CN-121983255-A - Disease seed cost optimization method and device, terminal equipment and storage medium

CN121983255ACN 121983255 ACN121983255 ACN 121983255ACN-121983255-A

Abstract

The invention provides a disease cost optimization method, a device, terminal equipment and a storage medium, wherein the method comprises the steps of obtaining disease resource consumption information in the present period; the disease resource consumption information comprises department disease resource consumption information and department doctor disease resource consumption information, the disease resource consumption information in the current stage is input into a disease standard cost model constructed in advance to conduct cost measurement and calculation to obtain corresponding standard cost information, the department disease resource consumption information and the standard cost information are subjected to difference detection, disease types affiliated to the department disease resource consumption information with the difference exceeding a preset threshold are judged to be abnormal disease types, first department doctor disease resource consumption information under the abnormal disease types is obtained, difference comparison results between the first department doctor disease resource consumption information and the corresponding standard cost information are generated, and a disease type cost optimization strategy in the next stage is generated according to the difference comparison results. The invention can realize efficient and accurate disease cost optimization decision.

Inventors

WU JIARONG
ZHANG LIEYUAN
XU LILI
HUANG FANG
Hu Yuerui
ZENG YUPING
FU HAOYANG

Assignees

广东省中医院（广州中医药大学第二附属医院、广州中医药大学第二临床医学院、广东省中医药科学院）

Dates

Publication Date: 20260505
Application Date: 20251211

Claims (10)

1. A method for optimizing disease cost, comprising: Acquiring disease resource consumption information of the current period, wherein the disease resource consumption information comprises department disease resource consumption information and department doctor disease resource consumption information; inputting the current disease resource consumption information into a pre-constructed disease standard cost model for cost measurement and calculation to obtain corresponding standard cost information, wherein the disease standard cost model is obtained by training based on a historical disease resource consumption information sample set; performing difference inspection on the department disease resource consumption information and the standard expense information, judging disease types affiliated to the department disease resource consumption information with the difference exceeding a preset threshold as abnormal disease types, acquiring first department doctor disease resource consumption information under the abnormal disease types, and generating a difference comparison result between the first department doctor disease resource consumption information and the corresponding standard expense information; and generating a disease cost optimization strategy of the next stage according to the difference comparison result.
2. The disease cost optimization method of claim 1, wherein the obtaining the current disease resource consumption information specifically comprises: Acquiring medical records data of inpatients in the current period; Dividing each case of medical records of inpatients into subordinate disease types, and respectively calculating medical resource consumption information in a disease type dimension and a doctor dimension to obtain department disease type resource consumption information and department doctor disease type resource consumption information, wherein the department disease type resource consumption information represents the medical cost average value of a same disease type lower department in each case, and the department doctor disease type resource consumption information represents the medical cost average value of different doctors in the same disease type lower department.
3. The method for optimizing seed costs of claim 1, further comprising constructing the historical seed resource consumption information sample set, specifically: acquiring medical records data of inpatients in the last period, wherein the medical records data of inpatients comprise basic information of patients, disease characteristics of the patients, medical resource characteristics and cost data of the patients; Preprocessing the medical records data of the inpatients, wherein the preprocessing comprises removing the cases of data missing or abnormal and removing the cases of abnormal cost of the patients; dividing each patient case data of the inpatients into subordinate disease types, and calculating the medical cost standard deviation of each case under each disease type one by one, wherein the medical cost standard deviation is obtained based on the difference between the medical cost of each case and the medical cost average value of the subordinate disease types in each case; and eliminating the medical patient medical records data of which the medical cost standard deviation exceeds a preset standard deviation threshold value to obtain a historical disease resource consumption information sample set corresponding to each disease, wherein the medical patient medical records data contained under each disease in the historical disease resource consumption information sample set exceed a preset quantity threshold value.
4. The disease cost optimization method of claim 1, wherein the disease standard cost model is trained based on a historical disease resource consumption information sample set, the disease standard cost model being trained by: Respectively establishing characteristic data and tag data for the data content of each historical disease resource consumption information sample set; setting the number d of neurons based on the number of the characteristic data, and setting the number of hidden layers to be 2; And constructing a network topology structure, and setting the normalized feature vector as X epsilon R d （X=[x 1 ,x 2 ,…,x d ] T ). Forward propagation is employed and the output of each layer is set, specifically: The 1 st hidden layer output is H 1 =σ(W 1 X+b 1 )（W 1 ∈R n1×d as weight matrix, b 1 ∈R n1 as bias vector); the 2 nd hidden layer output is H 2 =σ(W 2 H 1 +b 2 )（W 2 ∈R n2×n1 is weight matrix, b 2 ∈ Rn2 is bias vector); Output layer prediction value: =w 3 H 2 +b 3 （W3∈ R1×n2 is the weight matrix, b 3 e R is the bias vector, To predict cost); Dividing a historical disease resource consumption information sample set into a training set and a verification set according to a preset proportion, performing iterative training on the network topological structure based on the training set, and updating parameters of the network topological structure until an optimal loss function is obtained, wherein the loss function is obtained according to a difference value between a predicted value and a true value of the tag data, and the loss function has the following expression: ; wherein N is the number of samples, yi is the real cost of the ith sample, and lambda is the regularization coefficient; And verifying the average performance of the network topology structure based on the verification set repeatedly for a plurality of times, and calculating a decision coefficient, wherein the expression of the decision coefficient is as follows: ; Wherein the method comprises the steps of In order to determine the coefficient of the coefficient, Is the real cost average value; and when the decision coefficient exceeds a preset threshold, judging that the disease standard cost model of the current disease is trained.
5. The disease cost optimization method according to claim 4, wherein the establishing feature data and tag data for the data content of each of the historical disease resource consumption information sample sets respectively comprises: The characteristic data comprise patient gender, age, medical insurance type, main diagnosis, admission route, discharge mode, treatment mode, main operation, operation grade, operation times, whether entering an intensive care unit and rescue times; The tag data includes total cost, medicine cost, inspection cost, medical service cost, and consumable cost.
6. The disease fee optimizing method according to claim 5, wherein the obtaining the information about the consumption of resources of the first medical laboratory under the abnormal disease and generating the comparison result of the difference between the information about the consumption of resources of the first medical laboratory and the corresponding standard fee information specifically comprises: Acquiring information of resource consumption of first department doctors under the abnormal disease, wherein the information of resource consumption of the first department doctors is a medical cost average value of different doctors in the department under the abnormal disease, and the medical cost average value comprises an average total cost, an average medicine cost, an average inspection cost, an average medical service cost and an average consumable cost; establishing an initial hypothesis test based on the medical expense mean value of each doctor, wherein the initial hypothesis test is that the medical expense mean value of the doctor is not different from the standard expense mean value; performing difference test on the total cost of each doctor and the total cost in the standard cost information respectively, and judging that the initial hypothesis test is not established when the difference between the total cost and the standard cost exceeds a preset threshold value and that the disease resource consumption information of the current doctor is abnormal; And generating a difference comparison result between disease resource consumption information and the standard expense information based on the average value information of the medical expense of the doctor.
7. The disease fee optimizing method of claim 6, wherein the generating a next stage disease fee optimizing strategy according to the difference comparison result specifically comprises: generating a fee control result of each doctor according to the difference comparison result, sequencing the fee control results, and selecting a preset number of doctors with the top ranking to generate a doctor fee control attention list; generating a fee control result of each abnormal disease seed according to the difference comparison result and the fee control result, sequencing the fee control results, and selecting a preset number of disease seeds with the front sequencing to generate a disease seed fee control attention list; Generating a stage type expense optimization target according to a preset expense control time period and the expense control result; Generating a cost optimization strategy of the next period according to the doctor cost control attention list, the disease seed cost control attention list and the stage cost optimization target, wherein the cost optimization strategy comprises the step of monitoring the disease seed resource consumption condition of each doctor in the doctor cost control attention list in the next period in real time based on the stage cost optimization target.
8. The disease cost optimizing device is characterized by comprising an information acquisition module, a cost measuring and calculating module, a difference comparison module and a strategy generation module; The information acquisition module is used for acquiring disease resource consumption information in the current period, wherein the disease resource consumption information comprises department disease resource consumption information and department doctor disease resource consumption information; the cost measurement module is used for inputting the current disease resource consumption information into a pre-constructed disease standard cost model for cost measurement to obtain corresponding standard cost information, wherein the disease standard cost model is obtained by training based on a historical disease resource consumption information sample set; The difference comparison module is used for carrying out difference detection on the department disease resource consumption information and the standard expense information, judging the disease belonging to the department disease resource consumption information with the difference exceeding a preset threshold as an abnormal disease, acquiring first department doctor disease resource consumption information under the abnormal disease, and generating a difference comparison result between the first department doctor disease resource consumption information and the corresponding standard expense information; and the strategy generation module is used for generating a disease seed cost optimization strategy in the next stage according to the difference comparison result.
9. A terminal device comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the disease seed cost optimization method according to any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored computer program, wherein the computer program when run controls a device in which the computer readable storage medium is located to perform the disease seed cost optimization method according to any one of claims 1 to 7.

Description

Disease seed cost optimization method and device, terminal equipment and storage medium Technical Field The invention relates to the technical field of digital medical treatment, in particular to a disease cost optimization method, a device, terminal equipment and a storage medium. Background The current medical resource consumption situation is complex. On one hand, the aging of population and the change of disease spectrum lead the medical demands to be continuously raised, the total consumption of resources is increased, and on the other hand, the resources are unevenly distributed. The traditional hospital disease fee structure comprises medicine fee, examination fee, treatment fee, nursing fee, bed fee and the like, wherein the medicine fee and the examination fee occupy a relatively high proportion. The management mode is mainly based on the statistical analysis of the fact data actually occurring in the current period, so that doctors and operation departments can know the cost condition. Whether the cost structure is reasonable and the optimization direction depends on experience. However, this manner of management has significant drawbacks. The scientificity and the precision are lacking, the cost composition and the change rule are difficult to grasp by experience, and an effective optimization scheme cannot be formulated. How to scientifically adjust the disease cost structure to realize high-efficiency intelligent decision is a core requirement for the operation fine management of the next hospital. Disclosure of Invention The invention aims to provide a disease cost optimization method, a device, terminal equipment and a storage medium, so as to solve the technical problems and realize efficient and accurate disease cost optimization decision. In order to solve the technical problems, the invention provides a disease cost optimization method, which comprises the following steps: acquiring disease seed resource consumption information in the current period, wherein the disease seed resource consumption information comprises department disease seed resource consumption information and department doctor disease seed resource consumption information; Inputting the disease resource consumption information in the current period into a pre-constructed disease standard cost model for cost measurement and calculation to obtain corresponding standard cost information, wherein the disease standard cost model is obtained by training based on a historical disease resource consumption information sample set; Performing difference inspection on the department disease resource consumption information and the standard expense information, judging the disease species affiliated by the department disease resource consumption information with the difference exceeding a preset threshold as abnormal disease species, acquiring first department doctor disease resource consumption information under the abnormal disease species, and generating a difference comparison result between the first department doctor disease resource consumption information and the corresponding standard expense information; and generating a disease cost optimization strategy of the next stage according to the difference comparison result. In the scheme, the standard cost information fitting the actual situation is obtained by constructing the disease standard cost model which is trained based on historical data in advance and carrying out cost measurement and calculation by combining the current disease resource consumption information, so that a scientific and accurate reference basis is provided for subsequent cost management. The resource consumption information of the department diseases is compared with the standard expense information by using the difference test, so that the abnormal diseases with the difference exceeding the preset threshold can be accurately found, the key problem is focused, and the problem of expense abnormality can be conveniently solved. After the abnormal disease seeds are determined, the resource consumption information of the first department doctor under the abnormal disease seeds is further acquired, the difference is compared, analysis is conducted deep into a doctor level, the specific source of the cost abnormality is clear, and the follow-up optimization strategy is more targeted. Generating a disease cost optimization strategy of the next stage according to the difference comparison result, timely adjusting cost management measures according to actual conditions, reasonably optimizing disease cost, improving resource utilization efficiency and improving medical cost control level. In one implementation, the method for obtaining the resource consumption information of the disease seeds in the present period specifically includes: Acquiring medical records data of inpatients in the current period; dividing each case of the inpatient medical records into subordinate disease types, and respectively calculati