CN105933172A - Cloud computing based disease self-diagnosis service construction system - Google Patents

Abstract

The invention discloses a cloud computing based disease self-diagnosis service construction system, which comprises a data resource collection module, a task planning module, a credible combination assessment module and a disease diagnosis model generation module, wherein the data resource collection module comprises a modeling submodule, a resource replication submodule and a resource search submodule, and the credible combination assessment module comprises an assessment submodule and an assessment optimization submodule. The cloud computing based disease self-diagnosis service construction system is provided with the data resource collection module, thereby facilitating users to use data resources through matching service description information, further increasing the coverage rate of data resource information in the network, and improving the search efficiency of medical record data resources; and the cloud computing based disease self-diagnosis service construction system is provided with the credible combination assessment module in order to efficiently realize low-cost disease self-diagnosis services, the credibility of a big data service supporting cloud service combination scheme is improved, and storage and computing resources of the cloud are utilized in a maximum benefit manner. In addition, the assessment time is saved, and the assessment speed is improved.

Description

Disease self-diagnosis service construction system based on cloud computing

Technical Field

The invention relates to the field of medical treatment, in particular to a disease self-diagnosis service construction system based on cloud computing.

Background

The technical problems in two aspects need to be solved for managing medical big data resources and constructing medical big data service: on one hand, software in the medical field is skillfully developed by different manufacturers, and a wide range of heterogeneous vertebrae exist in the aspects of underlying technologies and business processes based on different software and hardware platforms. Although these systems store abundant medical big data resources (such as electronic medical records, medical images, etc.), the data sharing among different medical user groups is very difficult due to the information isolated island phenomenon formed by the heterogeneity among the systems. How to integrate the medical big data resources in these systems is a challenging problem; on the other hand, the medical big data service is constructed by using the abundant medical smart big data resources, the support of storage resources and calculation resources which are expanded according to needs is needed, and the problems of high cost investment and maintenance are brought to the processing and analysis of big data and the construction of the big data service.

Disclosure of Invention

In order to solve the problems, the invention provides a disease self-diagnosis service construction system based on cloud computing.

The purpose of the invention is realized by adopting the following technical scheme:

disease self-diagnosis service construction system based on cloud computing comprises:

(1) the data resource collection module is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of the disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) a trusted combination evaluation module: the system comprises a task planning module, a cloud service combination scheme selection module, a storage module and an evaluation optimization module, wherein the task planning module is used for executing evaluation of the cloud service combination scheme according to task planning of big data service generated by the task planning module, selecting an optimal cloud service combination scheme and providing storage and computing resources for disease self-diagnosis service;

the evaluation sub-module specifically executes the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling a utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting a task plan G obtained by a task planning module as { G ═ G }₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization sub-module can record the utility function value of the optimal cloud service combination scheme and the optimal combination cloud service scheme, the utility function value and the optimal combination cloud service scheme are used as samples for learning, and if a new optimal cloud service combination scheme appears, the function value is directly called;

(4) and the disease diagnosis model generation module is used for establishing an electronic medical record index according to the optimal cloud service combination scheme and calculating by adopting a big data analysis method to obtain a disease diagnosis model of the disease self-diagnosis model.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module comprises a modeling submodule, a resource copying submodule and a resource searching submodule which are connected in sequence, wherein the modeling submodule is used for modeling an overlay network formed by resource nodes under a cloud environment by adopting an unstructured peer-to-peer network, the resource copying submodule is used for copying resource information among all neighbor nodes in the overlay network, and the resource searching submodule is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module adopts a data resource information active replication protocol based on neighboring nodes when the resource information is replicated:

when x is_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmThe method comprises the steps of copying, judging whether a copy message is received or not according to the number information of the copy message when any node in the peer-to-peer network receives the copy message, discarding the copy message if the copy message is received, and updating according to the resource information and the node position information of the copy message if the copy message is received for the first timeAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the operation specifically executed by the resource searching submodule is as follows:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer node set is p_j×{x_j1,x_j2,…x_jm},j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1,x_j2,…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1，x_j2，…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

The invention has the beneficial effects that:

1. a data resource collection module, a task planning module, a credible combination evaluation module and a disease diagnosis model generation module are arranged, so that the construction of a disease self-diagnosis service system is realized;

2. the modeling submodule, the resource copying submodule and the resource searching submodule which are connected in sequence are arranged, the unstructured peer-to-peer network is used as a topological organization structure of data resource nodes in a cloud environment, and the data resources are packaged in a service mode, so that a user can conveniently use the data resources by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved;

3. in order to efficiently realize low-cost disease self-diagnosis service, a credible combination evaluation module is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud are used in the most profitable way, and an evaluation optimization submodule is adopted, so that the evaluation time is saved, and the evaluation speed is improved.

Drawings

The invention is further illustrated by means of the attached drawings, but the embodiments in the drawings do not constitute any limitation to the invention, and for a person skilled in the art, other drawings can be obtained on the basis of the following drawings without inventive effort.

FIG. 1 is a schematic diagram of the connection of modules of the present invention;

FIG. 2 is a flow chart of the operation of the evaluation sub-module of the present invention.

Reference numerals:

the system comprises a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3, a disease diagnosis model generation module 4, a modeling sub-module 11, a resource replication sub-module 12, a resource search sub-module 13, an evaluation sub-module 31 and an evaluation optimization sub-module 32.

Detailed Description

The invention is further described with reference to the following examples.

Example 1

Referring to fig. 1 and 2, the cloud computing-based disease self-diagnosis service construction system of the embodiment includes:

(1) the data resource collection module 1 is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module 2 is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module 3: the system comprises a task planning module (2), an evaluation module and a data analysis module, wherein the task planning module is used for executing the evaluation of a cloud service combination scheme according to the task planning of the big data service generated by the task planning module (2), selecting an optimal cloud service combination scheme, and providing storage and calculation resources for a disease self-diagnosis service, and the system comprises an evaluation submodule (31) and an evaluation optimization submodule (32);

the evaluation sub-module 31 specifically performs the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling the utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting the task plan G obtained by the task planning module 2 as { G ═ G₁，G₂，G₃}, corresponding toConstraint of C ═ C₁，C₂，C₃Each subtask G_vCorresponding cloud service resource poolSP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1，3]，ω∈[1，m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization submodule 32 can record the utility function value of the preferred cloud service combination scheme and the optimal combined cloud service

The scheme is used as a sample for learning, and if a new optimal cloud service combination scheme appears, a function value of the scheme is directly called;

(4) a disease diagnosis model generation module 4, which is used for establishing an electronic medical record index according to the optimal cloud service combination scheme,

and are

And calculating to obtain a disease diagnosis model of the disease self-diagnosis model by adopting a big data analysis method.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module 1 comprises a modeling submodule 11, a resource replication submodule 12 and a resource search submodule 13 which are connected in sequence, wherein the modeling submodule 11 is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule 12 is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule 13 is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1，x_i2，…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module 12 adopts an active replication protocol based on data resource information between neighboring nodes when replicating resource information:

when x is_iWhen joining the overlay network, x_iAnd { x_l1，x_l2，…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmThe method comprises the steps of copying, judging whether a copy message is received or not according to the number information of the copy message when any node in the peer-to-peer network receives the copy message, discarding the copy message if the copy message is received, and updating according to the resource information and the node position information of the copy message if the copy message is received for the first timeAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the resource search sub-module 13 specifically executes the following operations:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly selected peer node set asp_j×{x_j1，x_j2，…x_jm},j∈[1，n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1，x_j2，…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1，x_j2，…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1，x_j2，…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1，x_j2，…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

In the embodiment, a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3 and a disease diagnosis model generation module 4 are arranged, so that the construction of a disease self-diagnosis service system is realized; the modeling submodule 11, the resource replication submodule 12 and the resource searching submodule 13 which are connected in sequence are arranged, an unstructured peer-to-peer network is used as a topological organization structure of a data resource node in a cloud environment, and the data resource is packaged in a service mode, so that a user can conveniently use the data resource by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved; in order to efficiently realize low-cost disease self-diagnosis service, the credibility combination evaluation module 3 is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud end are used in the most profitable way, and the evaluation optimization submodule 32 is adopted, so that the evaluation time is saved, and the evaluation speed is improved; in this embodiment, the value α is 0.25, and the value β is 0.45, so that the medical record data resource searching efficiency is improved by 3.5%.

Example 2

Referring to fig. 1 and 2, the cloud computing-based disease self-diagnosis service construction system of the embodiment includes:

(1) the data resource collection module 1 is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module 2 is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module 3: the system comprises a task planning module (2), an evaluation module and a data analysis module, wherein the task planning module is used for executing the evaluation of a cloud service combination scheme according to the task planning of the big data service generated by the task planning module (2), selecting an optimal cloud service combination scheme, and providing storage and calculation resources for a disease self-diagnosis service, and the system comprises an evaluation submodule (31) and an evaluation optimization submodule (32);

the evaluation sub-module 31 specifically performs the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling the utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting the task plan G obtained by the task planning module 2 as { G ═ G₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization submodule 32 can record the utility function value of the preferred cloud service combination scheme and the optimal combined cloud service

The scheme is used as a sample for learning, and if a new optimal cloud service combination scheme appears, a function value of the scheme is directly called;

(4) a disease diagnosis model generation module 4, which is used for establishing an electronic medical record index according to the optimal cloud service combination scheme,

and are

And calculating to obtain a disease diagnosis model of the disease self-diagnosis model by adopting a big data analysis method.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module 1 comprises a modeling submodule 11, a resource replication submodule 12 and a resource search submodule 13 which are connected in sequence, wherein the modeling submodule 11 is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule 12 is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule 13 is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module 12 adopts an active replication protocol based on data resource information between neighboring nodes when replicating resource information:

when x is_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmMaking duplication, if any node in the peer-to-peer network receives a duplication message, judging according to the number information of the duplication messageIf the copy message is received, the copy message is discarded, and if the copy message is received for the first time, the resource information and the node position information of the copy message are updatedAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the resource search sub-module 13 specifically executes the following operations:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer node set is p_j×{x_j1,x_j2,…x_jm｝,j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1,x_j2，…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

In the embodiment, a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3 and a disease diagnosis model generation module 4 are arranged, so that the construction of a disease self-diagnosis service system is realized; the modeling submodule 11, the resource replication submodule 12 and the resource searching submodule 13 which are connected in sequence are arranged, an unstructured peer-to-peer network is used as a topological organization structure of a data resource node in a cloud environment, and the data resource is packaged in a service mode, so that a user can conveniently use the data resource by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved; in order to efficiently realize low-cost disease self-diagnosis service, the credibility combination evaluation module 3 is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud end are used in the most profitable way, and the evaluation optimization submodule 32 is adopted, so that the evaluation time is saved, and the evaluation speed is improved; in this embodiment, the value α is 0.26, and the value β is 0.46, so that the medical record data resource searching efficiency is improved by 3%.

Example 3

Referring to fig. 1 and 2, the cloud computing-based disease self-diagnosis service construction system of the embodiment includes:

(1) the data resource collection module 1 is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module 2 is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module 3: the system comprises a task planning module (2), an evaluation module and a data analysis module, wherein the task planning module is used for executing the evaluation of a cloud service combination scheme according to the task planning of the big data service generated by the task planning module (2), selecting an optimal cloud service combination scheme, and providing storage and calculation resources for a disease self-diagnosis service, and the system comprises an evaluation submodule (31) and an evaluation optimization submodule (32);

the evaluation sub-module 31 specifically performs the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling the utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting the task plan G obtained by the task planning module 2 as { G ═ G₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-h represents a parameter of a utility function in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization submodule 32 can record the utility function value of the preferred cloud service combination scheme and the optimal combined cloud service

The scheme is used as a sample for learning, and if a new optimal cloud service combination scheme appears, a function value of the scheme is directly called;

(4) a disease diagnosis model generation module 4, which is used for establishing an electronic medical record index according to the optimal cloud service combination scheme,

and are

And calculating to obtain a disease diagnosis model of the disease self-diagnosis model by adopting a big data analysis method.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module 1 comprises a modeling submodule 11, a resource replication submodule 12 and a resource search submodule 13 which are connected in sequence, wherein the modeling submodule 11 is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule 12 is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule 13 is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module 12 adopts an active replication protocol based on data resource information between neighboring nodes when replicating resource information:

when x is_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmThe method comprises the steps of copying, judging whether a copy message is received or not according to the number information of the copy message when any node in the peer-to-peer network receives the copy message, discarding the copy message if the copy message is received, and updating according to the resource information and the node position information of the copy message if the copy message is received for the first timeAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the resource search sub-module 13 specifically executes the following operations:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer node set is p_j×{x_j1,x_j2,…x_jm},j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1,x_j2,…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

In the embodiment, a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3 and a disease diagnosis model generation module 4 are arranged, so that the construction of a disease self-diagnosis service system is realized; the modeling submodule 11, the resource replication submodule 12 and the resource searching submodule 13 which are connected in sequence are arranged, an unstructured peer-to-peer network is used as a topological organization structure of a data resource node in a cloud environment, and the data resource is packaged in a service mode, so that a user can conveniently use the data resource by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved; in order to efficiently realize low-cost disease self-diagnosis service, the credibility combination evaluation module 3 is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud end are used in the most profitable way, and the evaluation optimization submodule 32 is adopted, so that the evaluation time is saved, and the evaluation speed is improved; in this embodiment, the value α is 0.27, and the value β is 0.47, so that the medical record data resource searching efficiency is improved by 3.2%.

Example 4

Referring to fig. 1 and 2, the cloud computing-based disease self-diagnosis service construction system of the embodiment includes:

(1) the data resource collection module 1 is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module 2 is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module 3: the system comprises a task planning module (2), an evaluation module and a data analysis module, wherein the task planning module is used for executing the evaluation of a cloud service combination scheme according to the task planning of the big data service generated by the task planning module (2), selecting an optimal cloud service combination scheme, and providing storage and calculation resources for a disease self-diagnosis service, and the system comprises an evaluation submodule (31) and an evaluation optimization submodule (32);

the evaluation sub-module 31 specifically performs the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling the utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting the task obtained by the task planning module 2Business plan G ═ G₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization submodule 32 can record the utility function value of the preferred cloud service combination scheme and the optimal combined cloud service

The scheme is used as a sample for learning, and if a new optimal cloud service combination scheme appears, a function value of the scheme is directly called;

(4) a disease diagnosis model generation module 4, which is used for establishing an electronic medical record index according to the optimal cloud service combination scheme,

and are

And calculating to obtain a disease diagnosis model of the disease self-diagnosis model by adopting a big data analysis method.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module 1 comprises a modeling submodule 11, a resource replication submodule 12 and a resource search submodule 13 which are connected in sequence, wherein the modeling submodule 11 is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule 12 is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule 13 is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module 12 adopts an active replication protocol based on data resource information between neighboring nodes when replicating resource information:

when x is_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmThe method comprises the steps of copying, judging whether a copy message is received or not according to the number information of the copy message when any node in the peer-to-peer network receives the copy message, discarding the copy message if the copy message is received, and updating according to the resource information and the node position information of the copy message if the copy message is received for the first timeAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the resource search sub-module 13 specifically executes the following operations:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer node set is p_j×{x_j1,x_j2,…x_jm},j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1,x_j2,…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

In the embodiment, a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3 and a disease diagnosis model generation module 4 are arranged, so that the construction of a disease self-diagnosis service system is realized; the modeling submodule 11, the resource replication submodule 12 and the resource searching submodule 13 which are connected in sequence are arranged, an unstructured peer-to-peer network is used as a topological organization structure of a data resource node in a cloud environment, and the data resource is packaged in a service mode, so that a user can conveniently use the data resource by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved; in order to efficiently realize low-cost disease self-diagnosis service, the credibility combination evaluation module 3 is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud end are used in the most profitable way, and the evaluation optimization submodule 32 is adopted, so that the evaluation time is saved, and the evaluation speed is improved; in this embodiment, the value α is 0.28, and the value β is 0.49, so that the medical record data resource searching efficiency is improved by 3.6%.

Example 5

Referring to fig. 1 and 2, the cloud computing-based disease self-diagnosis service construction system of the embodiment includes:

(1) the data resource collection module 1 is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module 2 is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module 3: the system comprises a task planning module (2), an evaluation module and a data analysis module, wherein the task planning module is used for executing the evaluation of a cloud service combination scheme according to the task planning of the big data service generated by the task planning module (2), selecting an optimal cloud service combination scheme, and providing storage and calculation resources for a disease self-diagnosis service, and the system comprises an evaluation submodule (31) and an evaluation optimization submodule (32);

the evaluation sub-module 31 specifically performs the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory recording, modeling the utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting the task plan G obtained by the task planning module 2 as { G ═ G₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\γ}}\right)={\mathrm{\Σ}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\Σ}}_{v=1}^3{\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }\left(k\right)}}\×w_k\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\ω}}{R}_h\right)\×x_{v\mathrm{\ω}-h}\≤C_k,1\≤k\≤3\\ {\mathrm{\Σ}}_{\mathrm{\ω}=1}^{m_v}{\mathrm{\Σ}}_{h=1}^{L_{v\mathrm{\ω}}}{x}_{v\mathrm{\ω}-h}=1,x_{v\mathrm{\ω}-h}\∈\left\{0,1\right\}\\ {\mathrm{\Σ}}_k^d{w}_k=1,w_k\∈\[0,1\]\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωRh is under SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization submodule 32 can record the utility function value of the preferred cloud service combination scheme and the optimal combination cloud service scheme, and learn the utility function value and the optimal combination cloud service scheme as a sample, and if a new preferred cloud service combination scheme appears, directly call the function value;

(4) and the disease diagnosis model generation module 4 is used for establishing an electronic medical record index according to the optimal cloud service combination scheme and obtaining a disease diagnosis model of the disease self-diagnosis model through calculation by adopting a big data analysis method.

The electronic medical record index comprises a medical record inverted index, a medical record filtering index and a medical record detail index, wherein the medical record inverted index is used for retrieving medical records with the same disease symptoms as the user from electronic medical record big data resources according to the disease symptoms of the user, the medical record filtering index is used for filtering medical records with different ages and sexes from the user according to the sex and the age of the user, and the medical record detail index is used for retrieving detailed contents of the medical records from the electronic medical record big data resources.

The data resource collection module 1 comprises a modeling submodule 11, a resource replication submodule 12 and a resource search submodule 13 which are connected in sequence, wherein the modeling submodule 11 is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule 12 is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule 13 is used for searching and matching electronic medical record data resources meeting application requirements;

let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m is less than n;

A. the resource replication sub-module 12 adopts an active replication protocol based on data resource information between neighboring nodes when replicating resource information:

when x is_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmMaking a copy if it is equivalentWhen any node in the network receives a copy message, judging whether the copy message is received or not according to the number information of the copy message, if so, discarding the copy message, and if first receiving, updating the copy message according to the resource information and the node position information of the copy messageAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals;

B. the resource search sub-module 13 specifically executes the following operations:

set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer node set is p_j×{x_j1,x_j2,…x_jm},j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf, ifIf not 0, calculating p by using Q learning algorithm_j×{x_j1,x_j2,…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\αQ}}_{learn}+\mathrm{\β}\×I\[\frac{N_{x_{j\mathrm{\μ}}}\left(t\right)\left(T_{x_{j\mathrm{\μ}}}-{T^{\′}}_{x_{j\mathrm{\μ}}}\right)}{{T^{\′}}_{x_{j\mathrm{\μ}}}\×T_{x_{j\mathrm{\μ}}}}\]\×\frac{1+N_{x_{j\mathrm{\μ}}}\left(t\right)}{T_{x_{j\mathrm{\μ}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

In the embodiment, a data resource collection module 1, a task planning module 2, a credible combination evaluation module 3 and a disease diagnosis model generation module 4 are arranged, so that the construction of a disease self-diagnosis service system is realized; the modeling submodule 11, the resource replication submodule 12 and the resource searching submodule 13 which are connected in sequence are arranged, an unstructured peer-to-peer network is used as a topological organization structure of a data resource node in a cloud environment, and the data resource is packaged in a service mode, so that a user can conveniently use the data resource by matching service description information, the coverage rate of data resource information in the network is further increased, and the efficiency of searching medical record data resources is improved; in order to efficiently realize low-cost disease self-diagnosis service, the credibility combination evaluation module 3 is arranged, the credibility of a cloud service combination scheme supporting big data service is improved, the storage and calculation resources of a cloud end are used in the most profitable way, and the evaluation optimization submodule 32 is adopted, so that the evaluation time is saved, and the evaluation speed is improved; in this embodiment, the value α is 0.3, and the value β is 0.5, so that the efficiency of searching medical record data resources is improved by 4%.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the protection scope of the present invention, although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims (5)

1. Disease self-diagnosis service construction system based on cloud computing is characterized by comprising the following steps:

(1) the data resource collection module is used for collecting electronic medical record data of patients in hospitals, clinics and medical software applications in the cloud according to the requirements of the disease self-diagnosis service to form electronic medical record big data resources;

(2) the task planning module is used for dividing the processing process of the electronic medical record big data resource into a data storage subtask, an index calculation subtask and a data processing analysis calculation subtask, matching a cloud service resource pool meeting the requirement of each subtask to form a cloud service combination scheme so as to obtain the storage resource or the calculation resource required in the big data processing process;

(3) the credible combination evaluation module is used for executing the evaluation of the cloud service combination scheme according to the task plan of the big data service generated by the task plan module, selecting the optimal cloud service combination scheme, providing storage and calculation resources for the disease self-diagnosis service,

(4) and the disease diagnosis model generation module is used for establishing an electronic medical record index according to the optimal cloud service combination scheme and calculating by adopting a big data analysis method to obtain a disease diagnosis model of the disease self-diagnosis model.

2. The cloud computing-based disease self-diagnosis service construction system according to claim 1, wherein the data resource collection module comprises a modeling submodule, a resource replication submodule and a resource search submodule which are connected in sequence, the modeling submodule is used for modeling an overlay network formed by resource nodes in a cloud environment by adopting an unstructured peer-to-peer network, the resource replication submodule is used for replicating resource information among all neighbor nodes in the overlay network, and the resource search submodule is used for searching and matching electronic medical record data resources meeting application requirements.

3. The cloud-computing-based disease self-diagnosis service construction system according to claim 1, wherein the electronic medical record index comprises an inverted medical record index, a medical record filtering index and a medical record detail index, the inverted medical record index is used for retrieving a medical record with the same disease symptom as the user from electronic medical record big data resources according to the disease symptom of the user, the medical record filtering index is used for filtering a medical record with age and sex different from age and sex of the user according to the sex and age of the user, and the medical record detail index is used for retrieving detailed content of the medical record from the electronic medical record big data resources.

4. According to claim2, the disease self-diagnosis service construction system based on cloud computing is characterized in that the resource replication submodule adopts a data resource information active replication protocol based on neighboring nodes when replicating resource information, and specifically comprises: let x_iFor a peer node in an unstructured peer-to-peer network, { x }_i1,x_i2,…x_imIs x_iThe set of neighboring nodes of (a) is,is a local resource pool, and is a local resource pool,is a neighbor node resource information pool, i ∈ [1, n]N is the total number of nodes contained in the peer-to-peer network, m represents the number of neighbor nodes, and m<n when x_iWhen joining the overlay network, x_iAnd { x_l1,x_l2,…x_lmEstablishment of a connection, x_iFurther in accordance withCreates a copy message of the resource information and forwards the copy message to all the neighbor nodes x_lmThe method comprises the steps of copying, judging whether a copy message is received or not according to the number information of the copy message when any node in the peer-to-peer network receives the copy message, discarding the copy message if the copy message is received, and updating according to the resource information and the node position information of the copy message if the copy message is received for the first timeAccording to the life value of the copy message, the copy message is determined to be forwarded or discarded, wherein the resource information needs to be synchronized between the neighbor nodes at regular intervals; the searching and matching of the electronic medical record data resources meeting the application requirements comprises the following steps: set up and initiate the inquiry request M_jIs x_jAt x_jAccording to the probability p in the neighbor node set_jRandomly picked peer-to-peer nodesSet as p_j×{x_j1,x_j2,…x_jm},j∈[1,n]；

When peer node x_iReceive x_jTransmitted query request M_jAt the same time, checkAndwhether it contains a request M satisfying the inquiry_jIf so, creating a response message of inquiry according to the electronic medical record data information and the position information of the peer node where the electronic medical record data information is locatedAnd according to x_jThe response information is transmitted to the mobile stationIs returned to x_jThen x is added_jIs decreased by 1 if x_jIs 0, the query request M is discarded_jIf not, calculating p by using Q learning algorithm_j×{x_j1,x_j2,…x_jmQ value of each peer node in the queue, will query the request M_jForward to p_j×{x_j1,x_j2,…x_jmThe node with the largest Q value in the (Q) }, the probability p_jThe value range when the network is idle is (5, 8)]The value range when the network is congested is [0,3 ];

the calculation formula for setting the Q value is as follows:

$Q_{new}=Q_{old}+{\mathrm{\&alpha;Q}}_{learn}+\mathrm{\&beta;}\&times;I\&lsqb;\frac{N_{x_{j\mathrm{\&mu;}}}\left(t\right)(T_{x_{j\mathrm{\&mu;}}}-{T^{\&prime;}}_{x_{j\mathrm{\&mu;}}})}{{T^{\&prime;}}_{x_{j\mathrm{\&mu;}}}\&times;T_{x_{j\mathrm{\&mu;}}}}\&rsqb;\&times;\frac{1+N_{x_{j\mathrm{\&mu;}}}\left(t\right)}{T_{x_{j\mathrm{\&mu;}}}}$

wherein Q is_newRepresenting the new value of Q, Q_oldDenotes the old value of Q, Q_learnIndicating the value learned, α indicating the learning rate, β indicating the congestion factor,indicating the time t node x_jμThe number of query request messages pending in the buffer queue,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time specified for processing a query request message,represents p_j×{x_j1,x_j2,…x_jmNode x in (c)_jμThe time actually required to process a query request message; function I [ x ]]At x>The value is 1 when 0, 0 when x is less than or equal to 0, and the value range of α is [0.25,0.3 ]]β is [0.45,0.5 ]]。

5. The cloud computing-based disease self-diagnosis service construction system according to claim 1, wherein the trusted combination evaluation module includes an evaluation sub-module and an evaluation optimization sub-module; the evaluation sub-module specifically executes the following operations:

A. according to cloud service resource pool SP_vAnd corresponding quality of serviceHistory of recording, enteringModeling a utility function X of the cloud service combination scheme, initializing each parameter of the utility function in the model, and setting a task plan G ═ G obtained by a task planning module₁,G₂,G₃}, corresponding toConstraint of C ═ C₁,C₂,C₃Each subtask G_vCorresponding cloud service resource pool SP_vTotal m_vIndividual service, for cloud service resource pool SP_vEach service SP in_vωWhich comprisesThe number of the history records is L_vωFrom SP_vThe formed Gamma feasible cloud service combination scheme is CS_γ，v∈[1,3]，ω∈[1,m_v]The definition model is:

$\left\{\begin{array}{c}X\left({\mathrm{CS}}_{\mathrm{\&gamma;}}\right)={\mathrm{\&Sigma;}}_k^3\frac{Q_{O_{S\max }}\left(k\right)-{\mathrm{\&Sigma;}}_{v=1}^3{\mathrm{\&Sigma;}}_{\mathrm{\&omega;}=1}^{m_v}{\mathrm{\&Sigma;}}_{h=1}^{L_{v\mathrm{\&omega;}}}{q}_d\left({\mathrm{SP}}_{v\mathrm{\&omega;}}{R}_h\right)\&times;x_{v\mathrm{\&omega;}-h}}{Q_{O_{S\max }}\left(k\right)-Q_{O_{S\min }}\left(k\right)}\&times;w_k\\ {\mathrm{\&Sigma;}}_{\mathrm{\&omega;}=1}^{m_v}{\mathrm{\&Sigma;}}_{h=1}^{L_{v\mathrm{\&omega;}}}{q}_k\left({\mathrm{SP}}_{v\mathrm{\&omega;}}{R}_h\right)\&times;x_{v\mathrm{\&omega;}-h}\&le;C_k,1\&le;k\&le;3\\ {\mathrm{\&Sigma;}}_{\mathrm{\&omega;}=1}^{m_v}{\mathrm{\&Sigma;}}_{h=1}^{L_{v\mathrm{\&omega;}}}{x}_{v\mathrm{\&omega;}-h}\mathrm{=1,}{x}_{v\mathrm{\&omega;}-h}\&Element;\{0,1\}\\ {\mathrm{\&Sigma;}}_k^d{w}_k=1,w_k\&Element;\&lsqb;0,1\&rsqb;\end{array}\right.$

wherein,in the k-th dimensionThe maximum value of the number of the first and second,in the k-th dimensionMinimum value, SP_vωR_hIs subordinate to SP_vωOne strip ofHistory, x_vω-hParameters representing utility functions in the model;

B. sequencing the feasible cloud service combination schemes according to the utility function values in the order from small to large, selecting the first Z feasible cloud service combination schemes as the preferred cloud service combination schemes, and setting the value of Z according to the application example;

C. calculating an average value of the utility function values of each group of the optimized cloud service combination schemes;

D. selecting an optimal cloud service combination scheme with the maximum average value of the utility function values as an optimal cloud service combination scheme;

the evaluation optimization sub-module can record the utility function value of the optimal cloud service combination scheme and the optimal combination cloud service scheme, the utility function value and the optimal combination cloud service scheme are used as samples to learn, and if a new optimal cloud service combination scheme appears, the function value is directly called.