CN117238507B - Intensive care monitoring system based on Internet of things - Google Patents
Intensive care monitoring system based on Internet of things Download PDFInfo
- Publication number
- CN117238507B CN117238507B CN202311489427.8A CN202311489427A CN117238507B CN 117238507 B CN117238507 B CN 117238507B CN 202311489427 A CN202311489427 A CN 202311489427A CN 117238507 B CN117238507 B CN 117238507B
- Authority
- CN
- China
- Prior art keywords
- dimension
- value
- description element
- moment
- current moment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 106
- 239000013598 vector Substances 0.000 claims abstract description 76
- 230000008859 change Effects 0.000 claims abstract description 49
- 238000011156 evaluation Methods 0.000 claims abstract description 47
- 230000002159 abnormal effect Effects 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims description 33
- 230000006870 function Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- UXFQFBNBSPQBJW-UHFFFAOYSA-N 2-amino-2-methylpropane-1,3-diol Chemical compound OCC(N)(C)CO UXFQFBNBSPQBJW-UHFFFAOYSA-N 0.000 claims description 4
- 101150035093 AMPD gene Proteins 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000000474 nursing effect Effects 0.000 abstract description 17
- 238000012545 processing Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 5
- 239000008280 blood Substances 0.000 description 3
- 230000004872 arterial blood pressure Effects 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000036760 body temperature Effects 0.000 description 2
- 230000035487 diastolic blood pressure Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 230000036387 respiratory rate Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000035488 systolic blood pressure Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to the technical field of physiological monitoring data processing, in particular to a severe nursing monitoring system based on the Internet of things. According to the invention, the physical sign stability evaluation of each dimension at the current moment is obtained according to the change and distribution characteristics of the physiological monitoring data of each dimension in the system state vector at all other moments in the history range of the current moment; obtaining a description element change weight corresponding to each dimension at the current time according to the difference of the description elements in the local data fluctuation interval of each dimension at the current time and the physical sign stability evaluation of each observation value sequence; further, adjusting the description element to obtain a weighted sigma point set at the current moment; and obtaining a predicted state vector at the next moment, and monitoring the abnormal state of the patient. According to the invention, by determining the accurate sigma points, the predicted value is more accurate, the abnormal state of the patient is accurately predicted, and the intelligence of intensive care monitoring is improved.
Description
Technical Field
The invention relates to the technical field of physiological monitoring data processing, in particular to a severe nursing monitoring system based on the Internet of things.
Background
The critical care can maintain the vital functions of the critical patients, prevent and alleviate complications, and provide support for the treatment of the patients. The monitoring means is the real-time monitoring patient's of wearing formula intelligent terminal real-time supervision patient's of thing networking sign index data, and the life state of estimation patient provides patient's real-time condition for medical personnel carries out patient's nursing during operation, helps improving nursing efficiency, reduces nursing cost. Because the illness state of the patient in severe nursing is serious, the change condition of the physiological index is interfered by the illness, the change of the monitoring value is more random, and the prediction accuracy is poor.
In the prior art, the UKF algorithm is adopted to predict the physiological index data, but in the actual nursing monitoring process, the physiological index data environments of patients are different, the permissible conditions of data fluctuation are different, so that the influence condition of the current moment data represented by the sigma point of the UKF is not integrated with the fluctuation condition of the data, the physiological monitoring data is predicted poorly, and the state of the patients cannot be evaluated accurately.
Disclosure of Invention
In order to solve the technical problem that a prediction result is poor due to the fact that proper sigma points cannot be determined, the invention aims to provide a severe nursing monitoring system based on the Internet of things, and the adopted technical scheme is as follows:
the invention provides a critical care monitoring system based on the Internet of things, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the following steps when executing the computer program:
acquiring physiological monitoring data of each patient at the current moment, and acquiring a sigma point set to be detected at the current moment by using a UKF algorithm according to a system state vector constructed by the physiological monitoring data at each moment;
acquiring description elements in each dimension of sigma points to be detected at all times in a history range of the current time, and constructing a description element curve of each dimension; obtaining a plurality of observation value sequences according to the change characteristics of the description element at the current moment in the description element curve of each dimension; obtaining a local data fluctuation interval of each dimension at the current moment according to the time characteristics in each observation value sequence;
taking the local data fluctuation interval as a time period to be analyzed, and obtaining a first feature stability evaluation corresponding to each moment according to the change and distribution characteristics of the physiological monitoring data corresponding to each dimension in the system state vector at all moments of the time period to be analyzed;
obtaining a description element change weight corresponding to each dimension at the current time according to the description element range of each dimension at the current time in the sigma point to be detected and the first feature stability evaluation; obtaining a weighted sigma point of the current moment according to the description element change weight values and the description elements corresponding to all dimensions at the current moment; obtaining a predicted state vector of the next moment according to the weighted sigma point of the current moment;
and monitoring the abnormal state of the patient according to the predicted state vector.
