Disclosure of Invention
In view of the above problems in the prior art, the present invention provides an intelligent pregnancy-induced hypertension monitoring device, which utilizes a vital sign standard value estimation module to determine the vital sign standard value of a monitored person by similarity matching with samples in a sample library according to the basic information of the monitored person and the normal value of the medically prescribed vital sign; and (3) taking the standard value of the vital sign as a control limit, and predicting and early warning the risk of pregnancy-induced hypertension of the monitor by acquiring the parameters of the vital sign of the monitor in real time. The system realizes targeted monitoring and early warning, and has high efficiency and high accuracy.
In order to achieve the purpose, the invention adopts the following technical scheme:
an intelligent pregnancy-induced hypertension syndrome monitoring device comprises an information input module, a data acquisition module, a vital sign standard value estimation module, a data storage module, a pregnancy-induced hypertension syndrome risk prediction module, an alarm module and a result display module;
the information input module is used for inputting basic information of a monitor;
the data acquisition module is used for acquiring vital sign parameters of a monitor;
the vital sign standard value estimation module is used for estimating the vital sign standard value of the monitored person according to the basic information of the monitored person and the normal value of the vital sign specified by medicine;
the data storage module is used for storing the data of the information input module and the data acquisition module and the vital sign standard value of the monitor;
the pregnancy-induced hypertension syndrome risk prediction module is used for predicting the pregnancy-induced hypertension syndrome risk of the monitor through the pregnancy-induced hypertension syndrome risk prediction model according to the basic information of the monitor, the vital sign parameters of the monitor and the vital sign standard values of the monitor;
the alarm module is used for carrying out early warning according to the pregnancy-induced hypertension risk of the monitored person;
and the result display module is used for displaying the pregnancy-induced hypertension risk of the monitored person and the vital sign parameters of the monitored person.
Further, the vital sign parameters include blood pressure data, pulse data, and heart rate data.
Furthermore, the blood vessel data of the monitored person are collected through blood vessel monitoring equipment, the blood vessel data comprise blood vessel blood pressure, blood vessel sclerosis index and blood vessel resistance index, the pulse data are collected through a pulse meter, and the heart rate data are collected through heart rate detection equipment.
Further, the basic information of the monitor includes name, age, weight, height, eating habit, medical history, family condition, medication condition and exercise condition of the monitor.
Further, the vital sign standard value estimation module comprises an estimation module and a sample library, and the estimation method of the vital sign standard value comprises the following steps:
s101, acquiring basic information of a monitor, and determining the weight of each index in the basic information;
s102, calculating the similarity between each index in the basic information of the monitor and each index in the sample library;
s103, acquiring the overall similarity between the basic information of the monitor and the sample library according to the weight of each index and the similarity of each index;
and S104, taking the vital sign standard value of the sample closest to the overall similarity of the basic information of the monitor in the sample library as the vital sign standard value of the monitor.
Further, the basic information of the sample and the corresponding vital sign standard value are stored in the sample library.
Further, the vital sign standard values of the monitored person include a vital sign standard value from 8 am to 8 pm, and a vital sign standard value from 8 pm to 8 pm.
Further, the pregnancy-induced hypertension syndrome risk prediction module predicts the pregnancy-induced hypertension syndrome risk of the monitored person through a pregnancy-induced hypertension syndrome risk prediction model, and comprises the following steps:
s201, acquiring on-line vital sign parameter monitoring data;
s202, establishing an autoregressive moving average fitting model of monitoring data when a monitor is in a normal state;
s203, calculating a residual error between the monitoring value and the vital sign standard value, and identifying the risk of pregnancy induction of the monitored person by using the residual error;
s204, forming a time sequence based on the monitoring data to estimate all potential risk points;
s205, analyzing the self-organization criticality of the monitor by taking the residual value as the characteristic quantity;
s206, establishing a control chart of the residual error according to the self-organization criticality;
and S207, alarming when the residual error value is out of limit, and realizing prediction of the risk of pregnancy-induced hypertension of the monitored person.
Advantageous effects
Compared with the prior art, the intelligent pregnancy-induced hypertension monitoring device provided by the invention has the following beneficial effects:
(1) the intelligent pregnancy-induced hypertension syndrome monitoring device provided by the invention comprises an information input module, a data acquisition module, a vital sign standard value estimation module, a data storage module, a pregnancy-induced hypertension syndrome risk prediction module, an alarm module and a result display module. Basic information of a monitor is input through an information input module, and then a vital sign standard value estimation module estimates a vital sign standard value of the monitor according to the basic information of the monitor and a medical specified vital sign normal value range; during actual monitoring, the data acquisition module acquires the vital sign parameters of the monitor in real time, and the pregnancy-induced hypertension syndrome risk prediction module monitors and warns the pregnancy-induced hypertension syndrome risk of the monitor according to the basic information of the monitor, the vital sign parameters of the monitor and the vital sign standard value of the monitor. The intelligent pregnancy-hypertension monitoring device provided by the invention can realize accurate and efficient monitoring and early warning of pregnancy-hypertension risks.
(2) In the intelligent pregnancy-induced hypertension monitoring device provided by the invention, the vital sign standard value estimation module comprises an estimation module and a sample library, wherein basic information of a sample and a corresponding vital sign standard value are stored in the sample library. The vital sign standard value estimation module firstly determines the weight of each index in the basic information according to the basic information of a monitor; then calculating the similarity between each index in the basic information of the monitor and each index in the sample library; then, acquiring the overall similarity between the basic information of the monitor and the sample library according to the weight of each index and the similarity of each index; and finally, taking the vital sign standard value of the sample closest to the overall similarity of the basic information of the monitor in the sample library as the vital sign standard value of the monitor. Aiming at the individual characteristics of different monitors, different vital sign standard values are determined, and monitoring errors caused by individual differences are eliminated, so that the monitoring effectiveness and accuracy are remarkably improved.
(3) The vital sign standard value estimation module of the intelligent pregnancy-hypertension monitoring device provided by the invention obtains the weight of each index through the particle swarm optimization algorithm, the accuracy is higher, the accuracy of the monitoring result is further improved, and the problems that the subjectivity of a traditional method utilizing an expert review method or an expert survey method is strong and the traditional method excessively depends on the experience of field experts are solved.
(4) The intelligent pregnancy-induced hypertension syndrome monitoring device provided by the invention firstly establishes an autoregressive moving average fitting model of monitoring data when a monitor is in a normal state, and then identifies the pregnancy-induced hypertension syndrome risk of the monitor by calculating the residual error between the monitoring value and the vital sign standard value, so that the intelligent pregnancy-induced hypertension syndrome monitoring device is more effective than the intelligent pregnancy-induced hypertension syndrome monitoring device which directly judges the pregnancy-induced hypertension syndrome risk from the monitoring value.
Detailed Description
The technical solutions of the embodiments of the present invention will be described clearly and completely below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments; all other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without any inventive step, are within the scope of the present invention.
Referring to fig. 1, the present invention provides an intelligent pregnancy-induced hypertension syndrome monitoring device, which comprises an information input module, a data acquisition module, a vital sign standard value estimation module, a data storage module, a pregnancy-induced hypertension syndrome risk prediction module, an alarm module, and a result display module. The information input module, the data acquisition module and the vital sign standard value estimation module are respectively connected with the data storage module, the information input module is also connected with the vital sign standard value estimation module, the data storage module is connected with the pregnancy-induced hypertension risk prediction module, and the pregnancy-induced hypertension risk prediction module is respectively connected with the alarm module and the result display module.
And the information input module is used for inputting the basic information of the monitor.
Preferably, the basic information of the monitor includes name, age, weight, height, eating habits, medical history, family condition, medication condition, exercise condition, etc. of the monitor, which are sequentially recorded as 1 st, 2 nd, … th indexes. The monitoring person has different basic information such as age, weight, height, eating habits, medical history, family condition, medication condition, exercise condition, etc., and generally has different individual states, and also has different values of vital sign parameters for the risk of pregnancy-induced hypertension. Therefore, the basic information of the monitoring person is input, so that the monitoring of pregnancy duration can be performed in a targeted manner according to the individual state of the monitoring person.
And the data acquisition module is used for acquiring the vital sign parameters of the monitor.
Preferably, the vital sign parameters include blood pressure data, pulse data and heart rate data.
Preferably, the blood vessel data of the monitored person is acquired by a blood vessel monitoring device, the blood vessel data comprises blood vessel pressure, vascular sclerosis index and vascular resistance index, the pulse data is acquired by a pulse meter, and the heart rate data is acquired by a heart rate detection device.
And the vital sign standard value estimation module is used for estimating the vital sign standard value of the monitored person according to the basic information of the monitored person and the medical normal value of the vital sign.
The medically prescribed vital sign normal values are: in the blood pressure data, the normal value of systolic pressure is 90-139 mmHg, the normal value of diastolic pressure is 60-89 mmHg, the normal value of pulse is 60-100 times/min, and the normal value of heart rate is 60-100 times/min.
Preferably, the samples in the sample library store basic information of the samples and corresponding vital sign standard values.
And the data storage module is used for storing the data of the information input module and the data acquisition module and the vital sign standard value of the monitor.
The pregnancy-induced hypertension syndrome risk prediction module is used for predicting the pregnancy-induced hypertension syndrome risk of the monitor through the pregnancy-induced hypertension syndrome risk prediction model according to the basic information of the monitor, the vital sign parameters of the monitor and the vital sign standard values of the monitor.
The alarm module is used for carrying out early warning according to the pregnancy-induced hypertension risk of the monitored person.
And the result display module is used for displaying the pregnancy-induced hypertension risk of the monitored person and the blood vessel data, the pulse data and the heart rate data of the monitored person.
Referring to fig. 2, the vital sign standard value estimation module includes an estimation module and a sample library, and the estimation method of the vital sign standard value includes the following steps:
s101, acquiring basic information of a monitor, and determining the weight of each index in the basic information;
wherein, the weight of each index is obtained by a particle swarm optimization algorithm, and the weight matrix of each index in the basic information is shown as formula (1):
W=[W1,W2,L,Wk](1)
in the formula, Wi(i ═ 1,2, …, k) represents the weight of the i-th index, and the matrix is used as one particle in the particle swarm optimization, and n particles are taken to constitute the particle swarm. Using a monitor prediction accuracy vector calculated by a first-time pass relation weighted K-nearest neighbor algorithm as an individual extreme value; and taking the extreme value with the highest accuracy in the individuals as a global extreme value, and taking the corresponding weight matrix as an optimal weight matrix. The specific acquisition process is as follows:
1. initializing the positions and the speeds of n particles by using random numbers, and setting a termination condition for the particle swarm optimization algorithm, wherein the termination condition is as follows: maximum iteration times are m times;
2. substituting the corresponding matrix of each particle into a relational weighting K-nearest neighbor algorithm, calculating the predicted accuracy of each monitor, and updating the individual extreme value and the global extreme value according to the accuracy;
3. updating the position and the speed of each particle according to the formulas (2) and (3);
vid k+1=vid k+c1random1 k(pbestid k-wid k)+c2random2 k(gbestd k-wid k) (2)
wid k+1=wid k+vid k+1(3)
in the formula, random1、random2Are each [0,1]A random number in between; v. ofid kIs the velocity of particle i in dimension d in the kth iteration; w is aid kIs the current position of particle i in the d-th dimension in the k-th iteration; c. C1、c2Two learning factors are respectively taken as c1、c2∈[0,4];pbestidIs the position of the particle i in the individual extreme point of the d-dimension; gbestdThe position of the global extreme point of the whole particle swarm in the d-dimension is taken as the position of the global extreme point.
(4) And (5) outputting W if the termination condition is met, otherwise, continuing to execute the step (2).
S102, calculating the similarity s between each index in the basic information of the monitor and each index in the sample library according to the formula (4);
in the formula, b and c are two thresholds of the value range of the index energy balance value respectively; a isr(x) The energy balance value of the r (r ═ 1,2, …, k) th index of the monitored x; a isr(xi) Is a sample xiAnd (i ═ 1,2, …, n) energy balance value of the r (r ═ 1,2, …, k) index.
S103, obtaining the overall similarity S between the basic information of the monitor and the sample library according to the weight of each index and the similarity of each index, as shown in formula (5):
in the formula, wrThe r (r ═ 1,2, …, k) index is weighted, and a smaller S indicates a higher similarity between the monitor and the sample.
And S104, taking the vital sign standard value of the sample closest to the overall similarity of the basic information of the monitor in the sample library as the vital sign standard value of the monitor.
Preferably, the vital sign standard values of the monitored person include a vital sign standard value from 8 am to 8 pm, and a vital sign standard value from 8 pm to 8 pm. In view of differences of vital sign parameters of a monitor in different time periods in a day, different vital sign standard values are set for different time periods, so that monitoring accuracy is improved.
Through the calculation of the vital sign standard values, different vital sign standard values can be determined according to individual characteristics of different monitors, and monitoring errors caused by individual differences are eliminated, so that the monitoring effectiveness and accuracy are remarkably improved.
Referring to fig. 3, the pregnancy-induced hypertension syndrome predicting module predicts the risk of pregnancy-induced hypertension syndrome of the monitored person through a pregnancy-induced hypertension syndrome risk prediction model, and comprises the following steps:
s201, acquiring on-line vital sign parameter monitoring data;
s202, establishing an autoregressive moving average fitting model of monitoring data when a monitor is in a normal state;
s203, calculating a residual error between the monitoring value and the vital sign standard value, and identifying the risk of pregnancy induction of the monitored person by using the residual error;
s204, forming a time sequence based on the monitoring data to estimate all potential risk points
S205, analyzing the self-organization criticality of the monitor by taking the residual value as the characteristic quantity;
s206, establishing a control chart of the residual error according to the self-organization criticality;
and S207, alarming when the residual error value is out of limit, and realizing prediction of the risk of pregnancy-induced hypertension of the monitored person.
The prediction model for risk of Pregnancy Induced Hypertension (PIH) based on-line monitoring data is characterized by that it inputs the time series formed from monitoring data mainly containing normal state of monitor, PIH state monitoring data time series and real-time monitoring data time series, and outputs a control chart with residual error sequence. The three most critical modules in the model are the establishment of an autoregressive moving average fitting model of normal state monitoring data, a calculation method of a residual error critical point of a self-organization critical state marked with the pregnancy height of a monitor and the calculation of a control limit of a residual error control chart. The detailed steps are as follows:
(1) and (3) selecting continuous time interval on-line monitoring data of a monitor in a normal state, and forming a time sequence { A (t) } by taking the monitoring data k with the equal interval length of a time point, wherein t is a,2a,3a, … and ka }. Establishing an autoregressive moving average fitting model expression F of the sequence A (t), wherein the expression F is shown as a formula (6):
where Q is the test statistic, p and Q are the order of the autoregressive moving average fitting model, and if the autoregressive model is selected, Q is 0, and if the moving average model is selected, p is 0.
(2) Selecting a plurality of pregnancy hypertension samples time sequences U (t), substituting the sequences into an expression F to calculate a model value Fu(t) calculating a corresponding time residual Su(t)=∣A(t)-Fu(t) using the residual sequence as the characteristic quantity to find out the residual critical point s for marking the pregnancy-induced hypertension critical state of the monitored person by statistical methodc。
(3) And forming a time sequence based on the monitoring data to estimate all potential risk points.
(4) If the critical point is found, selecting the residual pregnant woman with pregnancy-induced hypertension greater than the critical point as a sample, and establishing a control limit by using a weighted variance method; if no critical point is found, a 3-time standard deviation control limit is established by adopting a common method.
(5) Drawing a control chart according to the calculated control limit, generating a time sequence R (t) from the real-time monitoring data of a monitor, substituting the time sequence into an expression F to generate a fitting value Fr(t) and calculating the residual S at the corresponding timerAnd (t) drawing on the control chart, and sending a pregnancy-induced hypertension risk early warning once the residual error point exceeds the control limit or generates a risk point, thereby realizing the monitoring of the pregnancy-induced hypertension risk.
The identification method of the risk points comprises the following steps:
step S1: initial parameter estimation and risk point detection
(1) And forming a maximum likelihood estimation of a time sequence based on the monitoring data, selecting the sequence of the original time in the first iteration, and then selecting the sequence adjusted based on the information of the risk point.
(2) For each moment, a standard test statistic for the corresponding moment is calculated, and when the standard test statistic is greater than a specified value, the point is a potential risk point.
(3) And (4) if no risk point is detected, directly jumping to the step (4), otherwise, subtracting the influence of the risk point from the measured value and the residual sequence according to the type of the risk point, and returning to the step (2) to continuously detect whether a new risk point exists or not.
(4) If no risk point is found in the first iteration, the sequence is considered to have no risk point, and the detection is stopped. And (3) if the risk point is detected in the iterative process, returning to the step (1) to carry out maximum likelihood estimation again and carrying out subsequent calculation until no risk point appears in the detection of the current step (2).
Step S2: performing joint estimation of risk points and monitoring data
(1) And if a plurality of risk points are detected in the step one, estimating the magnitude of the risk effect of each risk point.
(2) And (3) calculating the statistic of each risk point, and when the minimum value of the absolute value of each risk point is not larger than the specified value in the step one, subtracting the point corresponding to the minimum value from the detected risk point data, and then returning to the step (1) for recalculating, otherwise, returning to the step (3).
(3) And (3) subtracting the influence of the risk point from the sequence based on the estimated value of the risk effect size in the last step (1) to obtain the monitoring data of the adjusted time sequence.
(4) And (3) calculating the maximum likelihood estimation value of the monitoring data based on the time sequence adjusted in the step (3), if the difference between the residual standard deviation of the maximum likelihood estimation value at the current time and the residual standard deviation of the maximum likelihood estimation value at the last time is not more than a specified error, jumping to the step (1) of the step (three), and if not, switching to the step (1) of the step to calculate more times.
Step S3: and calculating residual errors of the monitoring data of the time series based on the maximum likelihood estimation values of the monitoring data in the step two (4).
The monitoring method of the intelligent pregnancy-induced hypertension monitoring device provided by the invention comprises the following steps: basic information of a monitor is input through an information input module, and then a vital sign standard value estimation module estimates a vital sign standard value of the monitor according to the basic information of the monitor and a medical specified vital sign normal value range; during actual monitoring, the data acquisition module acquires the vital sign parameters of a monitor in real time, and the pregnancy-induced hypertension syndrome risk prediction module monitors the pregnancy-induced hypertension syndrome risk of the monitor according to the basic information of the monitor, the vital sign parameters of the monitor and the vital sign standard value of the monitor; when the monitoring result shows that the risk of pregnancy-induced hypertension is present, the alarm module sends out early warning signals in time, and the result display module displays the risk of pregnancy-induced hypertension of the monitored person and the vital sign parameters of the monitored person in real time. The intelligent pregnancy-hypertension monitoring device provided by the invention can realize accurate and efficient monitoring and early warning of pregnancy-hypertension risks.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.