[go: up one dir, main page]

CN114300106A - Medical resource allocation method and device and electronic equipment - Google Patents

Medical resource allocation method and device and electronic equipment Download PDF

Info

Publication number
CN114300106A
CN114300106A CN202111643554.XA CN202111643554A CN114300106A CN 114300106 A CN114300106 A CN 114300106A CN 202111643554 A CN202111643554 A CN 202111643554A CN 114300106 A CN114300106 A CN 114300106A
Authority
CN
China
Prior art keywords
user
information
resource allocation
medical resource
medical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111643554.XA
Other languages
Chinese (zh)
Inventor
申文武
龚姝
朱婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
West China Hospital of Sichuan University
Original Assignee
West China Hospital of Sichuan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by West China Hospital of Sichuan University filed Critical West China Hospital of Sichuan University
Priority to CN202111643554.XA priority Critical patent/CN114300106A/en
Publication of CN114300106A publication Critical patent/CN114300106A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Landscapes

  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

本发明实施例提供的医疗资源分配方法、装置及电子设备，涉及医疗资源分配技术领域。该医疗资源分配方法包括：获取用户的用户信息，其中，所述用户信息包括用户的描述信息、登记信息及疾病信息；根据所述用户信息得到所述用户的特征数据；根据预先构建的目标计算模型和所述特征数据计算得到资源分配参数；根据所述资源分配参数和当前的医疗资源信息对所述用户进行医疗资源分配处理。通过上述设置，可以提高医院的运营效率。

Figure 202111643554

The medical resource allocation method, device and electronic device provided by the embodiments of the present invention relate to the technical field of medical resource allocation. The medical resource allocation method includes: acquiring user information of a user, wherein the user information includes description information, registration information and disease information of the user; obtaining characteristic data of the user according to the user information; calculating according to a pre-built target The model and the feature data are calculated to obtain resource allocation parameters; medical resource allocation processing is performed on the user according to the resource allocation parameters and current medical resource information. Through the above settings, the operational efficiency of the hospital can be improved.

Figure 202111643554

Description

Medical resource allocation method and device and electronic equipment
This application is a divisional application, parent application number: 2019104167815, case name: medical resource management method, device and electronic equipment, application date of the mother application: year 2019, month 05 and day 20.
Technical Field
The invention relates to the technical field of medical resource allocation, in particular to a medical resource allocation method, a medical resource allocation device and electronic equipment.
Background
With the popularization of medical information systems such as hospital information systems in hospitals, medical information has entered the paperless era. The reasonable allocation and optimization of medical resources are the focus of long-term attention in medical reformation in China. In the prior art, medical resources cannot be reasonably utilized, so that the problem of resource waste exists, and the problem of low operating efficiency of hospitals exists.
Disclosure of Invention
In view of the above, an object of the present invention is to provide a method, an apparatus and an electronic device for allocating medical resources, so as to solve the problems in the prior art.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
a medical resource allocation method, comprising:
acquiring user information of a user, wherein the user information comprises description information, registration information and disease information of the user;
obtaining characteristic data of the user according to the user information;
calculating according to a pre-constructed target calculation model and the characteristic data to obtain a resource allocation parameter;
and performing medical resource allocation processing on the user according to the resource allocation parameters and the current medical resource information.
In a preferred option of the embodiment of the present invention, the step of calculating to obtain the resource allocation parameter according to the pre-constructed target calculation model and the feature data includes:
calculating the classification performance of a plurality of preset classification algorithms according to historical user information, and selecting a target classification algorithm according to the classification performance;
calculating a weight coefficient of each predetermined evaluation index according to historical user information and the target classification algorithm, wherein the description information, the registration information and the disease information respectively have at least one evaluation index;
calculating according to a pre-constructed target calculation model, the weight coefficient and the characteristic data to obtain a resource distribution parameter;
wherein the target calculation model is:
Figure BDA0003444427350000021
wherein P is the resource allocation parameter, n is the number of the evaluation indexes, aiCharacteristic data corresponding to the ith evaluation index, ciThe weight coefficient of the ith evaluation index.
In a preferred option of the embodiment of the present invention, the step of performing medical resource allocation processing on the user according to the resource allocation parameter and the current medical resource information includes:
judging whether medical resources meeting preset conditions exist according to the disease information of the user and the current medical resource information;
when the medical resource meeting the preset condition exists, distributing the medical resource to the user;
and when no medical resource meeting the preset condition exists, adjusting the medical resource reservation queuing list according to the user information of the user.
In a preferred option of the embodiment of the present invention, the medical resource allocation method further includes:
calculating the daily increase of the medical resources meeting the preset conditions according to a pre-constructed prediction model;
and performing pre-allocation processing on the medical resources for the users in the medical resource reservation queuing list according to the daily increment.
In a preferred option of the embodiment of the present invention, the medical resource allocation method further includes a step of constructing the prediction model in advance, and the step includes:
obtaining relative error data through an MSARIMA model and historical daily increment of medical resources meeting preset conditions;
a predictive model is constructed from the weighted markov chains and the relative error data.
In a preferred option of the embodiment of the present invention, the step of obtaining the relative error data through the msnarima model and the historical daily increment of the medical resource meeting the preset condition includes:
processing the historical daily increment of the medical resources meeting the preset conditions to obtain training data;
fitting the training data through an MSARIMA model to obtain fitting data;
and obtaining relative error data according to the training data and the fitting data.
In a preferred alternative of the embodiment of the present invention, the step of constructing a prediction model from the weighted markov chain and the relative error data includes:
processing the relative error data by a mean-square error grading method to obtain state sequence data;
calculating a one-step state transition probability matrix according to the state sequence data;
when the relative error data is in a Markov process, calculating a weight coefficient of a transition probability matrix of a corresponding time step of a Markov chain according to an autocorrelation coefficient of the relative error data in each time step;
and constructing a prediction model according to the weight coefficient and the one-step state transition probability matrix.
In a preferred option of the embodiment of the present invention, the step of performing medical resource allocation processing on the user according to the resource allocation parameter and the current medical resource information further includes:
judging whether the resource allocation parameters are in a preset interval or not;
when the resource allocation parameters are in a preset interval, allocating medical resources for the user according to the current medical resource information, and updating the current medical resource information;
and when the resource allocation parameter is not in the preset interval, adding the user into the medical resource reservation list according to the resource allocation parameter.
An embodiment of the present invention further provides a medical resource allocation apparatus, including:
the system comprises a user information acquisition module, a registration module and a disease information acquisition module, wherein the user information acquisition module is used for acquiring user information of a user, and the user information comprises description information, registration information and disease information of the user;
the characteristic data acquisition module is used for acquiring the characteristic data of the user according to the user information;
the resource allocation parameter calculation module is used for calculating to obtain resource allocation parameters according to a pre-constructed target calculation model and the characteristic data;
and the medical resource allocation module is used for allocating medical resources to the user according to the resource allocation parameters and the current medical resource information.
An embodiment of the present invention further provides an electronic device, which includes a memory and a processor, where the processor is configured to execute an executable computer program stored in the memory, so as to implement the medical resource allocation method described above.
According to the medical resource allocation method, the medical resource allocation device and the electronic equipment, the user information of the user is calculated according to the pre-constructed target calculation model to obtain the resource allocation parameters, and the medical resource allocation processing is performed on the user according to the resource allocation parameters, so that the operation efficiency of a hospital is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a block diagram of an electronic device according to an embodiment of the present invention.
Fig. 2 is a flowchart illustrating a medical resource allocation method according to an embodiment of the present invention.
Fig. 3 is a flowchart of step S300 according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of classification performance of five classification models provided in the embodiment of the present invention.
Icon: 10-an electronic device; 12-a memory; 14-a processor; 100-medical resource allocation device.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
As shown in fig. 1, an embodiment of the invention provides an electronic device 10. The electronic device 10 may include a memory 12, a processor 14, and a medical asset management device 100. The memory 12 and the processor 14 are electrically connected, directly or indirectly, to enable the transfer or interaction of data. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The medical resource management apparatus 100 includes at least one software function module which can be stored in the memory 12 in the form of software or firmware (firmware). The processor 14 is configured to execute executable computer programs stored in the memory 12, such as software functional modules and computer programs included in the medical asset management device 100, so as to implement a medical asset management method.
The Memory 12 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 14 may be an integrated circuit chip having signal processing capabilities. The Processor 14 may be a general-purpose Processor including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like.
It will be appreciated that the configuration shown in FIG. 1 is merely illustrative and that the electronic device 10 may include more or fewer components than shown in FIG. 1 or may have a different configuration than shown in FIG. 1.
With reference to fig. 2, an embodiment of the present invention further provides a medical resource allocation method applicable to the electronic device 10. Wherein the method steps defined by the flow related to the medical resource allocation method may be implemented by the electronic device 10. The specific process shown in fig. 2 will be described in detail below.
Step S100, user information of the user is obtained.
The user information may include description information, registration information, and disease information of the user. In particular, the descriptive information may include, but is not limited to, the user's gender, age, location of ownership, type of medical insurance, and clinician. The registration information may include, but is not limited to, season of admission, month of admission, date of admission, week of admission, time of admission, week of registration, time of registration, interval of registration, waiting time, remark information, and appointment care unit information. The disease information may include, but is not limited to, disease type and number of diseases.
And step S200, obtaining the characteristic data of the user according to the user information.
In detail, the user information of the user is subjected to data preprocessing to obtain the characteristic data of the user. For example, gender is divided into male and female, which may be assigned 1 and 2, respectively. The attribution areas are divided into provinces and provinces, and are respectively assigned with 1 and 2. Medical insurance types can be divided into 3 categories: medical insurance is provided; no medical insurance and other types, assigned values of 1, 2 and 3, respectively. The registration time and admission time can be divided into two periods: morning (00:00 am-12: 00am) and afternoon (12:01 am-23: 59pm), assigned values of 1 and 2, respectively. For disease types, physicians in the nephrology department and admission center managers can classify admitted patients for nephrology department into five subgroups based on outpatient diagnosis: vascular access (type 1), renal biopsy (type 2), peritoneal dialysis (type 3), post-renal transplant (type 4) and other diseases (type 5), assigned values of 1-5, respectively. The number of diseases is obtained by statistical calculation, and is divided into 4 types: contains only one disease, assigned a value of 1; two diseases are included, the value is 2, and the rest is similar; three diseases are included, with a value of 3, four and more diseases, with a value of 4.
And step S300, calculating according to a pre-constructed target calculation model and the characteristic data to obtain a resource allocation parameter.
In detail, the resource allocation parameter corresponds to a priority of the user.
And step S400, performing medical resource allocation processing on the user according to the resource allocation parameters and the current medical resource information.
The specific type of the medical resource is not limited, and the medical resource can be set according to the actual application requirement. For example, in this embodiment, the medical resource may be a bed.
Through the setting, the user information of the user is calculated according to the pre-constructed target calculation model to obtain the resource allocation parameter, and the medical resource allocation processing is carried out on the user according to the resource allocation parameter, so that the operation efficiency of the hospital is improved.
With reference to fig. 3, the step S300 may include a step S310, a step S320, and a step S330.
Step S310, calculating the classification performance of a plurality of preset classification algorithms according to historical user information, and selecting a target classification algorithm according to the classification performance.
In detail, the preset plurality of classification algorithms may include a logistic regression algorithm, a random forest rf (random forest) algorithm, a GBDT algorithm, and an xgboost (extreme Gradient boosting) algorithm.
The historical user information can include historical user information of 2014, 2015 and 2016, and the historical user information of 2014 can be used as a training set, the historical user information of 2015 can be used as a verification set, and the historical user information of 2016 can be used as a test set.
The classification performance of the four classification algorithms can be expressed by the area under the ROC curve, i.e., AUC. Among them, ROC (Receiver Operating Characteristic) is a relation curve depicting detection rate and false alarm rate under various threshold settings. AUC values calculated from the historical user information for the four classification algorithms are shown below.
Figure BDA0003444427350000081
The classification performance of the XGboost algorithm is relatively good.
Further, the calculation results of the models of all classification algorithms can be averaged by using an Averaging method to obtain a fusion model Ensemble, and the formula for calculating the fusion model is as follows:
Figure BDA0003444427350000082
where n represents the number of models, WeightiWeight, P, representing model iiAUC for model i is shown.
With reference to fig. 4, the models and the fusion models of the four algorithms can be cross-validated by ten folds to verify the classification performance. In detail, the historical user information is randomly divided into ten data subsets having the same capacity, one of which is used as a validation set and nine of which are used as a training set. The verification process was repeated ten times, each subset being used only once for verification. And after the verification process is finished, calculating the average value of the ten generated results to obtain the classification performance of each model.
As can be seen from fig. 4, the classification performance of the XGBoost algorithm is relatively good, and the XGBoost algorithm may be selected as the target classification algorithm.
Step S320, calculating the weight coefficient of each predetermined evaluation index according to the historical user information and the target classification algorithm.
Wherein the description information, the registration information, and the disease information each have at least one evaluation index. For example, the disease information may include two evaluation indicators, i.e., a disease type and a disease number.
And step S330, calculating to obtain resource distribution parameters according to a pre-constructed target calculation model, the weight coefficient and the characteristic data.
Wherein the target calculation model is:
Figure BDA0003444427350000091
wherein P is the resource allocation parameter, n is the number of the evaluation indexes, aiCharacteristic data corresponding to the ith evaluation index, ciThe weight coefficient of the ith evaluation index.
Further, the step S400 specifically includes the following steps: judging whether medical resources meeting preset conditions exist according to the disease information of the user and the current medical resource information; when the medical resource meeting the preset condition exists, distributing the medical resource to the user; and when no medical resource meeting the preset condition exists, adjusting the medical resource reservation queuing list according to the user information of the user.
In detail, the preset condition may be that a department corresponding to the disease information of the user has an empty bed. If the corresponding department has an empty bed, distributing the empty bed of the department to the user; if the corresponding department does not have a vacant bed and other departments have vacant beds, the vacant beds of other departments are allocated to the user; and if no vacant bed exists in all departments, adding the user information of the user into a medical resource reservation queuing list.
Further, the medical resource allocation method further includes: calculating the daily increase of the medical resources meeting the preset conditions according to a pre-constructed prediction model; and performing pre-allocation processing on the medical resources for the users in the medical resource reservation queuing list according to the daily increment.
In detail, the daily gain specifically refers to a bed gain corresponding to a daily discharge amount. By predicting the daily increment, the pre-allocation processing of medical resources can be carried out on the users in the medical resource reservation queuing list, a basis is provided for hospital bed allocation and patient admission scheduling decisions, and the hospital operation efficiency is further improved.
Further, the medical resource allocation method further includes a step of constructing the prediction model in advance, the step including: obtaining relative error data through an MSARIMA model and historical daily increment of medical resources meeting preset conditions; a predictive model is constructed from the weighted markov chains and the relative error data.
In detail, the step of obtaining relative error data through the msnaria model and the historical daily increase of medical resources meeting the preset conditions includes: processing the historical daily increment of the medical resources meeting the preset conditions to obtain training data; fitting the training data through an MSARIMA model to obtain fitting data; and obtaining relative error data according to the training data and the fitting data.
Specifically, training data with N observations is selected from the historical daily gain, and is represented as follows:
X＝{X(1)，X(2)，……，X(N)}；
fitting the training data by an MSARIMA model to obtain an increase variation relationship, which is expressed as follows:
Figure BDA0003444427350000101
Figure BDA0003444427350000102
in detail, the relative error data is represented as follows:
Figure BDA0003444427350000103
wherein the relative error data reflects the degree of fluctuation of the fitted sequence around the training data curve and also reflects the degree of dynamic time variation of the sequence.
Further, the step of constructing a predictive model from the weighted markov chains and the relative error data comprises: processing the relative error data by a mean-square error grading method to obtain state sequence data; calculating a one-step state transition probability matrix according to the state sequence data; when the relative error data is in a Markov process, calculating a weight coefficient of a transition probability matrix of a corresponding time step of a Markov chain according to an autocorrelation coefficient of the relative error data in each time step; and constructing a prediction model according to the weight coefficient and the one-step state transition probability matrix.
The mean-square-error classification divides the relative error data Y into i different states, the general form of the i-th state being Ei＝[Πi1,Πi2]. According to the central limit theorem, the sequence can be divided into five states:
Figure BDA0003444427350000111
and
Figure BDA0003444427350000112
therefore, which state each Y (k) in the sequence belongs to can be determined, and a one-step state transition probability matrix can be calculated according to the obtained state sequence data. Wherein
Figure BDA0003444427350000113
And S respectively represent a sample mean and a sample standard deviation of the relative error data Y, and the expression is as follows:
Figure BDA0003444427350000114
Figure BDA0003444427350000115
further, whether a stochastic process is a Markov process requires verification by statistical hypothesis testing. Typically, a discrete Markov chain sequence can be examined using the chi-square statistic. Suppose that
Figure BDA0003444427350000116
Representing the transfer frequency matrix (n)ij)q×qThe sum of all row elements in the jth row and the transfer frequency matrix (n)ij)q×qThe ratio of the sum of all elements. Recombination system slave state EiTransition to State EjProbability P ofijObtaining:
Figure BDA0003444427350000117
Figure BDA0003444427350000118
because the statistics obey a degree of freedom of (q-1)2Determining the significance level alpha to obtain
Figure BDA0003444427350000121
The value of (c). Defining:
Figure BDA0003444427350000122
wherein N isiTo representState EiThe total frequency of occurrence.
If it is not
Figure BDA0003444427350000123
The stochastic process is considered a markov process. If the relative error data Y is a Markov process, a weighted Markov chain model is introduced to obtain the predicted value of the verification set
Figure BDA0003444427350000124
Further, the autocorrelation coefficient and weight coefficient calculation expression is as follows:
Figure BDA0003444427350000125
Figure BDA0003444427350000126
wherein r is(m)The autocorrelation coefficient with the time step length of M is represented, and the maximum value of M is M. The vector to which the weight coefficients correspond can be represented as:
W(m)＝(w(1)，w(2)，…，w(m)，…，w(M))。
further, the predictive model may be represented as:
Figure BDA0003444427350000127
wherein,
Figure BDA0003444427350000128
indicating that the system is in state E at time kiTransition to state E at time k + m by m stepsjAnd satisfy
Figure BDA0003444427350000129
Figure BDA00034444273500001210
Indicating that the system is passing through m steps from state EiTransition to State EjFrequency of (1), NiRepresents state EiThe total frequency of occurrence.
Further, the step S400 may further include: judging whether the resource allocation parameters are in a preset interval or not; when the resource allocation parameters are in a preset interval, allocating medical resources for the user according to the current medical resource information, and updating the current medical resource information; and when the resource allocation parameter is not in the preset interval, adding the user into the medical resource reservation list according to the resource allocation parameter.
Specifically, the preset interval may be set according to experience of medical staff, and re-screening may be performed according to the emergency of the user, so as to allocate medical resources to the emergency user.
In another embodiment, the step S400 may further include: judging whether the resource allocation parameter is larger than a preset value; when the resource allocation parameter is larger than a preset value, allocating medical resources for the user according to the current medical resource information, and updating the current medical resource information; and when the resource distribution parameter is smaller than a preset value, adding the user into a medical resource reservation list according to the resource distribution parameter.
Further, an embodiment of the present invention further provides a medical resource management apparatus 100, which can be applied to the electronic device 10. The medical resource management apparatus 100 may include a user information acquisition module, a characteristic data acquisition module, a resource allocation parameter calculation module, and a medical resource allocation module.
The user information acquisition module is used for acquiring the user information of the user. In this embodiment, the user information acquiring module may be configured to perform step S100 shown in fig. 2, and reference may be made to the foregoing description of step S100 for relevant content of the user information acquiring module.
And the characteristic data acquisition module is used for acquiring the characteristic data of the user according to the user information. In this embodiment, the feature data acquiring module may be configured to execute step S200 shown in fig. 2, and reference may be made to the foregoing description of step S200 for relevant content of the feature data acquiring module.
And the resource allocation parameter calculation module is used for calculating and obtaining resource allocation parameters according to a pre-constructed target calculation model and the characteristic data. In this embodiment, the resource allocation parameter calculation module may be configured to execute step S300 shown in fig. 2, and reference may be made to the foregoing description of step S300 for relevant contents of the resource allocation parameter calculation module.
And the medical resource allocation module is used for allocating medical resources to the user according to the resource allocation parameters and the current medical resource information. In this embodiment, the medical resource allocation module may be configured to perform step S400 shown in fig. 2, and reference may be made to the foregoing description of step S400 for relevant contents of the medical resource allocation module.
In summary, according to the medical resource allocation method, the medical resource allocation device, and the electronic device 10 provided in the embodiments of the present invention, the user information of the user is calculated according to the pre-constructed target calculation model to obtain the resource allocation parameter, and the medical resource allocation processing is performed on the user according to the resource allocation parameter, so as to improve the operation efficiency of the hospital.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (9)

1. A method for medical resource allocation, the method comprising:
preprocessing user information of a user to obtain characteristic data of the user;
calculating the AUC of each preset classification algorithm according to the historical user information of the user;
according to the Averaging method, according to the formula
Figure FDA0003444427340000011
Averaging AUC of all the classification algorithms to obtain a fusion model, wherein n represents the number of the classification algorithms, WeightiRepresenting the weight, P, of the classification algorithm iiAUC representing classification algorithm i;
performing cross-folding verification on all the classification algorithms and the fusion model to verify the classification performance of each classification algorithm, and taking the classification algorithm with the optimal classification performance as a target classification algorithm;
calculating a weight coefficient of each predetermined evaluation index according to the historical user information and the target classification algorithm, wherein the user information comprises description information, registration information and disease information, and the description information, the registration information and the disease information respectively have at least one evaluation index;
determining the priority of the user according to the weight coefficient of each evaluation index, the characteristic data and a pre-constructed target calculation model, wherein the pre-constructed target calculation model is as follows:
Figure FDA0003444427340000012
wherein P is the priority of the user, n is the number of the evaluation indexes, aiCharacteristic data corresponding to the ith evaluation index, ciA weight coefficient for the ith evaluation index;
and performing medical resource allocation processing on the user according to the priority and the current medical resource information, wherein the medical resource is a bed.
2. The method of claim 1, wherein the step of performing a medical resource allocation process on the user based on the priority and current medical resource information comprises:
judging whether a department corresponding to the disease information of the user has a vacant bed;
if so, allocating the vacant bed of the department to the user;
if not, judging whether other departments have vacant beds or not;
if the free beds exist, the free beds of other departments are allocated to the user;
and if the user information does not exist, adding the user information of the user into a medical resource reservation queuing list.
3. The method of claim 1, wherein the descriptive information includes the user's gender, age, location of ownership, type of medical insurance, and clinician;
the registration information comprises admission season, admission month, admission date, admission week, admission time, registration week, registration interval time, waiting time, remark information and reservation nursing unit information;
the disease information includes a disease type and a disease number.
4. The method of claim 1, further comprising:
calculating daily increase of medical resources meeting preset conditions according to a pre-constructed prediction model, wherein the preset conditions are that vacant beds exist in departments corresponding to the disease information of the user, and the daily increase refers to bed increase corresponding to the daily discharge amount;
and performing pre-allocation processing on the medical resources for the users in the medical resource reservation queuing list according to the daily increment.
5. The method of claim 4, wherein the predictive model is constructed by:
selecting training data from historical daily increases of medical resources meeting preset conditions, wherein the training data is represented as:
X＝(X(1),X(2),……，X(N)}；
fitting the training data through an MSARIMA model to obtain an increase amount change relation, wherein the increase amount change relation is expressed as:
Figure FDA0003444427340000021
Figure FDA0003444427340000022
obtaining relative error data according to the training data and the increment variation relation, wherein the relative error data is used for representing the fluctuation degree of the fitted sequence around a training data curve and the dynamic time-varying degree of the sequence, and the relative error data is expressed as:
Figure FDA0003444427340000031
a predictive model is constructed from the weighted markov chains and the relative error data.
6. The method of claim 5, wherein the step of constructing a predictive model from the weighted markov chains and the relative error data comprises:
dividing the relative error data into a plurality of different states by mean-square-difference classification, the ith state being denoted as Ei＝[Πi1,Πi2]；
According to the central limit theorem, the sequence is divided into five states:
Figure FDA0003444427340000032
Figure FDA0003444427340000033
and
Figure FDA0003444427340000034
wherein,
Figure FDA0003444427340000035
and S respectively represent a sample mean and a sample standard deviation of the relative error data Y, and the expression is as follows:
Figure FDA0003444427340000036
Figure FDA0003444427340000037
determining the state to which each relative error data in the sequence belongs from the five states to obtain state sequence data;
calculating a one-step state transition probability matrix according to the state sequence data;
verifying whether the relative error data is a Markov process by a statistical hypothesis test;
when the relative error data is in a Markov process, calculating a weight coefficient of a transition probability matrix of a corresponding time step of a Markov chain according to an autocorrelation coefficient of the relative error data in each time step, wherein the autocorrelation coefficient is represented as:
Figure FDA0003444427340000038
the weighting coefficients are expressed as:
Figure FDA0003444427340000041
wherein r is(m)Representing step size m of time byThe maximum value of M is M, and the vector corresponding to the weight coefficient can be represented as:
W(m)＝(w(1),w(2),…，w(m)，…，w(M))；
constructing a prediction model according to the weight coefficient and the one-step state transition probability matrix, wherein the prediction model is as follows:
Figure FDA0003444427340000042
wherein,
Figure FDA0003444427340000043
indicating that the system is in state E at time kiTransition to state E at time k + m by m stepsjAnd satisfy
Figure FDA0003444427340000044
Figure FDA0003444427340000045
Indicating that the system is passing through m steps from state EiTransition to State EjFrequency of (1), NiRepresents state EiThe total frequency of occurrence.
7. The method of claim 6, wherein the step of verifying whether the relative error data is a Markov process via a statistical hypothesis test comprises:
calculating a transfer frequency matrix (n)ij)q×qThe sum of all row elements in the jth row and the transfer frequency matrix (n)ij)q×qRatio of the sum of all elements
Figure FDA0003444427340000046
Obtaining:
Figure FDA0003444427340000047
computing system slave state EiTransition to State EjProbability P ofijObtaining:
Figure FDA0003444427340000051
according to the transfer frequency matrix (n)ij)q×qThe sum of all row elements in the jth row and the transfer frequency matrix (n)ij)q×qRatio of the sum of all elements
Figure FDA0003444427340000057
And the system is in a slave state EiTransition to State EjProbability P ofijCalculating χ2The expression of (c) is:
Figure FDA0003444427340000052
wherein N isiRepresents state EiThe total frequency of occurrence;
degree of freedom (q-1) according to statistic2And significance level alpha, calculating
Figure FDA0003444427340000053
A value of (d);
if it is not
Figure FDA0003444427340000054
Considering the relative error data as a markov process;
if it is not
Figure FDA0003444427340000055
The relative error data is deemed not to be a markov process.
8. A medical resource allocation apparatus, characterized in that the apparatus comprises:
the characteristic data acquisition module is used for preprocessing user information of a user to obtain characteristic data of the user;
a resource allocation parameter calculation module to:
calculating the AUC of each preset classification algorithm according to the historical user information of the user;
according to the Averaging method, according to the formula
Figure FDA0003444427340000056
Averaging AUC of all the classification algorithms to obtain a fusion model, wherein n represents the number of the classification algorithms, WeightiRepresenting the weight, P, of the classification algorithm iiAUC representing classification algorithm i;
performing cross-folding verification on all the classification algorithms and the fusion model to verify the classification performance of each classification algorithm, and taking the classification algorithm with the optimal classification performance as a target classification algorithm;
calculating a weight coefficient of each predetermined evaluation index according to the historical user information and the target classification algorithm, wherein the user information comprises description information, registration information and disease information, and the description information, the registration information and the disease information respectively have at least one evaluation index;
determining the priority of the user according to the weight coefficient of each evaluation index, the characteristic data and a pre-constructed target calculation model, wherein the pre-constructed target calculation model is as follows:
Figure FDA0003444427340000061
wherein P is the priority of the user, n is the number of the evaluation indexes, aiCharacteristic data corresponding to the ith evaluation index, ciA weight coefficient for the ith evaluation index;
and the medical resource allocation module is used for allocating medical resources to the user according to the priority and the current medical resource information, wherein the medical resources are beds.
9. An electronic device comprising a memory and a processor, the processor being configured to execute an executable computer program stored in the memory to implement the medical resource allocation method of any one of claims 1-7.
CN202111643554.XA 2019-05-20 2019-05-20 Medical resource allocation method and device and electronic equipment Pending CN114300106A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111643554.XA CN114300106A (en) 2019-05-20 2019-05-20 Medical resource allocation method and device and electronic equipment

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910416781.5A CN110136816A (en) 2019-05-20 2019-05-20 Medical resource management method, device and electronic equipment
CN202111643554.XA CN114300106A (en) 2019-05-20 2019-05-20 Medical resource allocation method and device and electronic equipment

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201910416781.5A Division CN110136816A (en) 2019-05-20 2019-05-20 Medical resource management method, device and electronic equipment

Publications (1)

Publication Number Publication Date
CN114300106A true CN114300106A (en) 2022-04-08

Family

ID=67571337

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201910416781.5A Pending CN110136816A (en) 2019-05-20 2019-05-20 Medical resource management method, device and electronic equipment
CN202111643554.XA Pending CN114300106A (en) 2019-05-20 2019-05-20 Medical resource allocation method and device and electronic equipment

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201910416781.5A Pending CN110136816A (en) 2019-05-20 2019-05-20 Medical resource management method, device and electronic equipment

Country Status (1)

Country Link
CN (2) CN110136816A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114882986A (en) * 2022-07-12 2022-08-09 四川大学华西医院 Information distribution processing method and device and computer readable storage medium
CN117672482A (en) * 2023-12-19 2024-03-08 江苏道朴网络科技有限公司 An intelligent distribution management system for medical resources
CN118675721A (en) * 2024-08-22 2024-09-20 四川省医学科学院·四川省人民医院 Hospital bed allocation system for constructing disease difficulty level index based on probability measure

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110851725B (en) * 2019-11-15 2023-10-27 北京盛诺一家医院管理咨询有限公司 Medical resource recommendation method, system, computing device and storage medium
CN110706773A (en) * 2019-12-10 2020-01-17 四川大学华西医院 Patient admission management method and device
CN111312380A (en) * 2020-02-23 2020-06-19 河南省肿瘤医院 A hospital bed configuration monitoring system and monitoring method
CN112330039B (en) * 2020-11-12 2021-06-25 清华大学 Resource allocation method and device and storage medium
CN114582472B (en) * 2020-12-01 2024-08-27 中移(成都)信息通信科技有限公司 Information recommendation method, device, equipment and storage medium
CN114822785A (en) * 2021-01-18 2022-07-29 阿里巴巴集团控股有限公司 Medical resource allocation method, equipment and storage medium
CN113539457A (en) * 2021-07-16 2021-10-22 挂号网（杭州）科技有限公司 Medical resource recommendation method and device, electronic equipment and storage medium
CN114283932B (en) * 2022-03-03 2022-06-10 四川大学华西医院 Medical resource management method, device, electronic equipment and storage medium
CN114566269B (en) * 2022-03-09 2023-04-07 四川大学 Medical resource request method, device, electronic equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130326625A1 (en) * 2012-06-05 2013-12-05 Los Alamos National Security, Llc Integrating multiple data sources for malware classification
CN103440421A (en) * 2013-08-30 2013-12-11 上海普之康健康管理有限公司 Medical data processing method and system
US20150006433A1 (en) * 2013-03-15 2015-01-01 C4Cast.Com, Inc. Resource Allocation Based on Available Predictions
CN106897918A (en) * 2017-02-24 2017-06-27 上海易贷网金融信息服务有限公司 A kind of hybrid machine learning credit scoring model construction method
CN108039200A (en) * 2017-12-22 2018-05-15 东软集团股份有限公司 A kind of information recommendation method, device and storage medium, program product
CN108206049A (en) * 2017-12-29 2018-06-26 广东工业大学 The distribution method and system of a kind of medical resource

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130326625A1 (en) * 2012-06-05 2013-12-05 Los Alamos National Security, Llc Integrating multiple data sources for malware classification
US20150006433A1 (en) * 2013-03-15 2015-01-01 C4Cast.Com, Inc. Resource Allocation Based on Available Predictions
CN103440421A (en) * 2013-08-30 2013-12-11 上海普之康健康管理有限公司 Medical data processing method and system
CN106897918A (en) * 2017-02-24 2017-06-27 上海易贷网金融信息服务有限公司 A kind of hybrid machine learning credit scoring model construction method
CN108039200A (en) * 2017-12-22 2018-05-15 东软集团股份有限公司 A kind of information recommendation method, device and storage medium, program product
CN108206049A (en) * 2017-12-29 2018-06-26 广东工业大学 The distribution method and system of a kind of medical resource

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
HENRY ASANTE ANTWI 等: "mSARIMA and Prediction Markets in Forecasting Emergency Department Service", vol. 6, no. 1, pages 812 - 819 *
LI LUO 等: "Using machine-learning methods to support health-care professionals in making admission decisions", 《THE INTERNATIONAL JOURNAL OF HEALTH PLANNING AND MANAGEMENT》, vol. 34, no. 2, pages 1 - 11 *
TING ZHU 等: "Time-Series Approaches for Forecasting the Number of Hospital Daily Discharged Inpatients", 《IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS》, vol. 21, no. 2, pages 515 - 526 *
付亚军: "支持间歇性用户使用行为的SaaS资源管理方法研究", no. 5, pages 139 - 59 *
刘雪梅 等: "基于加权马尔可夫链和GM(1,1)模型的公共危机事件预测研究", vol. 33, no. 3, pages 33 - 37 *
白鹏飞 等: "基于多模型融合的互联网信贷个人信用评估方法", vol. 49, no. 6, pages 119 - 123 *
程望斌 等: "医院床位安排排队系统研究", 《湖南理工学院学报(自然科学版)》, vol. 23, no. 1, pages 24 - 27 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114882986A (en) * 2022-07-12 2022-08-09 四川大学华西医院 Information distribution processing method and device and computer readable storage medium
CN117672482A (en) * 2023-12-19 2024-03-08 江苏道朴网络科技有限公司 An intelligent distribution management system for medical resources
CN118675721A (en) * 2024-08-22 2024-09-20 四川省医学科学院·四川省人民医院 Hospital bed allocation system for constructing disease difficulty level index based on probability measure
CN118675721B (en) * 2024-08-22 2024-11-05 四川省医学科学院·四川省人民医院 Hospital bed allocation system for constructing disease difficulty level index based on probability measure

Also Published As

Publication number Publication date
CN110136816A (en) 2019-08-16

Similar Documents

Publication Publication Date Title
CN114300106A (en) Medical resource allocation method and device and electronic equipment
US11600390B2 (en) Machine learning clinical decision support system for risk categorization
CN104166667B (en) Analysis system and public health work support method
EP0875035B1 (en) Health data processing system
US20120150498A1 (en) Method and system for forecasting clinical pathways and resource requirements
CN119170182B (en) Full-course follow-up management system and method for obstetrics and gynecology patients
WO2015132903A1 (en) Medical data analysis system, medical data analysis method, and storage medium
WO2009048987A1 (en) Multi automated severity scoring
CN118969295B (en) Early warning system and method for high-risk pregnant women
US20170024523A1 (en) Requirement Forecast for Health Care Services
US20220319650A1 (en) Method and System for Providing Information About a State of Health of a Patient
Grieco et al. Performance assessment using mixed effects models: a case study on coronary patient care
CN112908452A (en) Event data modeling
US20200258624A1 (en) Method and system for centralized patient monitoring management
Filipe et al. Predict hourly patient discharge probability in Intensive Care Units using Data Mining
CN118039107A (en) Hospital informatization management system based on Internet and intelligent medical treatment
US10431339B1 (en) Method and system for determining relevant patient information
CN118098461A (en) Clinical patient state monitoring method and system based on artificial intelligence
Elitzur et al. Machine learning for optimal test admission in the presence of resource constraints
Ricciardi et al. Evaluation of different machine learning algorithms for predicting the length of stay in the emergency departments: a single-centre study
CN117038039A (en) Deep learning-based intelligent ward resource scheduling method and system
Harrison et al. Length of stay and imminent discharge probability distributions from multistage models: variation by diagnosis, severity of illness, and hospital
Saab et al. Comparison of Machine Learning Algorithms for Classifying Adverse-Event Related 30-ay Hospital Readmissions: Potential Implications for Patient Safety
JP2003526137A (en) Method and apparatus for automatically allocating staffing
CN117012356B (en) Operation distribution method and system based on operation research

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20220408