m：

Wherein, Λ_mIs represented by a characteristic value lambda₁，λ₂，...，λ_mComposed diagonal feature matrix to be processed, E_m＝(e₁，e₂，...，e_m) Is represented by a feature vector e₁，e₂，…，e_mAnd forming a feature vector matrix.

It should be noted that the number k of the principal elements may be determined by using the cumulative variance contribution rate, that is, the variance contribution rates corresponding to the feature vectors of the feature values are sequentially calculated from large to small, and are accumulated, so that when the cumulative variance contribution rate reaches the threshold for the first time, for example: the contribution rate reaches 90%, and the number of corresponding characteristic values is used as the preset number, namely the number of the principal elements.

And thirdly, obtaining theoretical monitoring statistics based on the sample scoring matrix.

Firstly, a first theoretical monitoring statistic in the theoretical monitoring statistics is obtained based on the sample score matrix and the diagonal feature matrix to be processed.

Wherein the standard square sum of each score vector in the sample score matrix is used as the first theoretical monitoring statistic.

Specifically, the first theoretical monitoring statistic T²Can be obtained based on the following formula

T²＝T_mΛ_m ^-1T_m ^T

The first theoretical monitoring statistic T can be monitored using the F distribution²Is calculated based on the control limit of

Wherein n is the number of samples used to build the pivot model, k is the number of pivot elements retained in the model, F_k，n-1，αIs the critical value of the F distribution under the condition corresponding to the test level alpha, the degree of freedom k, n-1.

And secondly, obtaining a second theoretical monitoring statistic in the theoretical monitoring statistics based on the eigenvalue in the diagonal eigenvalue matrix and the eigenvalue in the sample score matrix.

Wherein the second theoretical monitoring statistic represents the variation of the data which is not explained by the principal component model, and can represent the error between the variation trend of each data sample and the statistical model.

Specifically, the second theoretical monitoring statistic SPE can be obtained based on the following formula

The control limit for the second theoretical monitoring statistic SPE may be calculated by

Wherein the content of the first and second substances,

C_αis a critical value for normal distribution at a check level of α. Lambda [ alpha ]_jAnd representing the eigenvalue of the diagonal feature matrix to be processed used when the target data processing model is established.

Finally, the first theoretical monitoring statistic and the second theoretical monitoring statistic can be used as theoretical monitoring statistics, and the first theoretical monitoring statistic control limit and the second theoretical monitoring statistic control limit can be used as theoretical monitoring statistic threshold range.

S220, acquiring at least two target monitoring data for representing the user state of the target user.

And S230, processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data.

And S240, if the target monitoring statistic is within the range of the theoretical monitoring statistic threshold, marking the user state of the target user as a healthy state.

According to the technical scheme of the embodiment of the invention, the KECA is used for processing the data in the standard sample data set to obtain the sample score matrix, and further, the sample score matrix is processed by a statistical analysis method to obtain theoretical detection statistics to accurately determine the range of the theoretical monitoring statistics threshold, so that the problem of inaccurate judgment caused by judging whether the monitoring data is the data in a healthy state by using a single threshold in the prior art is solved, the accuracy of the theoretical detection statistics threshold is improved based on the KECA and the statistical analysis, the accuracy of user state judgment is improved, and the user experience is improved.

EXAMPLE III

Fig. 3 is a schematic flow chart of a user state monitoring method according to a third embodiment of the present invention, and reference may be made to the technical solution of the present embodiment for a determination method of target monitoring statistics and a specific method of user state determination in the embodiment based on the foregoing embodiments. The same or corresponding terms as those in the above embodiments are not explained in detail herein.

As shown in fig. 3, the method of this embodiment specifically includes the following steps:

s310, at least two kinds of target monitoring data used for representing the user state of the target user are obtained.

Specifically, physiological parameter information of a target user is acquired based on information acquisition equipment, and the physiological parameter information is subjected to standardization processing to obtain target monitoring data for representing the state of the user.

It can be understood that the physiological parameter information of the target user acquired based on the information acquisition device is classified according to different index types, namely, a list header. And further, determining target monitoring data according to the current physiological parameter information and the data mean and variance corresponding to the corresponding list head. Can be as follows: the difference between the current physiological parameter information and the corresponding data mean is divided by the corresponding variance.

In order to accurately measure the user state of the target user, the target monitoring data may include at least two of a blood pressure parameter, a blood oxygen saturation parameter, a blood glucose parameter, a heart rate parameter, a body fat rate parameter, a uric acid parameter, a urine routine parameter, a total cholesterol parameter, or a body temperature parameter. The units of the physiological parameter information variables are shown in table 1.

TABLE 1 physiological parameter information variables

And S320, analyzing and processing the target monitoring data based on the target data processing model to obtain a target health score vector corresponding to the target monitoring data.

Specifically, the target monitoring data is input into the target data processing model, and the target monitoring data may be subjected to the KECA statistical processing, that is, data (target monitoring data) outside the sample data set passes through

Is projected to U_mTo satisfy

Wherein the content of the first and second substances,

T′_meach vector in (a) as a target health score vector may be represented as

S330, processing the target health score vector based on the F distribution to obtain a first target monitoring statistic in the target monitoring statistics.

In particular, it can be based on the targetThe health score vector and the sample score matrix determine a first target monitoring statistic. Further, the first target monitoring statistic can be compared to the first theoretical monitoring statistic T using the F distribution²Are compared.

For a new sample point, the target health score vector, the first target monitoring statistic T²When the value of (2) is less than the control limit value, the target health score vector and the variables of the data vector matrix obey the same statistical distribution; on the contrary, it is indicated that the target health score vector and the data vector matrix do not obey the same statistical distribution, and the health condition of the target user may be abnormal at this time.

S340, processing each vector value in the target health score vectors to obtain a second target monitoring statistic in the target monitoring statistics.

Specifically, a second target monitoring statistic may be determined according to each vector value in the target health score vector, and then the second target monitoring statistic may be compared with the control limit of the second theoretical monitoring statistic SPE.

If the second target monitoring statistic of the target health score vector does not exceed the upper limit, the target user is in a healthy state currently; otherwise, the target user is currently in an unhealthy state and needs to be warned.

And S350, if the first target monitoring statistic is within a first theoretical monitoring statistic threshold value in the theoretical monitoring statistics and the second target monitoring statistic is within a second theoretical monitoring statistic threshold value range in the theoretical monitoring statistics, marking the user state of the target user as a healthy state.

Specifically, if the first target monitoring statistic conforms to the first theoretical monitoring statistic threshold range, it may be considered that the current target monitoring data of the target user and the sample index data in the sample data set belong to the same statistical distribution, and it may be considered that the user state of the current target user is a healthy state. If the second target monitoring statistic conforms to the second theoretical monitoring statistic threshold range, the current target monitoring data of the target user can be considered to be in the error range of the diagonal feature matrix to be processed used by the target data processing model, and the user state of the current target user can be considered to be a healthy state.

In order to avoid missing the marking and early warning of the unhealthy state of the target user, when the first target monitoring statistic and the second target monitoring statistic are within the threshold range of the theoretical monitoring statistic, marking the user state of the target user as a healthy state; and when at least one target monitoring statistic is out of the theoretical detection statistic threshold value, marking the user state of the target user as a non-healthy state, and carrying out early warning.

According to the technical scheme of the embodiment of the invention, the target health score vector obtained after target monitoring data are processed is subjected to statistical analysis based on KECA to obtain the first target monitoring statistic and the second target monitoring statistic, and if the first target monitoring statistic and the second target monitoring statistic are both within the range of the theoretical monitoring statistic threshold, the user state is marked as the healthy state, so that the problem that the user state is judged based on the change of univariate data in the prior art, the judgment is inaccurate, the frequent reminding or delayed reminding of the user is caused, and the user experience is poor is solved, the user state is monitored by comprehensively considering the multivariate data, the monitoring accuracy is improved, the user is reminded in time, and the technical effects of timely diagnosis and user experience of the user are improved.

Example four

As an optional implementation of the foregoing embodiments, fig. 4 is a schematic flowchart of a user status monitoring method provided in a fourth embodiment of the present invention. The same or corresponding terms as those in the above embodiments are not explained in detail herein.

As shown in fig. 4, the user status detection method can be divided into an offline modeling phase and an online monitoring phase.

An off-line modeling stage:

(1) and acquiring a sample data set, wherein the data in the sample data set is various index data in a healthy state. Various index data in the sample data set can be used as normal training data.

(2) And carrying out standardization processing on the normal training data, determining the mean value and the variance of each index type, and determining a standard sample data set.

(3) Based on the given kernel function and the parameters of the kernel function, a KECA model is constructed according to a standard sample data set, and a sample score matrix T is extracted_m＝[t₁，t₂，…，t_m]。

(4) Calculating the statistic of normal training data, namely a first theoretical monitoring statistic T according to the sample scoring matrix²And a second theoretical monitoring statistic SPE.

(5) And calculating to obtain the statistical quantity control limit according to the statistical quantity. That is, a first theoretical monitoring statistic control limit is determined from the first theoretical monitoring statistic, and a second theoretical monitoring statistic control limit is determined from the second theoretical monitoring statistic.

In the on-line monitoring stage, namely in the specific application stage:

(1) and acquiring physiological parameter information of a target user, and taking the physiological parameter information as online test data.

(2) And (3) carrying out standardization processing on the online test data according to the mean value and the variance of each index type determined in the offline modeling stage (2) and a standardization processing method to obtain target monitoring data.

(3) Processing target monitoring data through the KECA model constructed in the offline modeling stage (3) to obtain a target health score vector t'_i。

(4) And calculating monitoring statistics of the online test data according to the target health score vector, namely a first target monitoring statistic and a second target monitoring statistic.

(5) The first target monitoring statistic is compared to a first theoretical monitoring statistic control limit, and the second target monitoring statistic is compared to a second theoretical monitoring statistic control limit. If at least one target monitoring statistic exceeds the theoretical monitoring statistic control limit, the target user is proved to have a health problem, namely the user state is determined to be in a non-healthy state, a reminding prompt can be generated to remind the user at the moment, otherwise, the user state of the target user is considered to be in a healthy state, namely, the target user is in a healthy state, and the monitoring is returned to continue.

According to the technical scheme of the embodiment of the invention, in an off-line modeling stage, a KECA model is constructed based on a sample data set, and a statistical quantity control limit is calculated; in the on-line monitoring stage, a target health score vector is determined based on the physiological parameter information and the constructed KECA model, and then monitoring statistics are obtained. Whether a target user has a health problem is judged according to the relation between monitoring statistics and statistics control limit, the problem that in the prior art, when the user state is judged based on the change of single variable data, judgment is inaccurate, frequent reminding or delayed reminding of the user is caused, and user experience is poor is solved, the user state is monitored by comprehensively considering multivariable data, the monitoring accuracy is improved, the user is timely reminded, and the technical effects of timely diagnosis and user experience of the user are improved.

EXAMPLE five

Fig. 5 is a schematic structural diagram of a user status monitoring device according to a fifth embodiment of the present invention, where the device includes: a target monitoring data acquisition module 510, a target monitoring statistics determination module 520, and a health status flag module 530.

The target monitoring data obtaining module 510 is configured to obtain at least two types of target monitoring data used for representing a user state of a target user; a target monitoring statistic determination module 520, configured to process the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data; a health status labeling module 530, configured to label the user status of the target user as a health status if the target monitoring statistic is within a theoretical monitoring statistic threshold range; and the theoretical monitoring statistic threshold is obtained by processing the sample index data in the sample data set based on a pre-constructed target data processing model.

Optionally, the target monitoring data obtaining module 510 is specifically configured to collect physiological parameter information of the target user based on an information collecting device, and perform standardization processing on the physiological parameter information to obtain target monitoring data for representing a user state.

Optionally, the target monitoring data includes at least two of a blood pressure parameter, a blood oxygen saturation parameter, a blood glucose parameter, a heart rate parameter, a body fat rate parameter, a uric acid parameter, a urine routine parameter, a total cholesterol parameter, or a body temperature parameter.

Optionally, the target monitoring statistic determining module 520 is specifically configured to analyze and process the target monitoring data based on the target data processing model to obtain a target health score vector corresponding to the target monitoring data; processing the target health score vector based on the F distribution to obtain a first target monitoring statistic in the target monitoring statistics; and processing each vector value in the target health score vectors to obtain a second target monitoring statistic in the target monitoring statistics.

Optionally, the target monitoring statistic determining module 520 is further configured to mark the user status of the target user as a healthy status if the first target monitoring statistic and the second target monitoring statistic are both within the theoretical monitoring statistic threshold range.

Optionally, the apparatus further comprises: the theoretical monitoring statistic threshold value determining module is used for determining the theoretical monitoring statistic threshold value; the theoretical monitoring statistic threshold value determining module is specifically used for acquiring a sample data set and carrying out standardized processing on multiple index sample data in the sample data set to obtain a standard sample data set; the sample data set comprises a plurality of index sample data used for representing the user state; processing the standard sample data set based on a pre-constructed target data processing model to obtain a sample score matrix corresponding to the standard sample data set; and obtaining the theoretical monitoring statistic based on the sample scoring matrix.

Optionally, the theoretical monitoring statistic threshold determining module is further configured to acquire multiple kinds of index data indicating that the user state is a healthy state within a preset time period, and use the multiple kinds of index data as a group of sample data in the sample data set; determining a data mean value and a variance corresponding to the index data of the current list head aiming at the index data corresponding to the same list head in the sample data set; for each group of sample data, determining a data standard value of the current index data according to the current index data, the corresponding data mean value and the variance; for each group of sample data, obtaining standard sample data according to the data standard value corresponding to each list head in the current group of sample data; and obtaining the standard sample data set based on the standard sample data corresponding to each group of sample data.

Optionally, the theoretical monitoring statistic threshold determining module is further configured to determine a window probability density function based on the kernel function and the kernel function parameter; wherein the argument in the kernel function is standard sample data in the standard sample data set; carrying out sample mean value approximation processing on the standard sample data in the standard sample data set through the window probability density function to obtain a data vector matrix corresponding to the standard sample data set; obtaining a sample score matrix of the standard sample data set by performing dimensionality reduction transformation on the data vector matrix; wherein the dimensions of the sample score matrix are smaller than the dimensions of the data vector matrix.

Optionally, the theoretical monitoring statistic threshold determining module is further configured to process each index data and the data amount of the index data according to the window probability density function to obtain a to-be-converted data vector matrix; performing characteristic decomposition on a kernel matrix in the kernel function to obtain a characteristic decomposition vector corresponding to the kernel matrix; wherein the eigen decomposition vector is composed of a diagonal eigen matrix and a column eigen matrix; and processing the data vector matrix to be converted based on the characteristic decomposition vector to obtain a data vector matrix.

Optionally, the theoretical monitoring statistic threshold determining module is further configured to determine, according to the eigenvalue in the diagonal eigenvalue matrix, a diagonal eigenvalue matrix to be processed, where the diagonal eigenvalue matrix is composed of a preset number of target eigenvalues; and performing dimension reduction transformation on the diagonal feature matrix to be processed to obtain a sample score matrix of the standard sample data set.

Optionally, the theoretical monitoring statistic threshold determining module is further configured to obtain a first theoretical monitoring statistic in the theoretical monitoring statistics based on the sample score matrix and the diagonal feature matrix to be processed; and obtaining a second theoretical monitoring statistic in the theoretical monitoring statistics based on the eigenvalue in the diagonal eigenvalue matrix and the eigenvalue in the sample score matrix.

The user state monitoring device provided by the embodiment of the invention can execute the user state monitoring method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

It should be noted that, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiment of the invention.

EXAMPLE six

Fig. 6 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention. FIG. 6 illustrates a block diagram of an exemplary electronic device 60 suitable for use in implementing embodiments of the present invention. The electronic device 60 shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.

As shown in fig. 6, the electronic device 60 is in the form of a general purpose computing device. The components of the electronic device 60 may include, but are not limited to: one or more processors or processing units 601, a system memory 602, and a bus 603 that couples various system components including the system memory 602 and the processing unit 601.

Bus 603 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Electronic device 60 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by electronic device 60 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 602 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)604 and/or cache memory 605. The electronic device 60 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 606 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 603 by one or more data media interfaces. System memory 602 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 608 having a set (at least one) of program modules 607 may be stored, for example, in system memory 602, such program modules 607 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 607 generally perform the functions and/or methods of the described embodiments of the invention.

Electronic device 60 may also communicate with one or more external devices 609 (e.g., keyboard, pointing device, display 610, etc.), with one or more devices that enable a user to interact with electronic device 60, and/or with any devices (e.g., network card, modem, etc.) that enable electronic device 60 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 611. Also, the electronic device 60 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 612. As shown, the network adapter 612 communicates with the other modules of the electronic device 60 via the bus 603. It should be appreciated that although not shown in FIG. 6, other hardware and/or software modules may be used in conjunction with electronic device 60, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processing unit 601 executes various functional applications and data processing by running programs stored in the system memory 602, for example, implementing the user status monitoring method provided by the embodiment of the present invention.

EXAMPLE seven

An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a user status monitoring method.

The method comprises the following steps:

Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for embodiments of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for monitoring user status, comprising:

2. The method of claim 1, wherein obtaining at least two target monitoring data characterizing a user status of a target user comprises:

and acquiring physiological parameter information of the target user based on information acquisition equipment, and carrying out standardization processing on the physiological parameter information to obtain target monitoring data for representing the user state.

3. The method of claim 1 or 2, wherein the target monitoring data comprises at least two of a blood pressure parameter, a blood oxygen saturation parameter, a blood glucose parameter, a heart rate parameter, a body fat rate parameter, a uric acid parameter, a urine routine parameter, a total cholesterol parameter, or a body temperature parameter.

4. The method of claim 1, wherein the processing the target monitoring data based on a predetermined target data processing model to obtain target monitoring statistics corresponding to the target monitoring data comprises:

analyzing and processing the target monitoring data based on the target data processing model to obtain a target health score vector corresponding to the target monitoring data;

processing the target health score vector based on the F distribution to obtain a first target monitoring statistic in the target monitoring statistics;

and processing each vector value in the target health score vectors to obtain a second target monitoring statistic in the target monitoring statistics.

5. The method of claim 4, wherein said flagging the user status of the target user as healthy if the target monitoring statistic is within a theoretical monitoring statistic threshold range comprises:

marking the user status of the target user as healthy if the first target monitoring statistic is within a first theoretical monitoring statistic threshold of the theoretical monitoring statistics and the second target monitoring statistic is within a second theoretical monitoring statistic threshold range of the theoretical monitoring statistics.

6. The method of claim 1, further comprising:

determining the theoretical monitoring statistic threshold value;

the determining the theoretical monitoring statistic threshold value comprises:

acquiring a sample data set, and carrying out standardized processing on multiple index sample data in the sample data set to obtain a standard sample data set; the sample data set comprises a plurality of index sample data used for representing the user state;

processing the standard sample data set based on a pre-constructed target data processing model to obtain a sample score matrix corresponding to the standard sample data set;

and obtaining the theoretical monitoring statistic based on the sample scoring matrix.

7. The method according to claim 6, wherein the obtaining a sample data set and performing normalization processing on multiple index sample data in the sample data set to obtain a standard sample data set comprises:

acquiring various index data which indicates that the user state is a healthy state within a preset time length, and taking the various index data as a group of sample data in the sample data set;

determining a data mean value and a variance corresponding to the index data of the current list head aiming at the index data corresponding to the same list head in the sample data set;

for each group of sample data, determining a data standard value of the current index data according to the current index data, the corresponding data mean value and the variance;

for each group of sample data, obtaining standard sample data according to the data standard value corresponding to each list head in the current group of sample data;

and obtaining the standard sample data set based on the standard sample data corresponding to each group of sample data.

8. The method of claim 6, wherein the processing the standard sample data set based on a pre-constructed target data processing model to obtain a sample score matrix corresponding to the standard sample data set comprises:

determining a window probability density function based on the kernel function and the kernel function parameters; wherein the argument in the kernel function is standard sample data in the standard sample data set;

carrying out sample mean value approximation processing on the standard sample data in the standard sample data set through the window probability density function to obtain a data vector matrix corresponding to the standard sample data set;

obtaining a sample score matrix of the standard sample data set by performing dimensionality reduction transformation on the data vector matrix; wherein the dimensions of the sample score matrix are smaller than the dimensions of the data vector matrix.

9. The method of claim 8, wherein performing a sample mean approximation on the standard sample data in the standard sample data set through the window probability density function to obtain a data vector matrix corresponding to the standard sample data set, includes:

processing each index data and the data quantity of the index data according to the window probability density function to obtain a data vector matrix to be converted;

performing characteristic decomposition on a kernel matrix in the kernel function to obtain a characteristic decomposition vector corresponding to the kernel matrix; wherein the eigen decomposition vector is composed of a diagonal eigen matrix and a column eigen matrix;

and processing the data vector matrix to be converted based on the characteristic decomposition vector to obtain a data vector matrix.

10. The method according to claim 9, wherein obtaining the sample score matrix of the standard sample data set by performing a dimension reduction transformation on the data vector matrix comprises:

determining a diagonal feature matrix to be processed, which is formed by a preset number of target feature values, according to the feature values in the diagonal feature matrix;

11. The method of claim 10, wherein deriving the theoretical monitoring statistic based on the sample scoring matrix comprises:

obtaining a first theoretical monitoring statistic in the theoretical monitoring statistics based on the sample score matrix and the diagonal feature matrix to be processed;

and obtaining a second theoretical monitoring statistic in the theoretical monitoring statistics based on the eigenvalue in the diagonal eigenvalue matrix and the eigenvalue in the sample score matrix.

12. A user condition monitoring device, comprising:

13. An electronic device, characterized in that the electronic device comprises:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the user status monitoring method of any one of claims 1-11.

14. A storage medium containing computer executable instructions for performing the user status monitoring method of any one of claims 1-11 when executed by a computer processor.