Further, the method for acquiring the sigma point set to be detected comprises the following steps:
and taking the physiological monitoring data at all times in the history range of the current time and the system state vector at the current time as the input of a UKF algorithm to obtain a sigma point set to be detected at the current time.
Further, the method for acquiring the observation value sequence comprises the following steps:
obtaining maximum values of the description elements in each dimension at all time points through an AMPD algorithm;
and segmenting the whole description element sequence by taking the maximum value point of each description element as a segmentation point to obtain a plurality of sequence segments serving as a corresponding observation value sequence.
Further, the method for acquiring the local data fluctuation interval comprises the following steps:
and screening out the position time closest to the current time in the observation value sequence, wherein the time interval from the position time to the current time is used as the local data fluctuation interval of each dimension at the current time.
Further, the method for acquiring the first sign stability evaluation includes:
obtaining the sign according to an acquisition formula of sign stability evaluationThe acquisition formula of the stability evaluation and the physical sign stability evaluation comprises the following steps:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->At time, the%>First sign stability evaluation of each dimension; />Indicate->The>Individual physiological monitoring data; />Indicating the>The average value of all physiological monitoring data in each dimension; />Indicating the>Numerical standard deviation of all physiological monitoring data in each dimension; />Representing the normalization function.
Further, the method for obtaining the description element variation weight value comprises the following steps:
normalizing the description element range corresponding to the current moment based on the description element ranges corresponding to all moments in the local data fluctuation interval to obtain a first normalized value; taking a time period formed by the current time and all corresponding historical time as a time period to be analyzed, obtaining second sign stability evaluation of each time, normalizing the first sign stability evaluation according to the value range of all the second sign stability evaluation, and obtaining a second normalized value; and carrying out negative correlation mapping on the second normalized value, and adding the second normalized value with the first normalized value to obtain the variation weight of the description element.
Further, the method for acquiring the weighted sigma points comprises the following steps:
calculating the product of the description element variation weight value and the description element of each dimension to obtain a weighted description element of each dimension; and correspondingly replacing the description elements of the sigma points to be detected with the weighted description elements of all the dimensions to obtain the weighted sigma points at the current moment.
Further, the monitoring of the abnormal state of the patient according to the predicted state vector includes:
acquiring a priori maximum urgent value and a priori minimum urgent value corresponding to each physiological monitoring data; according to the sequence of each data dimension in the system state vector, the prior maximum urgent value corresponding to all the physiological monitoring data is formed into a maximum urgent monitoring vector, and the prior minimum urgent value corresponding to all the physiological monitoring data is formed into a minimum urgent monitoring vector;
calculating cosine similarity of the predicted state vector and the maximum emergency monitoring vector at the next moment to obtain a maximum vector included angle cosine value; calculating cosine similarity of the predicted state vector and the minimum emergency monitoring vector at the next moment to obtain a minimum vector included angle cosine value;
if at least one of the maximum vector included angle cosine value and the minimum vector included angle cosine value is larger than a preset threshold value, the patient is in an abnormal state.
Further, the method for performing negative correlation mapping on the second normalized value is to reverse the second normalized value.
Further, the preset threshold is 0.75.
The invention has the following beneficial effects:
the invention obtains the physiological monitoring data of each patient at the current moment, a system state vector constructed according to the physiological monitoring data of each moment, obtains a sigma point set to be detected at the current moment by using UFK algorithm, and analyzes the possible state change range under each state estimation; in order to judge the physical sign stability condition corresponding to the physiological monitoring data, according to the change and distribution characteristics of the physiological monitoring data of each dimension in the system state vector of all other moments in the history range of the current moment, obtaining the physical sign stability evaluation of each dimension at the current moment; the sign stability vector at the current moment is obtained, and the overall stability of the sign at the current moment can be known globally; obtaining an observation value sequence according to the change characteristics of the description elements of each dimension at the current moment; obtaining a local data fluctuation interval of each dimension at the current moment, and better knowing the change trend of the system state; obtaining a description element change weight corresponding to each dimension at the current moment according to the description element in the local data fluctuation interval of each dimension at the current moment and the physical sign stability evaluation of each observation value sequence; the weighted sigma point at the current moment is obtained by combining the description elements, the UKF filter performance is optimized, and the predicted state vector at the next moment is obtained; abnormal conditions of the patient are monitored. According to the invention, by determining the accurate sigma points, the predicted value is more accurate, the abnormal state of the patient is accurately predicted, and the effectiveness of nursing monitoring of the patient is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of an implementation method of a critical care monitoring system based on the internet of things according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of a critical care monitoring system based on the internet of things according to the invention, which is provided by combining the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of a severe nursing monitoring system based on the Internet of things, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of an implementation method of an intensive care monitoring system based on internet of things according to an embodiment of the present invention is shown, where the method specifically includes:
step S1: and acquiring physiological monitoring data of each patient at the current moment, and acquiring a sigma point set to be detected at the current moment by using a UFK algorithm according to a system state vector constructed by the physiological monitoring data at each moment.
In the embodiment of the invention, as the patient in serious nursing has serious illness and consciousness is not awake, the patient cannot express the state of the patient in time through language, usually the patient limb is provided with a sensor, and the physical sign index data of the patient, such as heart rate P, body temperature T, respiratory rate M and respiratory flow, are monitored in real timeBlood pressure (systolic pressure S, diastolic pressure D, mean arterial pressure A, central venous pressure C), blood oxygen saturation +.>Equal index numberAccording to the method, the life state of the patient is estimated, and the nursing efficiency is improved. The physiological monitoring data of each patient at the current moment is obtained, and the sigma point to be detected at the current moment is obtained by utilizing the UKF algorithm according to the system state vector constructed by the physiological monitoring data at each moment. In one embodiment of the invention, the system state vector is formulated as:
；
wherein,a system state vector representing each patient; p represents heart rate; t represents body temperature; m represents respiratory rate; />Representing respiratory flow; s represents the systolic pressure; d represents diastolic pressure; a represents mean arterial pressure; c represents the central venous pressure; />Represents blood oxygen saturation; the corner mark T represents the vector transpose.
It should be noted that, different types of physiological monitoring data of different nursing environments may be specifically set according to specific situations. In the embodiment of the invention, each sensor acquires time information through the same clock of the hospital center server; the data acquisition intervals of the sensors are the same, and the embodiment takes the acquisition interval of 0.1s as an example; in other embodiments of the present invention, the acquisition interval may be specifically set according to specific situations, which are not limited and described herein.
sigma points are a set of sampling points used to approximate the state variables of the nonlinear function propagation and measurement update process. These sampling points represent the range of possible state changes at each state estimate. By carrying out the propagation of the nonlinear function and the mapping of the observation model on the sigma points, the propagation result and the prediction observation value of the nonlinear function under each state estimation can be obtained, so that the state prediction and the state update are realized.
Preferably, in one embodiment of the present invention, the method for acquiring the sigma point set to be measured includes:
unscented kalman filter UKF is a nonlinear filtering algorithm that approximates the propagation and measurement update process of a nonlinear function by selecting a set of sigma points to be selected. The selection of the sigma points to be selected plays an important role in the performance and accuracy of the UKF, provides important estimation and update information in the UKF algorithm, and is beneficial to improving the accuracy and performance of filtering. And taking physiological monitoring data of all the moments in the historical range of the current moment and the system state vector of the current moment as the input of the UKF algorithm to obtain a sigma point set to be detected of the current moment.
It should be noted that, in one embodiment of the present invention, each sigma point under test and the system state vectorAre all 9x1 sized vectors.
It should be noted that, in the embodiment of the present invention, the unscented kalman filter UKF algorithm is a technical means well known to those skilled in the art, and will not be described herein.
Step S2: acquiring description elements in each dimension of sigma points to be detected at all times in a history range of the current time, and constructing a description element curve of each dimension; obtaining a plurality of observation value sequences according to the change characteristics of the description elements at the current moment in the description element curve of each dimension; and obtaining a local data fluctuation interval of each dimension at the current time according to the time characteristics in each observation value sequence.
The description element curve can intuitively show the change condition of the system state in each dimension, the change trend of the system state is better known by observing the change of the description element curve, predictions are made, the description element in each dimension in the sigma points to be detected at all times in the history range at the current moment is obtained, and the description element curve of each dimension is constructed; and obtaining a plurality of observation value sequences according to the change characteristics of the description element at the current moment in the description element curve of each dimension.
Preferably, in one embodiment of the present invention, the method for acquiring the observation sequence includes:
the observation value sequence can reflect the change characteristics of the description elements of each dimension, and the maximum value of the description elements of each dimension at all time points is obtained through an AMPD algorithm; and segmenting the whole description element sequence by taking the maximum value point of each description element as a segmentation point to obtain a plurality of sequence segments serving as a corresponding observation value sequence.
It should be noted that, the AMPD algorithm is a technical means well known to those skilled in the art, and will not be described herein.
In order to enable the selection and evaluation of sigma points to be more in line with the physiological monitoring value change generated by the short-time severe disease change of an actual patient in the nursing process, the influence condition represented by the sigma points can be more in line with the local monitoring value change near the current moment. And screening out the position moment closest to the current moment in the observation value sequence, and obtaining the local data fluctuation interval of each dimension at the current moment.
Preferably, in one embodiment of the present invention, the method for acquiring the local data fluctuation interval includes:
and screening out the position moment closest to the current moment in the observation value sequence, wherein the time interval from the position moment to the current moment is used as the local data fluctuation interval of each dimension at the current moment.
In one embodiment of the present invention, the following is the first embodimentTime is for example closest to the +.sup.th in the sequence of observations>One position of the moment is +.>，/>Represents a distance +.>The first, closest-to-moment, larger data interference change, therefore, is determined by +.>As->Local data fluctuation interval of individual dimension +.>。
Step S3: and taking the local data fluctuation interval as a time period to be analyzed, and obtaining a first feature stability evaluation corresponding to each moment according to the change and distribution characteristics of the physiological monitoring data corresponding to each dimension in the system state vector at all moments of the time period to be analyzed.
The illness state of the patient in the severe patients is serious, the change of the physiological monitoring data is random, and the physiological monitoring data is easy to change obviously in short time; analyzing the frequency of the change of the physiological monitoring data to judge the sign stability condition corresponding to the physiological monitoring data; and taking the local data fluctuation interval as a time period to be analyzed, and obtaining a first feature stability evaluation corresponding to each moment according to the change and distribution characteristics of the physiological monitoring data corresponding to each dimension in the system state vector at all moments of the time period to be analyzed.
Preferably, in an embodiment of the present invention, the method for acquiring the first feature stability evaluation includes:
obtaining a first physical stability evaluation according to an acquisition formula of the first physical stability evaluation, the acquisition formula of the first physical stability evaluation comprising:
；
wherein,indicate->At time, the%>First sign stability evaluation of each dimension; />Indicate->The>Individual physiological monitoring data; />Represents the period to be analyzed, th->The average value of all physiological monitoring data in each dimension; />Indicating the>Numerical standard deviation of all physiological monitoring data in each dimension; />Representing the normalization function.
In the acquisition formula for the first sign stability evaluation,indicating each +.>In the physiological monitoring data acquired at any time, the deviation degree of each physiological monitoring data from the mean value is measured by the standard deviation when each physiological monitoring data generates a larger difference relative to the mean value, and the larger the deviation degree is, the larger the abnormal influence possibly occurring in vital signs of patients is representedThe method comprises the steps of carrying out a first treatment on the surface of the By->The difference accumulation mode obtains the mean value of the deviation degree, which indicates that the current moment is passed +.>The number of times of the change on the accumulated time sequence is judged, and when the number of times of occurrence of the obvious deviation of the physiological monitoring data is larger, the more easily the physiological monitoring data is changed, the more unstable the vital sign condition represented by the physiological monitoring data of each patient is reflected; due to different increasing and decreasing trends of the physiological monitoring data change, the physiological monitoring data is obtained byThe cube processes, reduces the tiny difference, increases the obvious difference, simultaneously keeps the trend of the fluctuation of the physiological monitoring data, reflects the data direction of the change generated in each data dimension, and the physiological monitoring data change is usually accompanied with the change of the physiological environment, can analyze the change characteristics of the physiological environment of a patient through the trend direction of the data fluctuation, and increases the information corresponding to the change trend of the physical sign while judging the stable condition of the physical sign.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation and normalization method may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, and will not be described herein.
In order to facilitate analysis of the processed data, the overall stability of the sign at the current time can be globally known by summarizing the sign stability evaluations of each dimension. And the physical sign stability evaluation of all the dimensions at the current moment can be formed into a physical sign stability vector at the current moment according to the dimension sequence corresponding to each physiological monitoring data in the system state vector. In one embodiment of the invention, the physical stability vector is formulated as:
；
wherein,indicate->A physical sign stability vector at a time; />Indicate->At time, the%>Physical sign stability evaluation of each dimension; />Indicate->At time, the%>Physical sign stability evaluation of each dimension; the corner mark T represents the vector transpose.
Step S4: obtaining a description element change weight corresponding to each dimension at the current moment according to the description element range of each dimension at the current moment in the sigma point to be detected and the first feature stability evaluation; obtaining a weighted sigma point of the current moment according to the description element change weight and the description element corresponding to all dimensions at the current moment; and obtaining a predicted state vector of the next moment according to the weighted sigma point of the current moment.
In the UKF algorithm, the weights of the description elements are determined according to the closeness degree of the description elements and the real state, and the sigma point selection is influenced by the adjustment of the weights, so that the optimization of the filter performance is realized. And obtaining the change weight of the description element corresponding to each dimension at the current moment according to the description element range of each dimension at the current moment in the sigma point to be detected and the first feature stability evaluation.
Preferably, in one embodiment of the present invention, the method for acquiring the element change weight includes:
normalizing the description element range corresponding to the current moment based on the description element ranges corresponding to all moments in the local data fluctuation interval to obtain a first normalized value; taking a time period formed by the current time and all corresponding historical time as a time period to be analyzed, obtaining second sign stability evaluation of each time, normalizing the first sign stability evaluation according to the value range of all the second sign stability evaluation, and obtaining a second normalized value; and carrying out negative correlation mapping on the second normalized value, and adding the second normalized value with the first normalized value to obtain the variation weight of the description element. In one embodiment of the invention, the formula describing the element variation weight is expressed as:
；
wherein,indicate->The change weight of the description element corresponding to the kth dimension at the moment; />A description element range representing a kth dimension; />Is->Time-of-day correspondence +.>；/>Local data fluctuation interval representing the kth dimension +.>Maximum->A value; />Local data fluctuation interval representing the kth dimension +.>Minimum->A value; />Indicate->First sign stability evaluation of physiological monitoring data of a kth dimension at a moment; />Representing a dimension sequence number corresponding to any physiological monitoring data; />Indicate->Evaluation of the stability of the second sign of the individual dimension +.>；/>Second characterization stability evaluation of all dimensions at the time instant corresponding to each observation sequence B>Is the maximum value of (2); />Representing the second sign stability of all dimensions at the time instants corresponding to the respective observation sequences BQualitative assessment->Is a minimum of (2).
In the formula describing the element variation weight,for->The greater the normalization, i.e., the maximum time difference, the more severe the variation, the less the stability, the greater the fluctuation of the data, the more likely an anomaly will occur,the representation will->And (3) carrying out normalization, wherein the smaller the sign stability evaluation is, the more abnormal conditions exist, and the larger the variation weight of the description element is.
It should be noted that, in one embodiment of the present invention, the method of performing the negative correlation mapping on the second normalized value is to reverse the second normalized value. In other embodiments of the present invention, the positive-negative correlation may be constructed by other basic mathematical operations, and the specific means are technical means well known to those skilled in the art, and will not be described herein.
Through weighting adjustment of sigma points, proper sigma points are obtained, uncertainty and noise in a nonlinear system can be effectively approximated, adaptability of an algorithm is improved, and a more accurate estimation effect is obtained. And obtaining the weighted sigma point of the current moment according to the description element change weight and the description element corresponding to all the dimensions at the current moment.
Preferably, in one embodiment of the present invention, the method for acquiring weighted sigma points includes:
calculating the product of the description element variation weight value and the description element of each dimension to obtain a weighted description element of each dimension; and correspondingly replacing the description elements of the sigma points to be detected with the weighted description elements of all the dimensions to obtain the weighted sigma points at the current moment. In one embodiment of the invention, the formula for the weighted description element is expressed as:
；
wherein,a weighted description element representing a kth dimension; />A description element representing a kth dimension; />Represent the firstAnd the description element corresponding to the kth dimension changes the weight value at the moment.
In the formula of the weighted description element, the larger the description element variation weight of each dimension is, the larger the description element is, and the larger the weighted description element of each dimension is.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation may be constructed by other basic mathematical operations, and the specific means are technical means well known to those skilled in the art, which are not described herein.
In order to accurately estimate the system state at the next moment, a predicted state vector at the next moment is obtained according to the weighted sigma point at the current moment.
Step S5: the abnormal state of the patient is monitored based on the predicted state vector.
The predicted state vector can monitor the state change of the patient in real time, so that the health condition of the patient can be accurately judged in order to discover and treat the abnormal condition of the patient in time, the abnormal state of the patient is monitored according to the predicted state vector, and corresponding nursing measures are timely taken if the abnormal condition occurs.
Preferably, in one embodiment of the present invention, monitoring the abnormal state of the patient according to the predicted state vector comprises:
acquiring a priori maximum urgent value and a priori minimum urgent value corresponding to each physiological monitoring data; according to the sequence of each data dimension in the system state vector, constructing a maximum emergency monitoring vector by using the prior maximum emergency values corresponding to all the physiological monitoring data, and constructing a minimum emergency monitoring vector by using the prior minimum emergency values corresponding to all the physiological monitoring data; calculating cosine similarity of the predicted state vector and the maximum emergency monitoring vector at the next moment to obtain a cosine value of the maximum vector included angle; calculating cosine similarity of the predicted state vector and the minimum emergency monitoring vector at the next moment to obtain a minimum vector included angle cosine value; if at least one of the maximum vector included angle cosine value and the minimum vector included angle cosine value is larger than a preset threshold, the patient is in an abnormal state, the patient needs to be nursed, and the intelligent of patient nursing is improved. In one embodiment of the invention, taking a heartbeat as an example, physiological monitoring data exceeding 150 or falling below 60 would be considered abnormal, corresponding to a prior maximum urgency value of 150 and a prior minimum urgency value of 60.
It should be noted that, in one embodiment of the present invention, the magnitude of the preset threshold is 0.75. In other embodiments of the present invention, the magnitude of the preset threshold may also be specifically set according to specific situations, which is not limited and described herein. Cosine similarity is a technical means well known to those skilled in the art, and is not described herein.
In summary, the invention acquires the physiological monitoring data of each patient at the current moment, and acquires the sigma point set to be detected at the current moment by using the UKF algorithm according to the system state vector constructed by the physiological monitoring data at each moment; according to the change and distribution characteristics of physiological monitoring data of each dimension in the system state vector of all other moments in the history range of the current moment, obtaining the sign stability evaluation of each dimension at the current moment; further obtaining a physical sign stability vector at the current moment; obtaining an observation value sequence according to the change characteristics of the description elements of each dimension at the current moment; obtaining a local data fluctuation interval of each dimension at the current moment; obtaining a description element change weight corresponding to each dimension at the current time according to the difference of the description elements in the local data fluctuation interval of each dimension at the current time and the physical sign stability evaluation of each observation value sequence; the description elements are combined to obtain weighted sigma points at the current moment, and a predicted state vector at the next moment is obtained; abnormal conditions of the patient are monitored. According to the invention, by determining the accurate sigma points, the predicted value is more accurate, the abnormal state of the patient is accurately predicted, and the effectiveness of nursing monitoring of the patient is improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (6)
1. An intensive care monitoring system based on the internet of things, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the following steps when executing the computer program:
acquiring physiological monitoring data of each patient at the current moment, and acquiring a sigma point set to be detected at the current moment by using a UKF algorithm according to a system state vector constructed by the physiological monitoring data at each moment;
acquiring description elements in each dimension of sigma points to be detected at all times in a history range of the current time, and constructing a description element curve of each dimension; obtaining a plurality of observation value sequences according to the change characteristics of the description element at the current moment in the description element curve of each dimension; obtaining a local data fluctuation interval of each dimension at the current moment according to the time characteristics in each observation value sequence;
taking the local data fluctuation interval as a time period to be analyzed, and obtaining a first feature stability evaluation corresponding to each moment according to the change and distribution characteristics of the physiological monitoring data corresponding to each dimension in the system state vector at all moments of the time period to be analyzed;
obtaining a description element change weight corresponding to each dimension at the current time according to the description element range of each dimension at the current time in the sigma point to be detected and the first feature stability evaluation; obtaining a weighted sigma point of the current moment according to the description element change weight values and the description elements corresponding to all dimensions at the current moment; obtaining a predicted state vector of the next moment according to the weighted sigma point of the current moment;
monitoring the abnormal state of the patient according to the predicted state vector;
the method for acquiring the observation value sequence comprises the following steps:
obtaining maximum values of the description elements in each dimension at all time points through an AMPD algorithm;
segmenting the whole description element sequence by taking the maximum value point of each description element as a segmentation point to obtain a plurality of sequence segments serving as corresponding observation value sequences;
the method for acquiring the local data fluctuation interval comprises the following steps:
screening out the position moment closest to the current moment in the observation value sequence, wherein the time interval from the position moment to the current moment is used as the local data fluctuation interval of each dimension at the current moment;
the method for acquiring the first sign stability evaluation comprises the following steps:
obtaining a first physical stability evaluation according to an acquisition formula of the first physical stability evaluation, the acquisition formula of the first physical stability evaluation comprising:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->At time, the%>First sign stability evaluation of each dimension; />Indicate->The>Individual physiological monitoring data; />Indicating the>The average value of all physiological monitoring data in each dimension; />Indicating the>Numerical standard deviation of all physiological monitoring data in each dimension; />Representing a normalization function;
the method for acquiring the variation weight of the description element comprises the following steps:
normalizing the description element range corresponding to the current moment based on the description element ranges corresponding to all moments in the local data fluctuation interval to obtain a first normalized value; taking a time period formed by the current time and all corresponding historical time as a time period to be analyzed, obtaining second sign stability evaluation of each time, normalizing the first sign stability evaluation according to the value range of all the second sign stability evaluation, and obtaining a second normalized value; and carrying out negative correlation mapping on the second normalized value, and adding the second normalized value with the first normalized value to obtain the variation weight of the description element.
2. The intensive care monitoring system based on the internet of things of claim 1, wherein the method for acquiring the sigma point set to be tested comprises:
and taking the physiological monitoring data at all times in the history range of the current time and the system state vector at the current time as the input of a UKF algorithm to obtain a sigma point set to be detected at the current time.
3. The intensive care monitoring system based on the internet of things according to claim 1, wherein the method for acquiring the weighted sigma points comprises:
calculating the product of the description element variation weight value and the description element of each dimension to obtain a weighted description element of each dimension; and correspondingly replacing the description elements of the sigma points to be detected with the weighted description elements of all the dimensions to obtain the weighted sigma points at the current moment.
4. The internet of things-based critical care monitoring system of claim 1, wherein monitoring the abnormal state of the patient according to the predicted state vector comprises:
acquiring a priori maximum urgent value and a priori minimum urgent value corresponding to each physiological monitoring data; according to the sequence of each data dimension in the system state vector, the prior maximum urgent value corresponding to all the physiological monitoring data is formed into a maximum urgent monitoring vector, and the prior minimum urgent value corresponding to all the physiological monitoring data is formed into a minimum urgent monitoring vector;
calculating cosine similarity of the predicted state vector and the maximum emergency monitoring vector at the next moment to obtain a maximum vector included angle cosine value; calculating cosine similarity of the predicted state vector and the minimum emergency monitoring vector at the next moment to obtain a minimum vector included angle cosine value;
if at least one of the maximum vector included angle cosine value and the minimum vector included angle cosine value is larger than a preset threshold value, the patient is in an abnormal state.
5. The internet of things-based intensive care monitoring system of claim 1, wherein the method of inversely mapping the second normalized value is to reverse the second normalized value.
6. The internet of things-based critical care monitoring system of claim 4, wherein the preset threshold is 0.75.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311489427.8A CN117238507B (en) | 2023-11-10 | 2023-11-10 | Intensive care monitoring system based on Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311489427.8A CN117238507B (en) | 2023-11-10 | 2023-11-10 | Intensive care monitoring system based on Internet of things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117238507A CN117238507A (en) | 2023-12-15 |
| CN117238507B true CN117238507B (en) | 2024-01-30 |
Family
ID=89093139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311489427.8A Active CN117238507B (en) | 2023-11-10 | 2023-11-10 | Intensive care monitoring system based on Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117238507B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117442288B (en) * | 2023-12-20 | 2024-03-05 | 中国人民解放军总医院第八医学中心 | Intelligent regulation and control system for pressure value of electric tourniquet |
| CN118033409B (en) * | 2024-04-15 | 2024-06-14 | 三峡金沙江川云水电开发有限公司 | GCB arc extinguishing chamber switch resistance testing method |
| CN118098474A (en) * | 2024-04-19 | 2024-05-28 | 自贡市第一人民医院 | Data optimization processing method for medical nursing clinical monitoring equipment |
| CN118277744B (en) * | 2024-05-31 | 2024-08-13 | 水发投资集团有限公司 | Intelligent monitoring method for environment state of red sage root seedling culture based on data analysis |
| CN118299075A (en) * | 2024-05-31 | 2024-07-05 | 吉林大学 | Real-time physical sign monitoring system and method for comatose patient |
| CN118350809B (en) * | 2024-06-18 | 2024-09-10 | 大连乾曜科技有限公司 | Method and system for detecting concrete curing in winter |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114398817A (en) * | 2021-08-31 | 2022-04-26 | 中国石油大学(北京) | Method and device for dynamically estimating production operation condition of natural gas shaft |
| CN117009831A (en) * | 2023-10-07 | 2023-11-07 | 山东世纪阳光科技有限公司 | Fine chemical accident risk prediction assessment method |
-
2023
- 2023-11-10 CN CN202311489427.8A patent/CN117238507B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114398817A (en) * | 2021-08-31 | 2022-04-26 | 中国石油大学(北京) | Method and device for dynamically estimating production operation condition of natural gas shaft |
| CN117009831A (en) * | 2023-10-07 | 2023-11-07 | 山东世纪阳光科技有限公司 | Fine chemical accident risk prediction assessment method |
Non-Patent Citations (3)
| Title |
|---|
| "IsomapVSG-LIME：一种新的模型无关解释方法";向许等;《智能系统学报》;第18卷(第4期);全文 * |
| "基于无迹卡尔曼滤波的永磁无刷电机 转矩脉动抑制研究";程梁 等;《大电机技术》;全文 * |
| 基于极大后验估计和指数加权的自适应UKF滤波算法;赵琳;王小旭;孙明;丁继成;闫超;;自动化学报(07);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117238507A (en) | 2023-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117238507B (en) | Intensive care monitoring system based on Internet of things | |
| US10722179B2 (en) | Residual-based monitoring of human health | |
| CN110944577B (en) | Method and system for detecting blood oxygen saturation | |
| US7818131B2 (en) | Non-parametric modeling apparatus and method for classification, especially of activity state | |
| EP1658578B1 (en) | A system and method for detecting signal artifacts | |
| KR20150113700A (en) | System and method for diagnosis | |
| JP2020536623A (en) | Continuous monitoring of user health using mobile devices | |
| CN109564586B (en) | System monitor and system monitoring method | |
| JP6943287B2 (en) | Biometric information processing equipment, biometric information processing systems, biometric information processing methods, and programs | |
| CN113421654A (en) | Dynamic early warning and deep learning system for post-traumatic hemorrhagic shock | |
| CN114758786A (en) | Dynamic early warning system for post-traumatic hemorrhagic shock based on noninvasive parameters | |
| CN117379021B (en) | Old person health index monitoring system based on intelligent wearing equipment | |
| CA3139034A1 (en) | System and method for filtering time-varying data for physiological signal prediction | |
| Imhoff et al. | Pattern detection in intensive care monitoring time series with autoregressive models: Influence of the model order | |
| AU2019204388A1 (en) | Residual-based monitoring of human health | |
| Wong et al. | Probabilistic detection of vital sign abnormality with Gaussian process regression | |
| KR20220129283A (en) | System and method for notifying abnomal measurement state of bio-signal based on artificial intelligence algorithm | |
| CN116965785B (en) | Vital sign monitoring and analyzing method, system, terminal and storage medium | |
| CN112971791B (en) | Individuation physiological state monitoring and analyzing method and equipment | |
| CN112971792B (en) | Individual state monitoring and analyzing method and equipment based on continuous physiological data | |
| Fried et al. | Statistical Alarm Systems in Intensive Care Medicine | |
| CN116250823A (en) | Sleep breathing abnormality early warning method based on MSET and real-time dynamic baseline | |
| KR20220152010A (en) | Artificial intelligence system for acquiring sleep behavior information using wireless communication and estimating sleep stage using the same | |
| WO2023062118A1 (en) | Method and device for predicting frailty of a subject | |
| Fried et al. | Some statistical methods in intensive care online monitoring—a review |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240819 Address after: 11502, Building 10, Area A, Maple Leaf New City, Fenghui South Road, High tech Zone, Xi'an City, Shaanxi Province 710000 Patentee after: Xi'an Changshida Information Technology Co.,Ltd. Country or region after: China Address before: Room 201, 2nd Floor, No. 55-3 Yunong West Street, Ganjingzi District, Dalian City, Liaoning Province, 110000 Patentee before: Jinheng Technology (Dalian) Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right |