CN118352073A - Disease risk prediction method and system based on patient physical examination data decision tree analysis - Google Patents
Disease risk prediction method and system based on patient physical examination data decision tree analysis Download PDFInfo
- Publication number
- CN118352073A CN118352073A CN202311289928.1A CN202311289928A CN118352073A CN 118352073 A CN118352073 A CN 118352073A CN 202311289928 A CN202311289928 A CN 202311289928A CN 118352073 A CN118352073 A CN 118352073A
- Authority
- CN
- China
- Prior art keywords
- risk
- disease
- patient
- physical examination
- historical
- 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.)
- Pending
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 308
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 308
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000003066 decision tree Methods 0.000 title claims abstract description 27
- 238000013058 risk prediction model Methods 0.000 claims abstract description 88
- 238000001514 detection method Methods 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000007477 logistic regression Methods 0.000 claims description 10
- 238000012502 risk assessment Methods 0.000 claims description 9
- 238000003745 diagnosis Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 4
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 230000036541 health Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002565 electrocardiography Methods 0.000 description 1
- 210000003608 fece Anatomy 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 210000002700 urine Anatomy 0.000 description 1
Abstract
The invention discloses a disease risk prediction method and a disease risk prediction system based on analysis of a patient physical examination data decision tree, which relate to the technical field of intelligent medical treatment and comprise the following steps: establishing a risk prediction model corresponding to various diseases one by one; acquiring current risk indexes of patients suffering from various diseases; judging whether the current risk index of the patient suffering from the disease is larger than a preset value; for the diseases with low risk, acquiring a plurality of historical risk indexes of the diseases of the patient; calculating a disease risk prediction model of the patient in the next integrated detection period; and judging whether the patient has a disease risk in the next period based on the disease risk prediction model of the patient in the next integrated detection period. In summary, the invention has the advantages that: the physical examination period of the patient can be intelligently adjusted based on the risk variation trend of the patient, and the personalized establishment of an intelligent and efficient comprehensive physical examination scheme for the patient can be realized.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to a disease risk prediction method and system based on analysis of a patient physical examination data decision tree.
Background
Physical examination is performed by medical means and methods, and includes basic examination of clinical departments, including examination of medical equipment such as ultrasound, electrocardiography and radiation, and laboratory examination of blood and urine and feces around a human body. Health physical examination is a health-centered physical examination.
In the prior art, analysis of physical examination data only can realize whether abnormal signs exist or not definitely exist through the physical examination data, comprehensive predictive analysis of the physical examination data is lacking, intelligent prediction of disease risk of a patient is difficult to carry out according to the physical examination data of the patient, and targeted adjustment of a physical examination scheme of the patient according to a physical examination data decision tree of the patient cannot be realized.
Disclosure of Invention
In order to solve the technical problems, the technical scheme solves the problems that in the prior art, whether abnormal signs exist or not can only be clearly realized through physical examination data in analysis of the physical examination data, whether health risks exist or not is only realized, comprehensive predictive analysis of the physical examination data is lacking, intelligent prediction of the disease risk of a patient is difficult to carry out according to the physical examination data of the patient, and targeted adjustment of the physical examination scheme of the patient according to the physical examination data decision tree of the patient cannot be realized.
In order to achieve the above purpose, the invention adopts the following technical scheme:
A disease risk prediction method based on patient physical examination data decision tree analysis, comprising:
based on historical medical data, establishing a risk prediction model corresponding to various diseases one by one;
Obtaining physical examination data of a current physical examination of a patient, substituting the physical examination data of the current physical examination of the patient into a risk prediction model, and obtaining current risk indexes of the patient suffering from various diseases;
judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if so, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low;
Obtaining historical physical examination data of a patient for the disease with low risk, substituting the historical physical examination data of the patient into a risk prediction model, and obtaining a plurality of historical risk indexes of the patient with the disease;
Analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, and calculating a disease risk prediction model of the patient in the next integrated detection period based on the risk trend of the patient suffering from the disease;
Based on a disease risk prediction model of the patient in the next integrated detection period, judging whether the patient has a disease risk in the next period, if so, adjusting the physical detection period, and if not, not responding.
Preferably, the establishing a risk prediction model corresponding to each disease one to one based on the historical medical data specifically includes:
determining a number of relevant physical examination data associated with the disease based on the historical medical data;
acquiring a plurality of diagnosis data of a disease, and dividing historical medical data into diseased data and non-diseased data according to diagnosis results;
Acquiring specific numerical values of a plurality of relevant physical examination data relevant to the disease in the disease data and the non-disease data;
establishing a Logistic regression model between a plurality of relevant physical examination data related to the disease probability and the disease;
And solving unknown coefficients of a Logistic regression model between the illness probability and the plurality of relevant physical examination data related to the illness according to a maximum likelihood method based on specific numerical values of the plurality of relevant physical examination data related to the illness in the illness data and the non-illness data, and obtaining a risk prediction model corresponding to the illness.
Preferably, the obtaining physical examination data of a current physical examination of the patient, and substituting the physical examination data of the current physical examination of the patient into the risk prediction model, the obtaining the current risk index of the patient suffering from various diseases specifically includes:
retrieving specific numerical values of a plurality of related physical examination data related to the disease from physical examination data of a current physical examination of a patient;
invoking a risk prediction model corresponding to the disease;
substituting specific numerical values of a plurality of related physical examination data related to the disease into a risk prediction model corresponding to the disease, and calculating to obtain the current risk index of the patient suffering from various diseases.
Preferably, the analyzing the risk trend of the patient suffering from the disease based on the plurality of historical risk indexes of the patient suffering from the disease and the current risk index of the patient suffering from the disease specifically includes:
Setting a physical examination history analysis time limit, and acquiring historical risk indexes of the patient suffering from the diseases corresponding to all the historical physical examination data of the patient in the physical examination history analysis time limit;
According to the starting time of the physical examination history analysis time limit is a time zero point, respectively calculating the time intervals of the historical physical examination data and the time zero point of all patients in the physical examination history analysis time limit;
Calculating the risk trend of the patient suffering from the disease according to a risk trend index calculation formula based on the historical risk indexes of the patient suffering from the disease corresponding to all the historical physical examination data of the patient in the physical examination historical analysis time limit and the time intervals of the historical physical examination data and the moment zero point of all the patient in the physical examination historical analysis time limit;
the risk trend index calculation formula specifically comprises:
Wherein Z is the risk trend of the patient suffering from the disease, n is the total number of the historical physical examination data of the patient in the physical examination historical analysis time limit, t i is the moment corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit, and G i is the historical risk index of the patient suffering from the disease corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit.
Preferably, the calculating the disease risk prediction model of the patient in the next integrated detection period based on the risk trend of the patient suffering from the disease specifically includes:
Calculating the risk trend of the patient suffering from the disease in a plurality of continuous physical examination history analysis time periods, and recording the risk trend as a history risk trend;
numbering a plurality of historical risk trends according to the time limit of the physical examination historical analysis and the distance between the current intervals;
marking coordinate points corresponding to the historical risk trend and the number in a coordinate system by taking the number as an abscissa and taking the historical risk trend as an ordinate;
At least one known function model is called, a known function image is generated in a coordinate system, a plurality of function models which are preliminarily fitted with coordinate points corresponding to historical risk trends and numbers are preliminarily screened out through image recognition, and the function models are recorded as candidate function models;
calculating the average error of the candidate function model and the coordinate points corresponding to the historical risk trend and the serial number, and taking the function model with the minimum average error as a fitting model;
solving unknown parameters in the fitting model according to a least square method, obtaining a functional relation between a historical risk trend and a serial number, and marking the functional relation as F (x);
calculating a function value of F (m+1) based on a functional relation between the historical risk trend and the number, wherein m is the total number of the historical risk trend;
Establishing a disease risk prediction model of a patient in the next integrated detection period;
wherein, the expression of the disease risk prediction model of the patient in the following integrated detection period is as follows:
G(t)＝G0+F(m+1)×t
Wherein G (t) is a predicted value of a disease risk index of the patient in the next integrated detection period, G 0 is a current risk index of the patient suffering from the disease, and t is a time interval from the current moment.
Preferably, the determining whether the patient has a disease risk in the next cycle based on the disease risk prediction model of the patient in the next cycle specifically includes:
acquiring a period interval of the next physical examination of the patient;
Substituting the period interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination period to obtain a disease risk index prediction value in the next physical examination;
Judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
Preferably, the adjusting the physical examination period specifically includes:
Calculating a time interval from the current moment to the current moment when the output value of the disease risk prediction model of the patient in the next integrated detection period is equal to a preset value, and recording the time interval as an adjustment time interval;
the adjustment time interval is taken as the period interval of the next physical examination.
Furthermore, a disease risk prediction system based on analysis of a patient physical examination data decision tree is provided, which is used for implementing the disease risk prediction method based on analysis of the patient physical examination data decision tree, and the disease risk prediction system comprises:
the risk model module is used for establishing a risk prediction model corresponding to various diseases one by one based on historical medical data;
the physical examination risk analysis module is electrically connected with the first model module and is used for acquiring physical examination data of a patient's current physical examination, substituting the physical examination data of the patient's current physical examination into a risk prediction model, acquiring current risk indexes of the patient suffering from various diseases and judging whether the current risk indexes of the patient suffering from the diseases are larger than a preset value;
The prediction analysis module is electrically connected with the risk model module and the physical examination risk analysis module, and is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, calculating a disease risk prediction model of the patient in the next integrated examination period based on the risk trend of the patient suffering from the disease and judging whether the disease risk exists in the next period based on the disease risk prediction model of the patient in the next integrated examination period.
Optionally, the physical examination risk analysis module includes:
The risk calculation unit is used for retrieving specific values of a plurality of relevant physical examination data related to the disease from physical examination data of a patient, retrieving a risk prediction model corresponding to the disease, substituting the specific values of the plurality of relevant physical examination data related to the disease into the risk prediction model corresponding to the disease, and calculating to obtain a current risk index of the patient suffering from various diseases;
The first judging unit is used for judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if yes, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low.
Optionally, the prediction analysis module includes:
The risk trend calculation unit is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and current risk indexes of the patient suffering from the disease;
A risk prediction model unit for calculating a disease risk prediction model of the patient in a next integrated test period based on a risk trend of the patient suffering from the disease;
the risk prediction calculation unit is used for obtaining the cycle interval of the next physical examination of the patient, substituting the cycle interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination cycle, and obtaining a disease risk index prediction value in the next physical examination;
The second judging unit is used for judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
Compared with the prior art, the invention has the beneficial effects that:
The invention provides a disease risk prediction scheme based on analysis of a patient physical examination data decision tree, which is based on a Logistic regression model theory, analyzes the risk probability value of various diseases of a patient based on the physical examination data of the patient, establishes a physical examination health file for the historical physical examination data of the patient, comprehensively analyzes the risk variation trend of the diseases of the patient based on the historical physical examination data of the patient in the current life style of the patient, intelligently adjusts the physical examination period of the patient based on the risk variation trend of the diseases of the patient, further ensures that the patient can timely find the health state problem of the patient, and is convenient for establishing an intelligent and efficient comprehensive physical examination scheme for individuation of the patient.
Drawings
FIG. 1 is a flow chart of a disease risk prediction method based on patient physical examination data decision tree analysis;
FIG. 2 is a flow chart of a method for establishing a risk prediction model according to the present invention;
FIG. 3 is a flow chart of a method of obtaining current risk indicators of patients suffering from various diseases according to the present invention;
FIG. 4 is a flow chart of a method of analyzing a patient's risk trend of suffering from a disease according to the present invention;
FIG. 5 is a flow chart of a method of calculating a patient risk prediction model during a next screening cycle in accordance with the present invention;
FIG. 6 is a flowchart of a method of determining whether a patient is at risk for developing a disease in a next cycle according to the present invention;
FIG. 7 is a flow chart of a method for adjusting a physical examination period according to the present invention;
FIG. 8 is a block diagram of a system for predicting risk of a disease based on analysis of a decision tree of physical examination data of a patient according to the present invention.
Detailed Description
The following description is presented to enable one of ordinary skill in the art to make and use the invention. The preferred embodiments in the following description are by way of example only and other obvious variations will occur to those skilled in the art.
Referring to fig. 1, a disease risk prediction method based on patient physical examination data decision tree analysis includes:
based on historical medical data, establishing a risk prediction model corresponding to various diseases one by one;
Obtaining physical examination data of a current physical examination of a patient, substituting the physical examination data of the current physical examination of the patient into a risk prediction model, and obtaining current risk indexes of the patient suffering from various diseases;
judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if so, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low;
Obtaining historical physical examination data of a patient for the disease with low risk, substituting the historical physical examination data of the patient into a risk prediction model, and obtaining a plurality of historical risk indexes of the patient with the disease;
Analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, and calculating a disease risk prediction model of the patient in the next integrated detection period based on the risk trend of the patient suffering from the disease;
Based on a disease risk prediction model of the patient in the next integrated detection period, judging whether the patient has a disease risk in the next period, if so, adjusting the physical detection period, and if not, not responding.
According to the scheme, based on the Logistic regression model theory, through the physical examination data of the patient as a basis, the risk probability value of various diseases of the patient is analyzed, meanwhile, physical examination health files are established aiming at the historical physical examination data of the patient, comprehensive analysis is carried out on the risk variation trend of the patient under the current life style of the patient based on the historical physical examination data of the patient, and intelligent adjustment of the physical examination period of the patient aiming at the risk variation trend of the patient is realized.
Referring to fig. 2, the establishing a risk prediction model corresponding to each disease one-to-one based on the historical medical data specifically includes:
determining a number of relevant physical examination data associated with the disease based on the historical medical data;
acquiring a plurality of diagnosis data of a disease, and dividing historical medical data into diseased data and non-diseased data according to diagnosis results;
Acquiring specific numerical values of a plurality of relevant physical examination data relevant to the disease in the disease data and the non-disease data;
establishing a Logistic regression model between a plurality of relevant physical examination data related to the disease probability and the disease;
And solving unknown coefficients of a Logistic regression model between the illness probability and the plurality of relevant physical examination data related to the illness according to a maximum likelihood method based on specific numerical values of the plurality of relevant physical examination data related to the illness in the illness data and the non-illness data, and obtaining a risk prediction model corresponding to the illness.
Referring to fig. 3, the steps of obtaining physical examination data of a patient's current physical examination, substituting the physical examination data of the patient's current physical examination into a risk prediction model, and obtaining current risk indexes of the patient suffering from various diseases specifically include:
retrieving specific numerical values of a plurality of related physical examination data related to the disease from physical examination data of a current physical examination of a patient;
invoking a risk prediction model corresponding to the disease;
substituting specific numerical values of a plurality of related physical examination data related to the disease into a risk prediction model corresponding to the disease, and calculating to obtain the current risk index of the patient suffering from various diseases.
The Logistic regression model is a generalized linear regression analysis model and is commonly used in the fields of data mining, automatic disease diagnosis, economic prediction and the like. The Logistic regression model estimates the occurrence probability of an event according to a given autotransformer data set, and because the result is a probability, the range of the dependent variable is between 0 and 1.
Referring to fig. 4, the analyzing the risk trend of the patient suffering from the disease based on the plurality of historical risk indexes of the patient suffering from the disease and the current risk index of the patient suffering from the disease specifically includes:
Setting a physical examination history analysis time limit, and acquiring historical risk indexes of the patient suffering from the diseases corresponding to all the historical physical examination data of the patient in the physical examination history analysis time limit;
According to the starting time of the physical examination history analysis time limit is a time zero point, respectively calculating the time intervals of the historical physical examination data and the time zero point of all patients in the physical examination history analysis time limit;
Calculating the risk trend of the patient suffering from the disease according to a risk trend index calculation formula based on the historical risk indexes of the patient suffering from the disease corresponding to all the historical physical examination data of the patient in the physical examination historical analysis time limit and the time intervals of the historical physical examination data and the moment zero point of all the patient in the physical examination historical analysis time limit;
the risk trend index calculation formula specifically comprises:
Wherein Z is the risk trend of the patient suffering from the disease, n is the total number of the historical physical examination data of the patient in the physical examination historical analysis time limit, t i is the moment corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit, and G i is the historical risk index of the patient suffering from the disease corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit.
The risk trend shows that, whether the change law of patient's sick risk, it can be better reflects patient's life style can lead to the disease to take place, and the risk trend is bigger, then indicates that patient's sick risk probability improves faster, and patient's life style is more easily led to the emergence of disease, because under the normal condition, patient's life style is usually difficult to carry out quick transition, consequently need carry out intelligent adjustment patient's physical examination's interval to the change of risk trend to patient's health condition that can be more accurate assurance oneself.
Referring to fig. 5, the calculating a disease risk prediction model of the patient in the next integrated test period based on the risk trend of the patient suffering from the disease specifically includes:
Calculating the risk trend of the patient suffering from the disease in a plurality of continuous physical examination history analysis time periods, and recording the risk trend as a history risk trend;
numbering a plurality of historical risk trends according to the time limit of the physical examination historical analysis and the distance between the current intervals;
marking coordinate points corresponding to the historical risk trend and the number in a coordinate system by taking the number as an abscissa and taking the historical risk trend as an ordinate;
At least one known function model is called, a known function image is generated in a coordinate system, a plurality of function models which are preliminarily fitted with coordinate points corresponding to historical risk trends and numbers are preliminarily screened out through image recognition, and the function models are recorded as candidate function models;
calculating the average error of the candidate function model and the coordinate points corresponding to the historical risk trend and the serial number, and taking the function model with the minimum average error as a fitting model;
solving unknown parameters in the fitting model according to a least square method, obtaining a functional relation between a historical risk trend and a serial number, and marking the functional relation as F (x);
calculating a function value of F (m+1) based on a functional relation between the historical risk trend and the number, wherein m is the total number of the historical risk trend;
Establishing a disease risk prediction model of a patient in the next integrated detection period;
wherein, the expression of the disease risk prediction model of the patient in the following integrated detection period is as follows:
G(t)＝G0+F(m+1)×t
Wherein G (t) is a predicted value of a disease risk index of the patient in the next integrated detection period, G 0 is a current risk index of the patient suffering from the disease, and t is a time interval from the current moment.
Referring to fig. 6, the determining whether the patient has a risk of developing a disease in the next cycle based on the disease risk prediction model of the patient in the next cycle specifically includes:
acquiring a period interval of the next physical examination of the patient;
Substituting the period interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination period to obtain a disease risk index prediction value in the next physical examination;
Judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
Referring to fig. 7, the adjusting the physical examination period specifically includes:
Calculating a time interval from the current moment to the current moment when the output value of the disease risk prediction model of the patient in the next integrated detection period is equal to a preset value, and recording the time interval as an adjustment time interval;
the adjustment time interval is taken as the period interval of the next physical examination.
According to the method, the change trend of the patient's disease risk in the following time is researched and judged through fitting and calculating the functional relation between the historical risk trend and the serial number, the disease risk prediction model of the patient in the next integrated detection period is built based on the patient's disease risk, the disease risk value of the patient in the next integrated detection period is estimated through the disease risk prediction model of the patient in the next integrated detection period, the intelligent adjustment of the patient physical examination is carried out, the intelligent and efficient comprehensive physical examination scheme can be built for the patient in an individualized way, and the patient can acquire the health condition data of the patient more accurately and efficiently.
Further, referring to fig. 8, based on the same inventive concept as the disease risk prediction method based on the analysis of the patient physical examination data decision tree, the present disclosure proposes a disease risk prediction system based on the analysis of the patient physical examination data decision tree, including:
the risk model module is used for establishing a risk prediction model corresponding to various diseases one by one based on historical medical data;
the physical examination risk analysis module is electrically connected with the first model module and is used for acquiring physical examination data of a patient's current physical examination, substituting the physical examination data of the patient's current physical examination into a risk prediction model, acquiring current risk indexes of the patient suffering from various diseases and judging whether the current risk indexes of the patient suffering from the diseases are larger than a preset value;
The prediction analysis module is electrically connected with the risk model module and the physical examination risk analysis module, and is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, calculating a disease risk prediction model of the patient in the next integrated examination period based on the risk trend of the patient suffering from the disease and judging whether the disease risk exists in the next period based on the disease risk prediction model of the patient in the next integrated examination period.
The physical examination risk analysis module comprises:
The risk calculation unit is used for retrieving specific values of a plurality of relevant physical examination data related to the disease from physical examination data of a patient, retrieving a risk prediction model corresponding to the disease, substituting the specific values of the plurality of relevant physical examination data related to the disease into the risk prediction model corresponding to the disease, and calculating to obtain a current risk index of the patient suffering from various diseases;
The first judging unit is used for judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if yes, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low.
The predictive analysis module includes:
The risk trend calculation unit is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and current risk indexes of the patient suffering from the disease;
A risk prediction model unit for calculating a disease risk prediction model of the patient in a next integrated test period based on a risk trend of the patient suffering from the disease;
the risk prediction calculation unit is used for obtaining the cycle interval of the next physical examination of the patient, substituting the cycle interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination cycle, and obtaining a disease risk index prediction value in the next physical examination;
The second judging unit is used for judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
The disease risk prediction system based on the patient physical examination data decision tree analysis comprises the following using processes:
Step one: the risk model module establishes a risk prediction model corresponding to various diseases one by one based on the historical medical data;
Step two: the risk calculation unit retrieves specific values of a plurality of related physical examination data related to the disease from physical examination data of a patient, retrieves a risk prediction model corresponding to the disease, and then substitutes the specific values of the plurality of related physical examination data related to the disease into the risk prediction model corresponding to the disease, so as to calculate and obtain current risk indexes of the patient suffering from various diseases;
Step three: the first judging unit judges whether the current risk index of the patient suffering from the disease is larger than a preset value, if so, the risk of the patient suffering from the disease is judged to be high, and if not, the risk of the patient suffering from the disease is judged to be low;
Step four: the risk trend calculation unit analyzes the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and current risk indexes of the patient suffering from the disease;
step five: the risk prediction model unit calculates a disease risk prediction model of the patient in the next integrated detection period based on the risk trend of the patient suffering from the disease;
Step six: the risk prediction calculation unit acquires the cycle interval of the next physical examination of the patient and substitutes the cycle interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination cycle to obtain a disease risk index prediction value in the next physical examination;
step seven: the second judging unit judges whether the predicted value of the disease risk index is larger than a preset value or not in the next physical examination, if yes, the patient is judged to have the disease risk in the next period, and if not, the patient is judged to not have the disease risk in the next period.
In summary, the invention has the advantages that: the physical examination period of the patient can be intelligently adjusted based on the risk variation trend of the patient, and the personalized establishment of an intelligent and efficient comprehensive physical examination scheme for the patient can be realized.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention, which is defined by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. A disease risk prediction method based on patient physical examination data decision tree analysis, comprising:
based on historical medical data, establishing a risk prediction model corresponding to various diseases one by one;
Obtaining physical examination data of a current physical examination of a patient, substituting the physical examination data of the current physical examination of the patient into a risk prediction model, and obtaining current risk indexes of the patient suffering from various diseases;
judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if so, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low;
Obtaining historical physical examination data of a patient for the disease with low risk, substituting the historical physical examination data of the patient into a risk prediction model, and obtaining a plurality of historical risk indexes of the patient with the disease;
Analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, and calculating a disease risk prediction model of the patient in the next integrated detection period based on the risk trend of the patient suffering from the disease;
Based on a disease risk prediction model of the patient in the next integrated detection period, judging whether the patient has a disease risk in the next period, if so, adjusting the physical detection period, and if not, not responding.
2. The disease risk prediction method based on patient physical examination data decision tree analysis according to claim 1, wherein the establishing a risk prediction model corresponding to each disease one-to-one based on the historical medical data specifically comprises:
determining a number of relevant physical examination data associated with the disease based on the historical medical data;
acquiring a plurality of diagnosis data of a disease, and dividing historical medical data into diseased data and non-diseased data according to diagnosis results;
Acquiring specific numerical values of a plurality of relevant physical examination data relevant to the disease in the disease data and the non-disease data;
establishing a Logistic regression model between a plurality of relevant physical examination data related to the disease probability and the disease;
And solving unknown coefficients of a Logistic regression model between the illness probability and the plurality of relevant physical examination data related to the illness according to a maximum likelihood method based on specific numerical values of the plurality of relevant physical examination data related to the illness in the illness data and the non-illness data, and obtaining a risk prediction model corresponding to the illness.
3. The disease risk prediction method based on patient physical examination data decision tree analysis according to claim 2, wherein the steps of obtaining physical examination data of a patient's current physical examination, substituting the physical examination data of the patient's current physical examination into a risk prediction model, and obtaining current risk indexes of the patient suffering from various diseases specifically include:
retrieving specific numerical values of a plurality of related physical examination data related to the disease from physical examination data of a current physical examination of a patient;
invoking a risk prediction model corresponding to the disease;
substituting specific numerical values of a plurality of related physical examination data related to the disease into a risk prediction model corresponding to the disease, and calculating to obtain the current risk index of the patient suffering from various diseases.
4. A method for disease risk prediction based on a decision tree analysis of physical examination data of a patient according to claim 3, wherein the analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indicators of the patient suffering from the disease and a current risk indicator of the patient suffering from the disease specifically comprises:
Setting a physical examination history analysis time limit, and acquiring historical risk indexes of the patient suffering from the diseases corresponding to all the historical physical examination data of the patient in the physical examination history analysis time limit;
According to the starting time of the physical examination history analysis time limit is a time zero point, respectively calculating the time intervals of the historical physical examination data and the time zero point of all patients in the physical examination history analysis time limit;
Calculating the risk trend of the patient suffering from the disease according to a risk trend index calculation formula based on the historical risk indexes of the patient suffering from the disease corresponding to all the historical physical examination data of the patient in the physical examination historical analysis time limit and the time intervals of the historical physical examination data and the moment zero point of all the patient in the physical examination historical analysis time limit;
the risk trend index calculation formula specifically comprises:
Wherein Z is the risk trend of the patient suffering from the disease, n is the total number of the historical physical examination data of the patient in the physical examination historical analysis time limit, t i is the moment corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit, and G i is the historical risk index of the patient suffering from the disease corresponding to the ith historical physical examination data of the patient in the physical examination historical analysis time limit.
5. The disease risk prediction method based on the analysis of the decision tree of the physical examination data of the patient according to claim 4, wherein the calculating the disease risk prediction model of the patient in the next integrated examination period based on the risk trend of the patient suffering from the disease specifically comprises:
Calculating the risk trend of the patient suffering from the disease in a plurality of continuous physical examination history analysis time periods, and recording the risk trend as a history risk trend;
numbering a plurality of historical risk trends according to the time limit of the physical examination historical analysis and the distance between the current intervals;
marking coordinate points corresponding to the historical risk trend and the number in a coordinate system by taking the number as an abscissa and taking the historical risk trend as an ordinate;
At least one known function model is called, a known function image is generated in a coordinate system, a plurality of function models which are preliminarily fitted with coordinate points corresponding to historical risk trends and numbers are preliminarily screened out through image recognition, and the function models are recorded as candidate function models;
calculating the average error of the candidate function model and the coordinate points corresponding to the historical risk trend and the serial number, and taking the function model with the minimum average error as a fitting model;
solving unknown parameters in the fitting model according to a least square method, obtaining a functional relation between a historical risk trend and a serial number, and marking the functional relation as F (x);
calculating a function value of F (m+1) based on a functional relation between the historical risk trend and the number, wherein m is the total number of the historical risk trend;
Establishing a disease risk prediction model of a patient in the next integrated detection period;
wherein, the expression of the disease risk prediction model of the patient in the following integrated detection period is as follows:
G(t)＝G0+F(m+1)×t
Wherein G (t) is a predicted value of a disease risk index of the patient in the next integrated detection period, G 0 is a current risk index of the patient suffering from the disease, and t is a time interval from the current moment.
6. The disease risk prediction method based on the analysis of the decision tree of the physical examination data of the patient according to claim 5, wherein the determining whether the patient has a disease risk in the next cycle based on the disease risk prediction model of the patient in the next cycle specifically comprises:
acquiring a period interval of the next physical examination of the patient;
Substituting the period interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination period to obtain a disease risk index prediction value in the next physical examination;
Judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
7. The method for predicting risk of a disease based on decision tree analysis of physical examination data of a patient as claimed in claim 6, wherein said adjusting the physical examination cycle comprises:
Calculating a time interval from the current moment to the current moment when the output value of the disease risk prediction model of the patient in the next integrated detection period is equal to a preset value, and recording the time interval as an adjustment time interval;
the adjustment time interval is taken as the period interval of the next physical examination.
8. A disease risk prediction system based on patient physical examination data decision tree analysis, for implementing a disease risk prediction method based on patient physical examination data decision tree analysis as claimed in any one of claims 1 to 7, comprising:
the risk model module is used for establishing a risk prediction model corresponding to various diseases one by one based on historical medical data;
the physical examination risk analysis module is electrically connected with the first model module and is used for acquiring physical examination data of a patient's current physical examination, substituting the physical examination data of the patient's current physical examination into a risk prediction model, acquiring current risk indexes of the patient suffering from various diseases and judging whether the current risk indexes of the patient suffering from the diseases are larger than a preset value;
The prediction analysis module is electrically connected with the risk model module and the physical examination risk analysis module, and is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and the current risk indexes of the patient suffering from the disease, calculating a disease risk prediction model of the patient in the next integrated examination period based on the risk trend of the patient suffering from the disease and judging whether the disease risk exists in the next period based on the disease risk prediction model of the patient in the next integrated examination period.
9. The disease risk prediction system based on patient physical examination data decision tree analysis of claim 8, wherein the physical examination risk analysis module comprises:
The risk calculation unit is used for retrieving specific values of a plurality of relevant physical examination data related to the disease from physical examination data of a patient, retrieving a risk prediction model corresponding to the disease, substituting the specific values of the plurality of relevant physical examination data related to the disease into the risk prediction model corresponding to the disease, and calculating to obtain a current risk index of the patient suffering from various diseases;
The first judging unit is used for judging whether the current risk index of the patient suffering from the disease is larger than a preset value, if yes, judging that the risk of the patient suffering from the disease is high, and if not, judging that the risk of the patient suffering from the disease is low.
10. The disease risk prediction system based on patient physical examination data decision tree analysis of claim 8, wherein the predictive analysis module comprises:
The risk trend calculation unit is used for analyzing the risk trend of the patient suffering from the disease based on a plurality of historical risk indexes of the patient suffering from the disease and current risk indexes of the patient suffering from the disease;
A risk prediction model unit for calculating a disease risk prediction model of the patient in a next integrated test period based on a risk trend of the patient suffering from the disease;
the risk prediction calculation unit is used for obtaining the cycle interval of the next physical examination of the patient, substituting the cycle interval of the next physical examination of the patient into a disease risk prediction model of the patient in the next physical examination cycle, and obtaining a disease risk index prediction value in the next physical examination;
The second judging unit is used for judging whether the predicted value of the disease risk index in the next physical examination is larger than a preset value, if so, judging that the patient has a disease risk in the next period, and if not, judging that the patient does not have the disease risk in the next period.
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118352073A true CN118352073A (en) | 2024-07-16 |
Family
ID=
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200337580A1 (en) | Time series data learning and analysis method using artificial intelligence | |
| US6443889B1 (en) | Provision of decision support for acute myocardial infarction | |
| CN110739076A (en) | medical artificial intelligence public training platform | |
| CN112786204A (en) | Machine learning diabetes onset risk prediction method and application | |
| JP2023016928A (en) | Automated verification of medical data | |
| CN107145715B (en) | Clinical medicine intelligence discriminating gear based on electing algorithm | |
| CN113838577B (en) | Convenient layered old people MODS early death risk assessment model, device and establishment method | |
| CN112786203A (en) | Machine learning diabetic retinopathy morbidity risk prediction method and application | |
| CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
| CN115714022B (en) | Neonatal jaundice health management system based on artificial intelligence | |
| CN105787232B (en) | A kind of data processing method, device, health system platform and terminal | |
| CN111317458A (en) | Blood pressure detection system based on deep learning | |
| CN117238460A (en) | Illness state monitoring method and system based on nursing morning shift data | |
| CN117352175A (en) | Fracture intelligent auxiliary monitoring system for orthopedic nursing | |
| CN117116475A (en) | Method, system, terminal and storage medium for predicting risk of ischemic cerebral apoplexy | |
| CN118352073A (en) | Disease risk prediction method and system based on patient physical examination data decision tree analysis | |
| Alotaibi et al. | Early prediction of gestational diabetes using machine learning techniques | |
| CN115414054A (en) | Epilepsia electroencephalogram detection system based on feedforward pulse neural network | |
| CN114999647A (en) | Intelligent health monitoring method and system based on big data | |
| Deepa et al. | Experimental evaluation of artificial intelligence assisted heart disease prediction using deep learning principle | |
| CN110706805A (en) | Artificial intelligence medical diagnosis system based on feature selection | |
| AU2021102832A4 (en) | System & method for automatic health prediction using fuzzy based machine learning | |
| CN118016302B (en) | Health risk assessment method and system for multi-source data input | |
| Kim et al. | Untact Abnormal Heartbeat Wave Detection Using Non-Contact Sensor through Transfer Learning | |
| CN116452584B (en) | Neonatal retinopathy prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication |