CN115862894B - Coronary heart disease research method and system based on least square estimation and privacy protection - Google Patents
Coronary heart disease research method and system based on least square estimation and privacy protection Download PDFInfo
- Publication number
- CN115862894B CN115862894B CN202211620092.4A CN202211620092A CN115862894B CN 115862894 B CN115862894 B CN 115862894B CN 202211620092 A CN202211620092 A CN 202211620092A CN 115862894 B CN115862894 B CN 115862894B
- Authority
- CN
- China
- Prior art keywords
- coronary heart
- heart disease
- general
- data
- hospital
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to the technical field of coronary heart disease research, in particular to a coronary heart disease research method and a coronary heart disease research system based on least square estimation and privacy protection, wherein the method comprises the following steps: s1, carrying out differential privacy protection on the coronary heart disease data by a coronary heart disease special hospital, and transmitting the coronary heart disease data subjected to differential privacy protection to a research institution; s2, the plurality of general hospitals conduct differential privacy protection on own general disease data, and the general disease data subjected to differential privacy protection are sent to the research institution; s3, the research institution establishes an MP-ARMAX model of multiple participants according to the received data; s4, the research institution obtains relevant parameters of the coronary heart disease affected by other diseases according to the MP-ARMAX model. The invention can accurately study the condition of coronary heart disease affected by other diseases.
Description
Technical Field
The invention relates to the technical field of coronary heart disease research, in particular to a coronary heart disease research method and system based on least square estimation and privacy protection.
Background
Coronary atherosclerotic heart disease (coronary atherosclerotic heart disease) is abbreviated as coronary heart disease, and refers to a condition that due to abnormal lipid metabolism, lipids in blood deposit on an originally smooth intima of an artery, and some atheromatous lipid substances accumulate on the intima of the artery to form a white plaque, which is called an atherosclerosis lesion. These plaques gradually increase to narrow the arterial lumen, block blood flow, cause ischemia of the heart, and produce angina pectoris.
The heart is an important organ of the human body and acts as a pump that never stops working, and with each contraction of the heart, blood flow carrying oxygen and nutrients is delivered throughout the body via the aorta to supply the metabolic needs of the cells of each tissue. How does the heart's own oxygen and nutrition get? Two arteries are formed at the root of the aorta and are responsible for the blood circulation of the heart itself, called coronary arteries. Due to abnormal lipid metabolism, lipids in the blood deposit on the otherwise smooth intima of the artery, where some atheromatous lipid material accumulates to form white plaques, known as atherosclerotic lesions. Coronary heart disease is classified as latent: patients have coronary sclerosis, but lesions are lighter or have better collateral circulation, or patients have higher pain threshold and thus no pain symptoms; angina pectoris type: on the basis of coronary artery stenosis, a clinical syndrome of acute, transient ischemia and hypoxia of the heart muscle is caused by an increase in myocardial load; myocardial infarction type: on the basis of coronary artery lesions, the coronary blood supply is suddenly reduced or interrupted, so that the corresponding cardiac muscle is seriously and permanently subjected to acute ischemia to cause myocardial necrosis; heart failure: (ischemic cardiomyopathy) myocardial fibrosis, long-term insufficiency of blood supply to the myocardium, and malnutrition and atrophy of myocardial tissue, or fibrous tissue hyperplasia after large-area myocardial infarction; sudden death: the occurrence of sudden cardiac arrest in patients with classified criteria is caused by the occurrence of coronary artery spasm or embolism on the basis of atherosclerosis, which leads to acute ischemia of cardiac muscle, causes local electrophysiological disturbance, and causes temporary severe arrhythmia.
The main cause of coronary heart disease is coronary atherosclerosis, but the cause of atherosclerosis is not completely clear, and may be the result of the combined action of multiple factors. The risk factors for the occurrence of this disease are considered to be: age and sex (male over 45 years old, female over 55 years old or postmenopausal), family history (father and brothers die of heart disease before 55 years old, mother/sister before 65 years old), dyslipidemia (low density lipoprotein cholesterol LDL-C is too high, high density lipoprotein cholesterol HDL-C is too low), hypertension, uroglycoses, smoking, overweight, obesity, gout, lack of exercise, etc. That is, coronary heart disease and other diseases have many correlations and are greatly affected by other diseases. There is therefore a need for a method of studying the effects of coronary heart disease on other diseases.
Disclosure of Invention
The invention provides a coronary heart disease research method and a coronary heart disease research system based on least square estimation and privacy protection, which are used for researching the influence condition of other diseases on coronary heart disease. The technical scheme is as follows:
in one aspect, a coronary heart disease research method based on least squares estimation and privacy protection is provided, the method comprising:
s1, carrying out differential privacy protection on the coronary heart disease data by a coronary heart disease special hospital, and transmitting the coronary heart disease data subjected to differential privacy protection to a research institution;
s2, the plurality of general hospitals conduct differential privacy protection on own general disease data, and the general disease data subjected to differential privacy protection are sent to the research institution;
s3, the research institution establishes an MP-ARMAX model of multiple participants according to the received data;
and S4, calculating and obtaining relevant parameters of the coronary heart disease affected by other diseases by the research institution according to the MP-ARMAX model.
Optionally, the coronary heart disease special hospital in S1 performs differential privacy protection on its own coronary heart disease data, which specifically includes:
the coronary heart disease special hospital adds first privacy protection noise to the coronary heart disease data of the coronary heart disease special hospital, wherein the first privacy protection noise is obtained by following that the variance is 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Representing P 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protection, epsilon represents the privacy protectionBudget.
Optionally, the multiple general hospitals of S2 perform differential privacy protection on own general disease data, which specifically includes:
the plurality of general hospitals add a second privacy-preserving noise to the own general disease data, the second privacy-preserving noise being a noise with a following variance of 2b 2 (k) The plurality of general hospitals selecting the noise parameter b of the second privacy-preserving noise i Satisfies the following conditionsC i,2 Representing P i Is a sensitive degree of (a).
Optionally, the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the coronary heart disease special department hospital,historical general disease data for the first of said general hospitals,/i->Historical general disease data for a second of the general hospitals,Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, calculates the correlation parameter b of the coronary heart disease affected by other diseases i,k And a correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j ,k=1,...,q i ,j=1,...,p,q i And p is a known systemAn order.
Optionally, calculating the correlation parameter b i,k And a j The method specifically comprises the following steps:
At the time instant k +1,
wherein θ is k Is an estimated value of θ at time k, θ k+1 Is an estimate of theta at time k +1,representing a matrixT is the transpose of the corresponding matrix, pairAt any given small positive value α, +.> I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at 0 moment, and is obtained by continuous iterative calculation until K approaches infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
In another aspect, there is also provided a coronary heart disease research system based on least squares estimation and privacy protection, the system comprising: coronary disease specialty hospitals, a plurality of general hospitals and research institutions, wherein,
the special coronary heart disease hospital is used for carrying out differential privacy protection on the coronary heart disease data of the special coronary heart disease hospital and sending the coronary heart disease data subjected to differential privacy protection to the research institution;
the plurality of general hospitals are used for carrying out differential privacy protection on own general disease data and sending the general disease data subjected to differential privacy protection to the research institution;
the research institution is used for establishing an MP-ARMAX model of multiple participants according to the received data;
the research institution is also used for researching the condition that the coronary heart disease is influenced by other diseases according to the MP-ARMAX model.
Optionally, the coronary heart disease special hospital is specifically configured to add a first privacy protection noise to its own coronary heart disease data, where the first privacy protection noise follows a variance of 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Representing P 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protectionEpsilon is the privacy preserving budget.
Optionally, the plurality of general hospitals is specifically used for adding second privacy protection noise to own general disease data, wherein the second privacy protection noise is obtained by following a variance of 2b 2 (k) The plurality of general hospitals selecting the noise parameter b of the second privacy-preserving noise i Satisfies the following conditionsC i,2 Representation of Pi Is a sensitive degree of (a).
Optionally, the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the coronary heart disease special department hospital,historical general disease data for the first of said general hospitals,/i->Historical general disease data for a second of the general hospitals,Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, a correlation parameter b is calculated which indicates that the coronary heart disease is affected by other diseases i,k And a correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j ,k=1,...,q i ,j=1,...,p,q i And p is a known system order.
Optionally, the research institution is specifically used for calculatingDeriving the correlation parameter b i,k And a j The method specifically comprises the following steps:
At the time instant k +1,
wherein θ is k Is an estimated value of θ at time k, θ k+1 Is an estimate of theta at time k +1,representing a matrixT is the transpose of the corresponding matrix, α, < > for any given small positive value> I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at 0 moment, and is obtained by continuous iterative calculation until K approaches infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
The technical scheme provided by the invention has the beneficial effects that at least:
the research institution establishes MP-ARMAX models of multiple participants according to the received coronary heart disease data sent by the coronary heart disease special hospital and the received general disease data sent by the general hospitals, and calculates and obtains the relevant parameters of the coronary heart disease affected by other diseases according to the MP-ARMAX models, so that the condition of the coronary heart disease affected by other diseases can be accurately researched. In addition, the differential privacy protection is respectively carried out on the coronary heart disease data sent by the coronary heart disease special hospital and the general disease data sent by the plurality of general hospitals, so that the coronary heart disease data sent by the coronary heart disease special hospital and the general disease data sent by the plurality of general hospitals can be protected from being attacked by an attacker to reveal the privacy of a patient.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a coronary heart disease research method based on least square estimation and privacy protection provided by an embodiment of the invention;
fig. 2 is a block diagram of a coronary heart disease research system based on least square estimation and privacy protection according to an embodiment of the present invention.
Detailed Description
As shown in fig. 1, the embodiment of the invention provides a coronary heart disease research method based on least square estimation and privacy protection, which comprises the following steps:
s1, carrying out differential privacy protection on the coronary heart disease data by a coronary heart disease special hospital, and transmitting the coronary heart disease data subjected to differential privacy protection to a research institution;
s2, the plurality of general hospitals conduct differential privacy protection on own general disease data, and the general disease data subjected to differential privacy protection are sent to the research institution;
s3, the research institution establishes an MP-ARMAX model of multiple participants according to the received data;
and S4, calculating and obtaining relevant parameters of the coronary heart disease affected by other diseases by the research institution according to the MP-ARMAX model.
The following describes in detail a coronary heart disease research method based on least squares estimation and privacy protection according to an embodiment of the present invention with reference to fig. 2.
As shown in fig. 2, a coronary heart disease research system based on least squares estimation and privacy protection according to an embodiment of the present invention includes: coronary disease specialty hospitals, a plurality of general hospitals and research institutions.
In order to study the influence of other diseases on coronary heart disease, the research institution of the embodiment of the invention selects more specialized coronary heart disease data of the coronary heart disease special hospital and more comprehensive general disease data of a plurality of general hospitals as data sources for study. The embodiment of the invention provides a coronary heart disease research method based on least square estimation and privacy protection, which comprises the following steps:
s1, carrying out differential privacy protection on the coronary heart disease data by a coronary heart disease special hospital, and transmitting the coronary heart disease data subjected to differential privacy protection to a research institution;
the coronary heart disease data relates to privacy of a plurality of patients, and in order to protect the privacy from attack of an attacker and reveal the privacy of the patients, the coronary heart disease special hospital of the embodiment of the invention carries out differential privacy protection on the coronary heart disease data.
Optionally, the coronary heart disease special hospital in S1 performs differential privacy protection on its own coronary heart disease data, which specifically includes:
the coronary heart disease special hospital adds first privacy protection noise to the coronary heart disease data of the coronary heart disease special hospital, wherein the first privacy protection noise is obtained by following that the variance is 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Representing P 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protection and epsilon is the privacy protection budget.
S2, the plurality of general hospitals conduct differential privacy protection on own general disease data, and the general disease data subjected to differential privacy protection are sent to the research institution;
the general disease data also relates to the privacy of a plurality of patients, and in order to protect the privacy from attack of an attacker and reveal the privacy of the patients, the general hospitals of the embodiment of the invention carry out differential privacy protection on the general disease data.
Optionally, the multiple general hospitals of S2 perform differential privacy protection on own general disease data, which specifically includes:
the plurality of general hospitals add a second privacy-preserving noise to the own general disease data, the second privacy-preserving noise being a noise with a following variance of 2b 2 (k) The plurality of general hospitals selecting the noise parameter b of the second privacy-preserving noise i Satisfies the following conditionsC i,2 Representing P i Is a sensitive degree of (a).
S3, the research institution establishes an MP-ARMAX model of multiple participants according to the received data;
optionally, the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the coronary heart disease special department hospital,historical general disease data for the first of said general hospitals,/i->Historical general disease data for a second of the general hospitals,Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, the correlation parameter b of the coronary heart disease affected by other diseases can be calculated i,k And a correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j ,k=1,...,q i ,j=1,...,p,q i And p is a known system order.
As can be seen from equation (1), the MP-amax model of the multi-participant of the embodiment of the present invention:
1) The running average term input of the multi-participant MP-ARMAX model is changed from a single participant to multiple participants, as compared to the traditional autoregressive Average Running Model (ARMA);
2) Regression term effects are often ignored in correlation regression analysis for traditional medical fields, where regression term a is considered 1 y k +a 2 y k-1 +···+a p y k-p+1 That is, not only the direct relation between coronary heart disease data and general disease data is considered, but also the coronary heart disease of patients is affected by the history coronary heart disease in the process of other diseasesThe effects of the situation are all related to each other.
3) The specific model style refers to formula (1) and comprises regression terms, a moving average term input by multiple participants and system noise.
And S4, calculating and obtaining relevant parameters of the coronary heart disease affected by other diseases by the research institution according to the MP-ARMAX model.
Optionally, calculating the correlation parameter b i,k And a j The method specifically comprises the following steps:
At the time instant k +1,
wherein θ is k Is an estimated value of θ at time k, θ k+1 Is theta at the momentAn estimated value of k +1,representing a matrixT is the transpose of the corresponding matrix, α, < > for any given small positive value> I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at 0 moment, and is obtained by continuous iterative calculation until K approaches infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
θ 0 Is the initial value of theta at 0 moment, which can be any preset value, and then theta is obtained through continuous iterative calculation according to formulas (2), (3) and (4) based on least square estimation and differential privacy protection k Approximation values at different moments, wherein the approximation values are more and more close to the true value along with the increase of the K value until the K tends to infinity, and the obtained theta k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
In another aspect, as shown in fig. 2, there is also provided a coronary heart disease research system 200 based on least squares estimation and privacy protection, the system comprising: coronary disease specialty hospitals, a plurality of general hospitals and research institutions, wherein,
the coronary heart disease special department hospital 201 is configured to perform differential privacy protection on own coronary heart disease data, and send the coronary heart disease data subjected to differential privacy protection to the research institution;
the plurality of general hospitals 202 are configured to perform differential privacy protection on own general disease data, and send the general disease data subjected to differential privacy protection to the research institution;
the research institution 203 is configured to establish an MP-amax model of multiple participants according to the received data;
the research institution is also used for researching the condition that the coronary heart disease is influenced by other diseases according to the MP-ARMAX model.
Optionally, the coronary heart disease special hospital is specifically configured to add a first privacy protection noise to its own coronary heart disease data, where the first privacy protection noise follows a variance of 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Representing P 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protection and epsilon is the privacy protection budget.
Optionally, the plurality of general hospitals is specifically used for adding second privacy protection noise to own general disease data, wherein the second privacy protection noise is obtained by following a variance of 2b 2 (k) The plurality of general hospitals selecting the noise parameter b of the second privacy-preserving noise i Satisfies the following conditionsC i,2 Representing P i Is a sensitive degree of (a).
Optionally, the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the coronary heart disease special department hospital,historical general disease data for the first of said general hospitals,/i->Historical general disease data for a second of the general hospitals,Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, a correlation parameter b is calculated which indicates that the coronary heart disease is affected by other diseases i,k And a correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j ,k=1,...,q i ,j=1,...,p,q i And p is a known system order.
Optionally, the research institution is specifically configured to calculate the correlation parameter b i,k And a j The method specifically comprises the following steps:
At the time instant k +1,
wherein θ is k Is an estimated value of θ at time k, θ k+1 Is an estimate of theta at time k +1,representing a matrixT is the transpose of the corresponding matrix, α, < > for any given small positive value> I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at 0 moment, and is obtained by continuous iterative calculation until K approaches infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (8)
1. A coronary heart disease research method based on least squares estimation and privacy protection, the method comprising:
s1, carrying out differential privacy protection on the coronary heart disease data by a coronary heart disease special hospital, and transmitting the coronary heart disease data subjected to differential privacy protection to a research institution;
s2, the plurality of general hospitals conduct differential privacy protection on own general disease data, and the general disease data subjected to differential privacy protection are sent to the research institution;
s3, the research institution establishes an MP-ARMAX model of multiple participants according to the received data;
s4, the research institution calculates and obtains relevant parameters of the coronary heart disease affected by other diseases according to the MP-ARMAX model;
the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the special department of coronary heart disease hospital, u 1,k 、u 1,k-1 、...u 1,k+1-q1 For historical general disease data of the first said general hospital, u 2,k 、u 2,k-1 、...u 2,k+1-q2 For historical general disease data of a second of said general hospitals, u m,k 、u m,k-1 、...u m,k+1-qm Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, calculates the correlation parameter b of the coronary heart disease affected by other diseases i,k I=1,..m, m is the number of multiple general hospitals, k=1 and, q i ,q i Is a known system order and the correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j J=1,..p, p is a known system order.
2. The method according to claim 1, wherein the coronary heart disease specific hospital of S1 performs differential privacy protection on its own coronary heart disease data, specifically comprising:
the coronary heart disease special department hospital is used for treating the coronary heart disease of the patientThe data adds a first privacy preserving noise, the first privacy preserving noise being a data with a following variance of 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Coronary heart disease data P representing the coronary heart disease special department hospital 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protection and epsilon is the privacy protection budget.
3. The method according to claim 2, wherein the S2 plurality of general hospitals differential privacy preserving their own general disease data, specifically comprising:
the plurality of general hospitals add a second privacy-preserving noise to the own general disease data, the second privacy-preserving noise being a noise with a following variance of 2b 2 (k) The plurality of general hospitals selecting the noise parameter b of the second privacy-preserving noise i Satisfies the following conditionsC i,2 Data P representing the plurality of general hospital general diseases i Is a sensitive degree of (a).
4. A method according to claim 3, wherein the correlation parameter b is calculated i,k And a j The method specifically comprises the following steps:
θ=[a 1 ,...,a p ,b 1,1 ,...,b 1,q1 ,...,b n,1 ,...,b m,qm ] T
At the time instant k +1,
wherein y is k Is the true value of coronary heart disease data of the special department of coronary heart disease hospital, eta k Is the first privacy preserving noise added to the system,is coronary heart disease data of coronary heart disease special department hospital added with first privacy protection noise, u i,k Is the true value of the general disease data of a plurality of general hospitals, and xi i,k Is added second privacy preserving noise, +.>Is the estimated value of theta at the time k, theta, which is the data of the plurality of general hospital general diseases added with the second privacy preserving noise k+1 Is an estimate of θ at time k+1, +.>Representing a matrixT is the transpose of the corresponding matrix, α, < > for any given small positive value>I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at the moment 0, and is obtained by continuous iterative calculation until k tends to infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
5. A coronary heart disease research system based on least squares estimation and privacy preservation, the system comprising: coronary disease specialty hospitals, a plurality of general hospitals and research institutions, wherein,
the special coronary heart disease hospital is used for carrying out differential privacy protection on the coronary heart disease data of the special coronary heart disease hospital and sending the coronary heart disease data subjected to differential privacy protection to the research institution;
the plurality of general hospitals are used for carrying out differential privacy protection on own general disease data and sending the general disease data subjected to differential privacy protection to the research institution;
the research institution is used for establishing an MP-ARMAX model of multiple participants according to the received data;
the research institution is also used for researching the condition that the coronary heart disease is influenced by other diseases according to the MP-ARMAX model;
the multi-participant MP-amax model comprises:
wherein y is k+1 Is coronary heart disease data at time k+1 of the coronary heart disease special hospital, y k 、y k-1 、...y k-p+1 Is the historical coronary heart disease data of the special department of coronary heart disease hospital, u 1,k 、u 1,k-1 、...u 1,k+1-q1 For historical general disease data of the first said general hospital, u 2,k 、u 2,k-1 、...u 2,k+1-q2 For historical general disease data of a second of said general hospitals, u m,k 、u m,k-1 、...u m,k+1-qm Historical general disease data for the mth general hospital, w k+1 Is systematic noise, and based on the above data, calculates the correlation parameter b of the coronary heart disease affected by other diseases i,k I=1,..m, m is the number of multiple general hospitals, k=1 and, q i ,q i Is a known system order and the correlation parameter a of the coronary heart disease affected by the historical coronary heart disease data j J=1,..p, p is a known system order.
6. The system according to claim 5, wherein the coronary heart disease specialty hospital is specifically configured to add a first privacy preserving noise to its own coronary heart disease data, the first privacy preserving noise being compliant with a variance of 2b 2 (k) Is used for selecting the noise parameter b of the first privacy protection noise by the special hospital for coronary heart disease 0 Satisfies the following conditionsC 1 Coronary heart disease data P representing the coronary heart disease special department hospital 0 Is a degree of sensitivity of (2); delta represents the degree of privacy protection and epsilon is the privacy protection budget.
7. The system according to claim 6, wherein the plurality of general hospitals is specifically configured to add a second privacy-preserving noise to own general disease data, the second privacy-preserving noise being compliant with a variance of 2b 2 (k) Is a Laplace distribution of the plurality of general hospitalsNoise parameter b of the second privacy preserving noise i Satisfies the following conditionsC i,2 Data P representing the plurality of general hospital general diseases i Is a sensitive degree of (a).
8. The system according to claim 7, wherein the research institution is specifically configured to calculate the correlation parameter b i,k And a j The method specifically comprises the following steps:
θ=[a 1 ,...,a p ,b 1,1 ,...,b 1,q1 ,...,b n,1 ,...,b m,qm ] T
At the time instant k +1,
wherein y is k Is the true value of coronary heart disease data of the special department of coronary heart disease hospital, eta k Is the first privacy preserving noise added to the system,is coronary heart disease data of coronary heart disease special department hospital added with first privacy protection noise, u i,k Is the true value of the general disease data of a plurality of general hospitals, and xi i,k Is added second privacy preserving noise, +.>Is the estimated value of theta at the time k, theta, which is the data of the plurality of general hospital general diseases added with the second privacy preserving noise k+1 Is an estimate of θ at time k+1, +.>Representing a matrixT is the transpose of the corresponding matrix, α, < > for any given small positive value>I is an identity matrix, the dimension is the same as the dimension of the unknown parameter theta 0 Is the initial value of theta at the moment 0, and is obtained by continuous iterative calculation until k tends to infinity k The final approximation of the value θ, and thereby the correlation parameter b i,k And a j Is a similar value to (a) in the above.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211620092.4A CN115862894B (en) | 2022-12-15 | 2022-12-15 | Coronary heart disease research method and system based on least square estimation and privacy protection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211620092.4A CN115862894B (en) | 2022-12-15 | 2022-12-15 | Coronary heart disease research method and system based on least square estimation and privacy protection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115862894A CN115862894A (en) | 2023-03-28 |
CN115862894B true CN115862894B (en) | 2023-06-30 |
Family
ID=85673456
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211620092.4A Active CN115862894B (en) | 2022-12-15 | 2022-12-15 | Coronary heart disease research method and system based on least square estimation and privacy protection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115862894B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103246794A (en) * | 2012-02-01 | 2013-08-14 | 苏州人为峰软件科技有限公司 | Method for designing forgetting factor of least square method for blood glucose concentration prediction |
CN105608389A (en) * | 2015-10-22 | 2016-05-25 | 广西师范大学 | Differential privacy protection method of medical data dissemination |
CN106339714A (en) * | 2016-08-10 | 2017-01-18 | 上海交通大学 | Multi-layer differential privacy embedded decision tree model-based privacy risk control method |
CN114664457A (en) * | 2022-03-23 | 2022-06-24 | 河南大学 | Clinical path establishing and optimizing method meeting differential privacy constraints |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SE0400456D0 (en) * | 2004-02-26 | 2004-02-26 | Lars Liljeryd | Multiparameter metabolic monitoring, a method and device for the improvement of management and control in borderline or manifest type 2 diabetes |
CN114005541B (en) * | 2021-11-24 | 2023-07-18 | 珠海全一科技有限公司 | Dynamic dry eye early warning method and system based on artificial intelligence |
CN114417242B (en) * | 2021-12-20 | 2023-03-24 | 淮阴工学院 | Big data detection system for livestock and poultry activity information |
-
2022
- 2022-12-15 CN CN202211620092.4A patent/CN115862894B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103246794A (en) * | 2012-02-01 | 2013-08-14 | 苏州人为峰软件科技有限公司 | Method for designing forgetting factor of least square method for blood glucose concentration prediction |
CN105608389A (en) * | 2015-10-22 | 2016-05-25 | 广西师范大学 | Differential privacy protection method of medical data dissemination |
CN106339714A (en) * | 2016-08-10 | 2017-01-18 | 上海交通大学 | Multi-layer differential privacy embedded decision tree model-based privacy risk control method |
CN114664457A (en) * | 2022-03-23 | 2022-06-24 | 河南大学 | Clinical path establishing and optimizing method meeting differential privacy constraints |
Non-Patent Citations (2)
Title |
---|
一种基于支持向量回归和动态特征选择的梨黑星病预测方法;辜丽川;钟金琴;张友华;李绍稳;;计算机科学(07);全文 * |
人工胰脏中数据驱动个体血糖代谢模型的辨识;李鹏;祝楠楠;郁磊;王弼陡;;仪器仪表学报(03);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN115862894A (en) | 2023-03-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Brady | The role of obesity in the development of left ventricular hypertrophy among children and adolescents | |
Palmieri et al. | Aortic root dilatation at sinuses of valsalva and aortic regurgitation in hypertensive and normotensive subjects: The Hypertension Genetic Epidemiology Network Study | |
Blankstein et al. | Female gender is an independent predictor of operative mortality after coronary artery bypass graft surgery: contemporary analysis of 31 Midwestern hospitals | |
Johansson et al. | The no-touch vein graft for coronary artery bypass surgery preserves the left ventricular ejection fraction at 16 years postoperatively: long-term data from a longitudinal randomised trial | |
CN115862894B (en) | Coronary heart disease research method and system based on least square estimation and privacy protection | |
Trimarco et al. | Therapeutic concordance improves blood pressure control in patients with resistant hypertension | |
Édes et al. | Acute, total occlusion of the left main stem: coronary intervention options, outcomes, and recommendations | |
Wyse | Lenient versus strict rate control in atrial fibrillation: Some devils in the details | |
Lo et al. | Safety and effectiveness of a next-generation contact force catheter: results of the TactiSense trial | |
Krishnan et al. | Intra-atrial right coronary artery and its ablation implications | |
Netuka et al. | A trial of complete withdrawal of anticoagulation therapy in the heartmate 3 pump | |
McCarthy | Outcomes after coronary artery bypass: getting better all the time | |
Gradman | SPRINT: to whom do the results apply? | |
Nozari et al. | Outcome of coronary artery bypass grafting in patients without major risk factors and patients with at least one major risk factor for coronary artery disease | |
Kumar et al. | TCT CONNECT-401 Stent Underexpansion Is Associated With High Wall Shear Stress: A Biomechanical Analysis of the Shear Stent Study | |
Cecconi et al. | Anterior ST-segment elevation secondary to right coronary occlusion: The sheep in wolf's clothing | |
de Jaegere et al. | New conduction abnormalities after transcatheter aortic valve replacement: an innocent bystander or a serious adverse event indeed? | |
Symalla et al. | A Long-Term Counterpulsation Heart Assist System Used as a Bridge to Decision in Advanced Congestive Heart Failure | |
Jeremias et al. | Assessing post-percutaneous coronary intervention physiology: is hyperemia necessary? | |
Cortese et al. | TCTAP A-068 Long-term Follow-up from the NANOLUTE Registry on the Performance of Sirolimus Coated Balloon for the Treatment of In-Stent Restenosis | |
Powell et al. | Bedside VA-ECMO Cannulation for a Patient with CTEPH and RV Failure | |
Sobitniak et al. | Comparative characteristics of the effectiveness of surgical treatment of various forms of atrial fibrillation | |
Yokoi et al. | Clinical Characteristics and Short-Term Outcomes in Patients With Cardiogenic Shock Undergoing Mechanical Circulatory Support Escalation From Intra-Aortic Balloon Pump to Impella: From the J-PVAD Registry | |
Abu-Hilal et al. | Interrupted inferior vena cava syndrome discovered incidentally after minimally invasive mitral valve repair in a 31-year-old female patient: A case report | |
Tong et al. | Extracorporeal Left Ventricular Assist Device as a Bridge to Surgery for Ventricular Septal Rupture After Acute Myocardial Infarction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |