Movatterモバイル変換


[0]ホーム

URL:


CN118315059A - Human health data analysis model and method based on Bayesian algorithm - Google Patents

Human health data analysis model and method based on Bayesian algorithm
Download PDF

Info

Publication number
CN118315059A
CN118315059ACN202410416794.3ACN202410416794ACN118315059ACN 118315059 ACN118315059 ACN 118315059ACN 202410416794 ACN202410416794 ACN 202410416794ACN 118315059 ACN118315059 ACN 118315059A
Authority
CN
China
Prior art keywords
target
target personnel
state
target person
personnel
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.)
Granted
Application number
CN202410416794.3A
Other languages
Chinese (zh)
Other versions
CN118315059B (en
Inventor
王萧雷
王忱
完阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Yijingyun Technology Co ltd
Original Assignee
Guangzhou Yijingyun Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Yijingyun Technology Co ltdfiledCriticalGuangzhou Yijingyun Technology Co ltd
Priority to CN202410416794.3ApriorityCriticalpatent/CN118315059B/en
Publication of CN118315059ApublicationCriticalpatent/CN118315059A/en
Application grantedgrantedCritical
Publication of CN118315059BpublicationCriticalpatent/CN118315059B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

The invention relates to the technical field of human health data analysis, in particular to a human health data analysis model and method based on a Bayesian algorithm, comprising a data acquisition module, a priori probability analysis module, a physiological analysis module, a habit analysis module, a conditional probability analysis module, a disease risk prediction analysis module and a display terminal; according to the invention, the probability analysis is carried out by acquiring the basic data parameters, the physiological data parameters and the habit data parameters of the target personnel and utilizing the Bayesian algorithm, so that the comprehensive and accurate assessment of the physical state of the target personnel and the prediction of the disease risk are realized, the potential health risk is found in advance, and more accurate and effective health management and prevention measures are provided for the nursing home.

Description

Human health data analysis model and method based on Bayesian algorithm
Technical Field
The invention relates to the technical field of human health data analysis, in particular to a human health data analysis model and method based on a Bayesian algorithm.
Background
Along with rapid development of technology and increasing perfection of medical technology, human health data has become one of the focuses of current society, especially in places where elderly people such as nursing homes concentrate, health monitoring and data analysis of the elderly people are important, however, the current nursing homes still have a plurality of defects in health monitoring;
At present, although the aged is subjected to measurement and monitoring of physical data every day, the data usually stay at the basic physiological index level, such as blood pressure, blood sugar and the like, medical staff can usually only carry out simple judgment according to the single data, but cannot comprehensively and deeply know the physical state of the aged, the single monitoring mode often leads to the situation that the best intervention time is missed when the aged really has problems, and more importantly, the current health monitoring method lacks comprehensive analysis and probability prediction capability of various data of the aged, so that the probability trend of the physical diseases of the aged cannot be predicted in advance, and the defect of the prediction capability cannot enable the aged to provide a personalized health management scheme for the aged, and cannot effectively prevent and treat the aged before the diseases occur.
Disclosure of Invention
In order to overcome the defects in the background technology, the embodiment of the invention provides a human health data analysis model and a human health data analysis method based on a Bayesian algorithm, which can effectively solve the problems related to the background technology.
The aim of the invention can be achieved by the following technical scheme:
a human health data analysis method based on a Bayesian algorithm comprises the following steps:
Step one: acquiring human body data information corresponding to a target person to obtain human body data information corresponding to the target person, wherein the human body data information comprises basic data parameters, physiological data parameters and habit data parameters;
Step two: the method is used for monitoring basic data parameters corresponding to the target personnel, so that prior probability analysis is carried out on the physical state corresponding to the target personnel, and prior probability corresponding to the target personnel is obtained;
step three: the physiological data parameter monitoring device is used for monitoring physiological data parameters corresponding to the target personnel in the unit time period, so that the physiological state corresponding to the target personnel in the unit time period is evaluated and analyzed to obtain a physiological state evaluation coefficient corresponding to the target personnel;
Step four: the habit data parameter monitoring system is used for monitoring habit data parameters corresponding to the target personnel in the unit time period, so that habit states corresponding to the target personnel in the unit time period are evaluated and analyzed, and habit state evaluation coefficients corresponding to the target personnel are obtained;
Step five: the method comprises the steps of receiving a physiological state evaluation coefficient and a habit state evaluation coefficient corresponding to a target person, and performing conditional probability analysis on a physical state corresponding to the target person in a unit time period to obtain a conditional probability corresponding to the target person;
Step six: the method is used for carrying out posterior probability analysis on the physical state corresponding to the target personnel, so that disease risk prediction is carried out on the physical state corresponding to the target personnel, and high trend disease risk, medium trend disease risk and no trend disease risk are obtained.
Further, the prior probability analysis is performed on the physical state corresponding to the target person, and the specific analysis is as follows:
The height value, the age value, the weight value, the sex value and the family disease history value corresponding to the target personnel are obtained by extracting the height, the age, the weight, the sex and the family disease history in the basic data parameters corresponding to the target personnel and respectively assigning the height, the age, the weight, the sex and the family disease history values, and are respectively calibrated into RT1, RT2, RT3, RT4 and RT5 according to a set data model: Obtaining a basic body value JBZ corresponding to a target person, wherein e1, e2, e3, e4 and e5 all represent natural constants, yz represents a preset correction factor, and eta 1, eta 2, eta 3, eta 4 and eta 5 respectively represent a height value, an age value, a weight value, a sex value and a family disease history value weight coefficient;
And carrying out matching analysis on the basic body values corresponding to the target personnel and the body state tables stored in the cloud database to obtain body state grades corresponding to the target personnel, wherein each basic body value corresponding to the target personnel corresponds to one body state grade, and meanwhile, the basic body values are matched with the prior probabilities corresponding to the body state grades, so that the prior probabilities corresponding to the target personnel are obtained, and the prior probabilities are marked as P (Hi).
Further, the physiological state corresponding to the target person in the unit time period is evaluated and analyzed, and the specific analysis is as follows:
the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period are obtained by extracting the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period, and meanwhile, the numerical values of the highest heart rate, the lowest heart rate, the average heart rate, the highest body temperature, the lowest body temperature, the average body temperature, the highest blood pressure, the lowest blood pressure, the average blood pressure, the highest respiratory rate, the lowest respiratory rate, the average respiratory rate, the highest pulse beating times, the lowest pulse beating times and the average pulse beating times are extracted from the physiological data parameters, and are respectively calibrated into Qmax、Qmin、Qpjz、Wmax、Wmin、Wpjz、Emax、Emin、Epjz、Ymax、Ymin、Ypjz、Umax、Umin、Upjz, according to the formula: and obtaining a physiological state evaluation coefficient delta 1 corresponding to the target person, wherein lambda 1, lambda 2, lambda 3, lambda 4 and lambda 5 are expressed as set weight factors.
Further, the habit state corresponding to the target person in the unit time period is evaluated and analyzed, and the specific analysis is as follows:
The method comprises the steps of obtaining the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period by extracting the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period, calibrating the dining time length, the activity time length and the rest time length into CS, DS and XS respectively, extracting the numerical values of the three, and carrying out normalization processing according to the formula: And obtaining a habit state evaluation coefficient delta 2 corresponding to the target person, wherein CS*、DS* and XS* respectively represent a reference dining time length, a reference activity time length and a reference rest time length, and mu 1, mu 2 and mu 3 are represented as set weight factors.
Further, the conditional probability analysis is performed on the physical state corresponding to the target person in the unit time period, and the specific analysis is as follows:
The physical inertia value corresponding to the target person is obtained by modulating the values of the physiological state evaluation coefficient delta 1 and the habit state evaluation coefficient delta 2 corresponding to the target person to perform normalization calculation;
matching and analyzing the physical inertia state values corresponding to the target personnel with a physical inertia state table stored in a cloud database to obtain physical inertia state levels corresponding to the target personnel, wherein each physical inertia state level corresponds to one physical inertia state level, and meanwhile, the physical inertia state levels are matched with the physical states corresponding to the physical inertia state levels to obtain the physical states corresponding to the target personnel;
Assuming that Hi represents all possible physical states, the probability of occurrence of event E can be obtained by summing all possible physical states, according to the formula: p (E) =ΣiP(E|Hi)×P(Hi), resulting in the probability of occurrence of event E, P (e|hi) being the probability of occurrence of event E given the body state Hi, P (Hi) being the prior probability;
According to the Bayesian formula: And obtaining the conditional probability corresponding to the target personnel.
Further, the disease risk prediction is performed on the physical state corresponding to the target person, and the specific analysis is as follows:
The prior probability and the conditional probability corresponding to the target personnel are called, and the posterior probability corresponding to the target personnel is calculated according to a Bayesian formula;
Setting a posterior probability threshold corresponding to a target person, comparing and analyzing the posterior probability corresponding to the target person with the posterior probability threshold, judging that the target person has high trend disease risk when the posterior probability corresponding to the target person is larger than the posterior probability threshold, judging that the target person has medium trend disease risk when the posterior probability corresponding to the target person is equal to the posterior probability threshold, and judging that the target person has no trend disease risk when the posterior probability corresponding to the target person is smaller than the posterior probability threshold.
Further, a human health data analysis model based on a bayesian algorithm includes:
the data acquisition module is used for acquiring human body data information corresponding to the target personnel to obtain the human body data information corresponding to the target personnel, wherein the human body data comprises basic data parameters, physiological data parameters and habit data parameters;
The prior probability analysis module is used for monitoring basic data parameters corresponding to the target personnel to obtain basic body values corresponding to the target personnel, so that prior probability analysis is carried out on body states corresponding to the target personnel to obtain prior probabilities corresponding to the target personnel;
the physiological analysis module is used for monitoring physiological data parameters corresponding to the target personnel in the unit time period, so that the physiological states corresponding to the target personnel in the unit time period are evaluated and analyzed to obtain physiological state evaluation coefficients corresponding to the target personnel;
the habit analysis module is used for monitoring habit data parameters corresponding to the target personnel in the unit time period, so that habit states corresponding to the target personnel in the unit time period are evaluated and analyzed, and habit state evaluation coefficients corresponding to the target personnel are obtained;
the conditional probability analysis module is used for receiving the physiological state evaluation coefficient and the habit state evaluation coefficient corresponding to the target personnel, so that the physical state corresponding to the target personnel in the unit time period is subjected to conditional probability analysis, and the conditional probability corresponding to the target personnel is obtained;
the disease risk prediction analysis module is used for performing posterior probability analysis on the physical state corresponding to the target personnel, so that disease risk prediction is performed on the physical state corresponding to the target personnel, and high trend disease risk, medium trend disease risk and trend-free disease risk are obtained.
Compared with the prior art, the invention has the beneficial effects that:
According to the invention, through collecting various data parameters of target personnel, richer and comprehensive health information can be obtained, compared with the traditional method which only depends on a single physiological index, the method can reflect the physical condition of an individual more comprehensively, so that the accuracy of health evaluation is improved, secondly, the Bayesian algorithm is utilized to carry out prior probability analysis, conditional probability analysis and posterior probability analysis, individual differences and historical data of the individual can be fully considered, the health evaluation result is more personalized and accurate, thereby realizing the purpose of providing targeted health management advice for the individual and reducing the disease risk, and furthermore, the method can predict the disease risk of the individual in advance, including high trend disease risk, medium trend disease risk and no trend disease risk, so that the individual can know the health condition of the individual timely, and corresponding preventive measures are taken, so that the occurrence of diseases is effectively avoided or delayed.
Drawings
For the purpose of facilitating understanding to those skilled in the art, the invention is further described below with reference to the drawings;
FIG. 1 is a block diagram of the method of the present invention
Fig. 2 is a general block diagram of the system of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on the embodiments of the present invention, are intended to be within the scope of the present invention.
As shown in fig. 1, a human health data analysis method based on a bayesian algorithm includes the following steps:
Step one: and (3) data acquisition: the human body data acquisition device is used for acquiring human body data information corresponding to a target person to obtain the human body data information corresponding to the target person, wherein the human body data comprises basic data parameters, physiological data parameters and habit data parameters, and the specific acquisition mode is as follows:
Through setting up the target personnel data bullet window, before target personnel gets into the nursing home, automatic pop-up target personnel basic data fills the window, and it specifically can be: "please report your height, age, weight, sex and family history", obtain the basic data parameters corresponding to the target person;
The heart rate corresponding to the target person in the unit time period is monitored and obtained through a heart rate sensor, the body temperature corresponding to the target person in the unit time period is monitored and obtained through a temperature sensor, the blood pressure corresponding to the target person in the unit time period is monitored and obtained through a portable blood pressure monitor, the respiratory rate corresponding to the target person in the unit time period is monitored and obtained through an intelligent bracelet, the pulse beating times corresponding to the target person in the unit time period is monitored and obtained through a pulse meter, and the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times corresponding to the target person in the unit time period are respectively obtained, so that physiological data parameters corresponding to the target person in the unit time period are obtained;
Acquiring meal videos corresponding to target persons in a unit time period through an intelligent camera to obtain meal videos corresponding to the target persons in the unit time period, analyzing the meal videos corresponding to the target persons in the unit time period frame by frame according to monitoring time points to obtain meal images corresponding to the target persons in the unit time period, extracting action features corresponding to the target persons in the monitoring time points in the unit time period from the meal images, simultaneously matching the action features corresponding to the target persons in the monitoring time points in the unit time period with the action features corresponding to the set meal action, and when the action features corresponding to the target persons in the monitoring time points in the unit time period are successfully matched with the action features corresponding to the set meal action for the first time, recording the monitored time points successfully matched for the first time as meal starting time points, and simultaneously recording the monitored time points successfully matched for the last time as meal ending time points, so as to obtain meal duration corresponding to the target persons in the unit time period based on the meal starting time points and the meal ending time points corresponding to the target persons in the unit time period.
Wherein the action features are pouring water, holding dishes, taking things, clamping dishes into the mouth, chewing at the mouth, clamping dishes, using soup ladle, etc. The corresponding action characteristics of the dining action are as follows: the hand is used for clamping the dish in the mouth, chewing the dish at the mouth, and using a soup ladle;
The method comprises the steps that the motion state corresponding to a target person at each monitoring time point in a unit time period is monitored through a motion bracelet, if the motion state at a certain monitoring time point is an active state, the step count corresponding to the target person at the monitoring time point is displayed on the motion bracelet and is increased compared with the step count of the previous active state, if the motion state at the certain monitoring time point is a rest state, the step count corresponding to the target person at the monitoring time point on the motion bracelet is unchanged, therefore, the motion state corresponding to the target person at each monitoring time point in the unit time period can be identified based on the step count displayed on the motion bracelet, if the motion state at the certain monitoring time point is an active state, the monitoring time point is recorded as an active time point, and if the motion state at the certain monitoring time point is a rest state, the monitoring time point is recorded as a rest time point, and the activity duration and the rest duration corresponding to the target person at the unit time point are obtained through statistics based on each active time point and each rest time point corresponding to the target person at the unit time point in the unit time period;
Thereby obtaining habit data parameters corresponding to the target personnel in a unit time period;
step two: prior probability analysis: the method is used for monitoring basic data parameters corresponding to the target personnel, so that the prior probability analysis is carried out on the physical state corresponding to the target personnel, and the specific analysis is as follows:
The height value, the age value, the weight value, the sex value and the family disease history value corresponding to the target personnel are obtained by extracting the height, the age, the weight, the sex and the family disease history in the basic data parameters corresponding to the target personnel and respectively assigning the height, the age, the weight, the sex and the family disease history values, and are respectively calibrated into RT1, RT2, RT3, RT4 and RT5 according to a set data model: Obtaining a basic body value JBZ corresponding to a target person, wherein e1, e2, e3, e4 and e5 all represent natural constants, yz represents a preset correction factor, eta 1, eta 2, eta 3, eta 4 and eta 5 respectively represent a height value, an age value, a weight value, a sex value and a family disease history value weight coefficient, eta 2 > eta 5 > eta 3 > eta 1 > eta 4, and the weight coefficient is used for balancing the duty ratio weight of each item of data in formula calculation so as to promote the accuracy of a calculation result;
Performing matching analysis on the basic body values corresponding to the target personnel and the body state tables stored in the cloud database to obtain body state grades corresponding to the target personnel, wherein each basic body value corresponding to the target personnel corresponds to one body state grade, and meanwhile, the basic body values are matched with the prior probabilities corresponding to the body state grades, so that the prior probabilities corresponding to the target personnel are obtained, and the prior probabilities are marked as P (Hi);
Wherein, P (Hi) represents the prior probability of a specific physical state, which is based on a large amount of historical data and statistical information, reflecting the possibility that the individual is in a specific physical state without any additional information, i represents a specific physical state such as hypertension, diabetes, heart disease, etc.; for example, in a population of a particular age group, the prevalence of hypertension is 20%, then for a random individual of that age group, the prior probability P of suffering from hypertension (h_hypertension) is 0.2;
step three: physiological analysis: the device is used for monitoring physiological data parameters corresponding to the target personnel in the unit time period, so that the physiological states corresponding to the target personnel in the unit time period are evaluated and analyzed, and the specific analysis is as follows:
the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period are obtained by extracting the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period, and meanwhile, the numerical values of the highest heart rate, the lowest heart rate, the average heart rate, the highest body temperature, the lowest body temperature, the average body temperature, the highest blood pressure, the lowest blood pressure, the average blood pressure, the highest respiratory rate, the lowest respiratory rate, the average respiratory rate, the highest pulse beating times, the lowest pulse beating times and the average pulse beating times are extracted from the physiological data parameters, and are respectively calibrated into Qmax、Qmin、Qpjz、Wmax、Wmin、Wpjz、Emax、Emin、Epjz、Ymax、Ymin、Ypjz、Umax、Umin、Upjz, according to the formula: obtaining a physiological state evaluation coefficient delta 1 corresponding to the target person, wherein lambda 1, lambda 2, lambda 3, lambda 4 and lambda 5 are expressed as set weight factors;
Step four: habit analysis: the habit data monitoring system is used for monitoring habit data parameters corresponding to target personnel in a unit time period, so that habit states corresponding to the target personnel in the unit time period are evaluated and analyzed, and the specific analysis is as follows:
The method comprises the steps of obtaining the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period by extracting the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period, calibrating the dining time length, the activity time length and the rest time length into CS, DS and XS respectively, extracting the numerical values of the three, and carrying out normalization processing according to the formula: obtaining a habit state evaluation coefficient delta 2 corresponding to the target person, wherein CS*、DS* and XS* respectively represent a reference dining time length, a reference activity time length and a reference rest time length, and mu 1, mu 2 and mu 3 are represented as set weight factors;
Step five: conditional probability analysis: the method is used for receiving the physiological state evaluation coefficient and the habit state evaluation coefficient corresponding to the target person, so that the physical state corresponding to the target person in a unit time period is subjected to conditional probability analysis, and the specific analysis is as follows:
The values of the physiological state evaluation coefficient delta 1 and the habit state evaluation coefficient delta 2 corresponding to the target personnel are called for normalization processing, and the formula is used for: Obtaining a physical inertia value ZPZ corresponding to a target person, wherein delta 1max and delta 2max respectively represent the maximum value of a physiological state evaluation coefficient and the maximum value of a habit state evaluation coefficient, and a1 and a2 respectively represent the physiological state evaluation coefficient and the weight coefficient of the habit state evaluation coefficient corresponding to the target person;
matching and analyzing the physical inertia state values corresponding to the target personnel with a physical inertia state table stored in a cloud database to obtain physical inertia state levels corresponding to the target personnel, wherein each physical inertia state level corresponds to one physical inertia state level, and meanwhile, the physical inertia state levels are matched with the physical states corresponding to the physical inertia state levels to obtain the physical states corresponding to the target personnel;
assuming that Hi represents all possible physical states, the probability of occurrence of event E can be obtained by summing all possible physical states, according to the formula: p (E) = ΣiP(E|Hi)×P(Hi), resulting in the probability of occurrence of event E, P (e|hi) being the probability of occurrence of event E given the body state Hi, P (Hi) being the prior probability;
According to the Bayesian formula: Obtaining a conditional probability corresponding to a target person, wherein P (H|E) is the conditional probability, namely the probability that an individual is in a physical state H (such as hypertension) under the condition that an event E (such as specific physiological data and habit data) is observed; p (e|h) is a likelihood function representing the probability of observing event E under conditions where the individual is in physical state H, which is typically derived based on medical studies and statistical data; p (H) is a priori probability, i.e. the probability that an individual is in a physical state H before no data is observed; p (E) is the probability of event E occurring;
step six: disease risk prediction analysis: the method is used for carrying out posterior probability analysis on the physical state corresponding to the target personnel, so that disease risk prediction is carried out on the physical state corresponding to the target personnel, and the specific analysis is as follows:
By calling the prior probability and the conditional probability corresponding to the target person, according to a Bayesian formula: obtaining posterior probability corresponding to the target personnel;
Setting a posterior probability threshold corresponding to a target person, comparing and analyzing the posterior probability corresponding to the target person with the posterior probability threshold, judging that the target person has high trend disease risk when the posterior probability corresponding to the target person is larger than the posterior probability threshold, judging that the target person has medium trend disease risk when the posterior probability corresponding to the target person is equal to the posterior probability threshold, and judging that the target person has no trend disease risk when the posterior probability corresponding to the target person is smaller than the posterior probability threshold;
if the target person is judged to have high-trend disease risk, generating an early warning notification signal, and sending the disease risk condition to a corresponding monitoring management target person in a set G1 time period according to the generated early warning notification signal, wherein the monitoring management target person performs corresponding examination according to the disease risk condition, and adjusts the diet condition of the target person;
if the target person is judged to have the middle trend disease risk, generating an early warning attention signal, and sending the disease risk condition to the corresponding monitoring management target person in a set G2 time period according to the generated early warning attention signal so as to improve the attention degree of the target person;
As shown in fig. 2, the human health data analysis model based on the bayesian algorithm comprises a data acquisition module, a priori probability analysis module, a physiological analysis module, a habit analysis module, a conditional probability analysis module, a disease risk prediction analysis module and a display terminal;
the data acquisition module is used for acquiring human body data information corresponding to a target person to obtain human body data information corresponding to the target person, wherein the human body data comprises basic data parameters, physiological data parameters and habit data parameters, the basic data parameters comprise height, age, weight, gender and family disease history, the physiological data parameters comprise heart rate, body temperature, blood pressure, respiratory rate and pulse beating times, and the habit data parameters comprise dining time length, activity time length and rest time length;
the prior probability analysis module is used for monitoring basic data parameters corresponding to the target personnel to obtain basic body values corresponding to the target personnel, so that prior probability analysis is carried out on body states corresponding to the target personnel to obtain prior probabilities corresponding to the target personnel;
The physiological analysis module is used for monitoring physiological data parameters corresponding to the target personnel in the unit time period, so that the physiological states corresponding to the target personnel in the unit time period are evaluated and analyzed, and physiological state evaluation coefficients corresponding to the target personnel are obtained;
The habit analysis module is used for monitoring habit data parameters corresponding to the target personnel in the unit time period, so that habit states corresponding to the target personnel in the unit time period are evaluated and analyzed, and habit state evaluation coefficients corresponding to the target personnel are obtained;
The conditional probability analysis module is used for receiving the physiological state evaluation coefficient and the habit state evaluation coefficient corresponding to the target person, so that the conditional probability analysis is carried out on the physical state corresponding to the target person in the unit time period, and the specific analysis is as follows:
carrying out normalization calculation processing by calling the values of the physiological state evaluation coefficient and the habit state evaluation coefficient corresponding to the target personnel to obtain a physical inertia value corresponding to the target personnel;
matching and analyzing the physical inertia state values corresponding to the target personnel with a physical inertia state table stored in a cloud database to obtain physical inertia state levels corresponding to the target personnel, wherein each physical inertia state level corresponds to one physical inertia state level, and meanwhile, the physical inertia state levels are matched with the physical states corresponding to the physical inertia state levels to obtain the physical states corresponding to the target personnel;
assuming that Hi represents all possible physical states, the probability of occurrence of event E can be obtained by summing all possible physical states, according to the formula: p (E) = ΣiP(E|Hi)×P(Hi), resulting in the probability of occurrence of event E, P (e|hi) being the probability of occurrence of event E given the body state Hi, P (Hi) being the prior probability;
According to the Bayesian formula: Obtaining a conditional probability corresponding to a target person, wherein P (H|E) is the conditional probability, namely the probability that an individual is in a physical state H under the condition that an event E is observed; p (e|h) is a likelihood function representing the probability of observing event E under conditions where the individual is in physical state H, which is typically derived based on medical studies and statistical data; p (H) is a priori probability, i.e. the probability that an individual is in a physical state H before no data is observed; p (E) is the probability of event E occurring;
the disease risk prediction analysis module is used for performing posterior probability analysis on the physical state corresponding to the target personnel, so that disease risk prediction is performed on the physical state corresponding to the target personnel, and the specific analysis is as follows:
The prior probability and the conditional probability corresponding to the target personnel are called, and the posterior probability corresponding to the target personnel is obtained through calculation according to a Bayesian formula;
Setting a posterior probability threshold corresponding to a target person, comparing and analyzing the posterior probability corresponding to the target person with the posterior probability threshold, judging that the target person has high trend disease risk when the posterior probability corresponding to the target person is larger than the posterior probability threshold, judging that the target person has medium trend disease risk when the posterior probability corresponding to the target person is equal to the posterior probability threshold, and judging that the target person has no trend disease risk when the posterior probability corresponding to the target person is smaller than the posterior probability threshold;
If the target person is judged to have high-trend disease risk, generating an early warning notification signal, sending the disease risk condition to a corresponding monitoring management target person in a set G1 time period according to the generated early warning notification signal, arranging corresponding examination by the monitoring management target person according to the disease risk condition, adjusting the diet condition of the target person, and displaying and notifying at a display terminal;
If the target person is judged to have the middle trend disease risk, an early warning attention signal is generated, the disease risk status is sent to the corresponding monitoring management target person within the set G2 time period according to the generated early warning attention signal, so that the attention degree of the target person is improved, and the display terminal is used for displaying and notifying.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.

Claims (7)

The height value, the age value, the weight value, the sex value and the family disease history value corresponding to the target personnel are obtained by extracting the height, the age, the weight, the sex and the family disease history in the basic data parameters corresponding to the target personnel and respectively assigning the height, the age, the weight, the sex and the family disease history values, and are respectively calibrated into RT1, RT2, RT3, RT4 and RT5 according to a set data model: Obtaining a basic body value JBZ corresponding to a target person, wherein e1, e2, e3, e4 and e5 all represent natural constants, yz represents a preset correction factor, and eta 1, eta 2, eta 3, eta 4 and eta 5 respectively represent a height value, an age value, a weight value, a sex value and a family disease history value weight coefficient;
the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period are obtained by extracting the heart rate, the body temperature, the blood pressure, the respiratory rate and the pulse beating times in the physiological data parameters corresponding to the target person in the unit time period, and meanwhile, the numerical values of the highest heart rate, the lowest heart rate, the average heart rate, the highest body temperature, the lowest body temperature, the average body temperature, the highest blood pressure, the lowest blood pressure, the average blood pressure, the highest respiratory rate, the lowest respiratory rate, the average respiratory rate, the highest pulse beating times, the lowest pulse beating times and the average pulse beating times are extracted from the physiological data parameters, and are respectively calibrated into Qmax、Qmin、Qpjz、Wmax、Wmin、Wpjz、Emax、Emin、Epjz、Ymax、Ymin、Ypjz、Umax、Umin、Upjz, according to the formula: and obtaining a physiological state evaluation coefficient delta 1 corresponding to the target person, wherein lambda 1, lambda 2, lambda 3, lambda 4 and lambda 5 are expressed as set weight factors.
The method comprises the steps of obtaining the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period by extracting the dining time length, the activity time length and the rest time length in the habit state corresponding to the target person in the unit time period, calibrating the dining time length, the activity time length and the rest time length into CS, DS and XS respectively, extracting the numerical values of the three, and carrying out normalization processing according to the formula: And obtaining a habit state evaluation coefficient delta 2 corresponding to the target person, wherein CS*、DS* and XS* respectively represent a reference dining time length, a reference activity time length and a reference rest time length, and mu 1, mu 2 and mu 3 are represented as set weight factors.
CN202410416794.3A2024-04-082024-04-08Human health data analysis model and method based on Bayesian algorithmActiveCN118315059B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202410416794.3ACN118315059B (en)2024-04-082024-04-08Human health data analysis model and method based on Bayesian algorithm

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202410416794.3ACN118315059B (en)2024-04-082024-04-08Human health data analysis model and method based on Bayesian algorithm

Publications (2)

Publication NumberPublication Date
CN118315059Atrue CN118315059A (en)2024-07-09
CN118315059B CN118315059B (en)2024-10-15

Family

ID=91725109

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202410416794.3AActiveCN118315059B (en)2024-04-082024-04-08Human health data analysis model and method based on Bayesian algorithm

Country Status (1)

CountryLink
CN (1)CN118315059B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US20110319724A1 (en)*2006-10-302011-12-29Cox Paul GMethods and systems for non-invasive, internal hemorrhage detection
CN103164616A (en)*2013-02-022013-06-19杭州卓健信息科技有限公司Intelligent hospital guide system and intelligent hospital guide method
CN109864745A (en)*2019-01-082019-06-11国家康复辅具研究中心A kind of novel risk of stroke appraisal procedure and system
CN115188473A (en)*2022-07-102022-10-14天津中智云海软件科技有限公司Health management system and working method thereof
CN117038082A (en)*2023-08-212023-11-10福能(福州)健康体检中心有限公司Traditional Chinese medicine health state analysis and evaluation system and device based on big data

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US20110319724A1 (en)*2006-10-302011-12-29Cox Paul GMethods and systems for non-invasive, internal hemorrhage detection
CN103164616A (en)*2013-02-022013-06-19杭州卓健信息科技有限公司Intelligent hospital guide system and intelligent hospital guide method
CN109864745A (en)*2019-01-082019-06-11国家康复辅具研究中心A kind of novel risk of stroke appraisal procedure and system
CN115188473A (en)*2022-07-102022-10-14天津中智云海软件科技有限公司Health management system and working method thereof
CN117038082A (en)*2023-08-212023-11-10福能(福州)健康体检中心有限公司Traditional Chinese medicine health state analysis and evaluation system and device based on big data

Also Published As

Publication numberPublication date
CN118315059B (en)2024-10-15

Similar Documents

PublicationPublication DateTitle
CN115083607B (en)Human health index condition monitoring and analyzing method, system and storage medium
CN115547500B (en)Hemodialysis patient's health monitoring equipment and system
CN118039166B (en)Health monitoring management method based on medical big data
CN114758786B (en)Post-traumatic hemorrhagic shock dynamic early warning system based on noninvasive parameters
CN117874689B (en) A method for intelligent processing of heart rate monitoring data
CN119028512B (en)Obstetrical real-time nursing optimization method based on big data
CN106777872A (en)A kind of method that life of elderly person quality evaluation is carried out based on intelligent wearing technology
CN114098655B (en) A kind of intelligent sleep risk monitoring method and system
CN109564586A (en)System monitor and system monitoring method
CN118016302B (en)Health risk assessment method and system for multi-source data input
US10037412B2 (en)Patient health state compound score distribution and/or representative compound score based thereon
CN115662631A (en)AI intelligence discrimination-based nursing home management system
CN118098606A (en) A warning method, medium and system based on physical examination data
CN119049642B (en) An intelligent auxiliary system and method for dermatological rehabilitation nursing
CN118315059B (en)Human health data analysis model and method based on Bayesian algorithm
CN115956903A (en)Method for judging state abnormity of target object and storage medium
CN119581034A (en) A remote monitoring and nursing early warning system based on deep learning
CN120032885A (en) Intelligent nursing risk assessment and early warning system and method
CN118507075B (en) A method and system for processing urological postoperative vital sign monitoring data
CN119864166A (en)Prediction system and method for gestational hypertension risk
CN118280607B (en)Intelligent critical index monitoring method and system based on clinical data
CN110427848B (en)Mental analysis system
US20220253629A1 (en)Health caring system and health caring method
CN110211703A (en)A kind of sign abnormality alarming method and intelligent mattress system
CN115153548A (en)System for monitoring change of severity of depression and evaluation model construction method thereof

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp