WO2020107899A1

WO2020107899A1 - Medical cost prediction method, device and equipment, and computer-readable storage medium

Info

Publication number: WO2020107899A1
Application number: PCT/CN2019/096633
Authority: WO
Inventors: 黄越; 陈明东
Original assignee: 平安医疗健康管理股份有限公司
Priority date: 2018-11-30
Filing date: 2019-07-19
Publication date: 2020-06-04
Also published as: CN109616216A

Abstract

Disclosed by the present application are a medical cost prediction method, device and equipment, and a computer-readable storage medium, the method comprising: obtaining multiple pieces of historical visit data, comparing each piece of the historical visit data to a preset identifier one by one, and filtering out multiple pieces of concurrent feature data corresponding to preset chronic disease complications; reading concurrent type information and medication type information in each piece of concurrent feature data, and classifying the each piece of concurrent feature data according to each piece of concurrent type information and medication type information so as to form a correspondence of the each piece of concurrent type information and each piece of medication type information to a medical expense in the each piece of concurrent feature data; and when diagnosis information of a hospitalized patient is received, comparing the diagnosis information to the correspondence, and predicting a target medical expense corresponding to the diagnosis information. In the present solution, a target medical cost for diagnosis information is predicted on the basis of a correspondence formed by big data in a medical institution, and the predicted target medical cost is more accurate and effective.

Description

Medical expense prediction method, device, equipment and computer readable storage medium The

This application requires the priority of the Chinese patent application submitted to the China Patent Office on November 30, 2018 with the application number 201811462249.9 and the invention titled "Medical Expense Prediction Method, Device, Equipment and Computer-readable Storage Medium" Incorporated by reference in the application.

Technical field

This application mainly relates to the technical field of medical systems, and in particular, to a medical expense prediction method, device, equipment, and computer-readable storage medium.

Background technique

At present, there are many types of complications of chronic diseases. The complications of diabetes alone are as many as 100 kinds. Combining hypertension complications, hyperlipidemia complications, coronary heart disease complications, etc. Variety. Different types of patients with chronic disease complications or patients with the same type of chronic disease complications will use different treatment and drug schemes due to differences in individual patients, resulting in different medical costs between different patients.

Chronic disease complications require long-term medical treatment. Accurately predicting the medical expenses required by patients during the treatment process is conducive to the planning of long-term treatment programs for patients to stabilize their conditions; however, there is currently no effective prediction of medical expenses for chronic disease complications The mechanism makes the prediction of medical expenses of patients with various chronic disease complications inaccurate, which leads to the inaccuracy of the long-term treatment plan planned by the patients and affects the stability of the patients.

Summary of the invention

The main purpose of the present application is to provide a medical cost prediction method, device, equipment, and computer-readable storage medium, aiming to solve the problem of lack of an effective prediction mechanism and inaccurate prediction for medical costs of chronic disease complications in the prior art.

In order to achieve the above object, the present application provides a medical expense prediction method, which includes the following steps:

Obtain multiple copies of historical consultation data from medical institutions, compare each of the historical consultation data with preset identifiers one by one, and filter the historical consultation data carrying the preset identifier in each of the historical consultation data into Multiple pieces of concurrent characteristic data corresponding to preset chronic disease complications, wherein each piece of concurrent characteristic data at least includes information on the type of concurrency, medication type, and medical expenses;

Read the concurrency type information and the medication type information in each of the concurrency characteristic data, and classify each of the concurrency characteristic data according to each of the concurrency type information and the medication type information to form each of the concurrency Correspondence between type information and each type of medication information and medical expenses in each piece of concurrent characteristic data;

When receiving the diagnosis information of the visiting patient, the diagnosis information and the corresponding relationship are compared, and a target medical cost corresponding to the diagnosis information is predicted.

In addition, in order to achieve the above object, the present application also proposes a medical expense prediction device, the medical expense prediction device includes:

The obtaining module is used to obtain multiple historical consultation data from the medical institution, and compare each of the historical consultation data with the preset identifier one by one, and compare each historical consultation data with the preset identifier The historical visit data is screened into multiple pieces of concurrent characteristic data corresponding to the preset chronic disease complications, wherein each of the concurrent characteristic data includes at least information about the type of concurrency, medication type, and medical expenses;

A classification module, used to read the concurrency type information and the medication type information in each of the concurrent feature data, and classify each of the concurrent feature data according to each of the concurrency type information and the medication type information, Forming a correspondence between each of the concurrency type information and each of the medication type information and the medical expenses in each of the concurrency characteristic data;

The prediction module is configured to compare the diagnosis information and the corresponding relationship when receiving the diagnosis information of the visiting patient, and predict the target medical expenses corresponding to the diagnosis information.

In addition, in order to achieve the above object, the present application also provides a medical expense prediction device, the medical expense prediction device includes: a memory, a processor, a communication bus, and a medical expense prediction program stored on the memory;

The communication bus is used to implement connection communication between the processor and the memory;

The processor is used to execute the medical expense prediction program to implement the steps of the aforementioned medical expense prediction method.

In addition, in order to achieve the above object, the present application also provides a computer-readable storage medium that stores one or more programs, and the one or more programs may be used by one or more processors Steps performed to realize the medical cost prediction method as described above.

The medical cost prediction method of this embodiment grabs multiple concurrent characteristic data corresponding to preset chronic disease complications from a large amount of historical visit data to characterize the information of patients with various chronic disease complications during the visit process ; Then classify the concurrent feature data based on the concurrent type information and medication type information read from the concurrent feature data; because each concurrent feature data also involves the medical expenses of the patient at the time of the visit, the classification forms each concurrency type information and medication Correspondence between type information and medical expenses; follow-up will compare the received diagnostic information characterizing the disease and the corresponding relationship received by the visiting patient to determine the corresponding relationship with the disease characterized by the diagnostic information in various medication programs The medical expenses on the list have realized the prediction of the target medical expenses corresponding to the diagnostic information. Because the concurrency feature data comes from a large amount of real and effective historical visit data, the corresponding relationship formed has a high accuracy, thereby improving the accuracy of predicting the medical expenses of patients with various chronic disease complications.

BRIEF DESCRIPTION

FIG. 1 is a schematic flowchart of a first embodiment of a medical expense prediction method of this application;

2 is a schematic diagram of functional modules of the first embodiment of the medical expense prediction apparatus of the present application;

FIG. 3 is a schematic diagram of the device structure of the hardware operating environment involved in the method of the embodiment of the present application.

The implementation, functional characteristics and advantages of the present application will be further described in conjunction with the embodiments and with reference to the drawings.

detailed description

It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

This application provides a medical cost forecasting method.

Please refer to FIG. 1, which is a schematic flowchart of a first embodiment of a medical expense prediction method of this application. In this embodiment, the medical expense prediction method includes:

Step S10: Acquire multiple pieces of historical consultation data from the medical institution, compare each piece of the historical consultation data with the preset identifier one by one, and compare each piece of the historical consultation data with the historical identifier carrying the preset identifier The data screening is a plurality of concurrent characteristic data corresponding to the preset chronic disease complications, wherein each of the concurrent characteristic data at least includes information about the type of concurrency, medication type and medical expenses;

The medical expense prediction method of the present application is applied to a server, and is suitable for predicting the medical expenses of patients diagnosed with various chronic disease complications through the server; where chronic disease complications are other diseases caused by the development of chronic diseases, while chronic diseases are Refers to diseases that do not constitute infection and have long-term accumulation to form disease morphological damage; such as diabetes, hypertension, coronary heart disease, asthma, chronic gastritis, etc.; taking the most typical diabetes in chronic diseases as an example, its complications include atherosclerosis and Cardio-cerebrovascular diseases, diabetic nephropathy, neuropathy, ocular diseases, combined with tuberculosis, acute infection, etc. The medical expenses predicted in this embodiment are the costs that patients with chronic disease complications may need to visit a medical institution. Medical institutions include but are not limited to general hospitals, traditional Chinese medicine hospitals, specialty hospitals and other types of hospitals and clinics , Health centers, pharmacies, etc. The types of chronic disease complications among patients with various chronic disease complications are different, and the medication options for the same type of chronic disease complications also vary greatly, resulting in the treatment of patients with various types of chronic disease complications. The medical expenses are different. However, for the same type of chronic disease complications and similar treatment regimens, the medical costs are similar; therefore, the current medical expenses of various types of chronic disease complications and various medication regimens can be used to treat current patients. The medical costs of the types of chronic disease complications characterized in the diagnostic information are predicted. For example, in the past, when treatment of hypertension heart disease and heart failure was treated with plan A, 90% of the patients' medical expenses were between a1 and a2, so that the medical expense interval can be used as a reference benchmark to characterize the concurrency of chronic diseases in the diagnostic information. The type of symptom is hypertension heart disease and heart failure, and the medical cost of treatment with plan A is predicted to be between a1 and a2.

Understandably, when treating patients, each medical institution will record the data related to the illness during the treatment of each patient; such as the type of illness, the time of treatment, the treatment method, the type of medication, the amount of various medications, the medical expenses, etc. ; Use this kind of data recorded in medical institutions as historical visit data, which includes historical visit data of chronic disease complications, and also includes historical visit data of other types of diseases. There is a communication connection between the server and each medical institution. Before predicting the medical expenses of the patients, it is necessary to obtain the historical visit data recorded in the medical institution for the treatment of various diseases of each patient, and from the recorded From each historical visit data, historical visit data related to chronic disease complications are selected. Specifically, the server sends a request to each medical institution to obtain historical visit data. After receiving the request, each medical institution transmits multiple historical visit data recorded in it to the server, and one of the patients sees the visit data at once Corresponding to a piece of historical consultation data.

Because the historical consultation data uploaded by various medical institutions include the historical consultation data of chronic disease complications and the historical consultation data of other chronic disease complications, the historical consultation corresponding to chronic disease complications needs to be extracted from each historical consultation data data. Various types of preset chronic disease complications are preset, and each preset chronic disease complication is characterized by a preset identifier, and one preset identifier characterizes a preset chronic disease complication; each historical consultation data is summarized one by one The preset identifiers are compared to determine historical visit data corresponding to preset chronic disease complications. Use each historical visit data corresponding to the preset chronic disease complication as the concurrent characteristic data corresponding to the preset chronic disease complication, and after comparing and determining the concurrent feature data existing in each historical visit data, grab from each historical visit data Take each piece of concurrent characteristic data corresponding to the preset chronic complication. Each piece of concurrent characteristic data includes at least the concurrent type, medication type, medication amount, medical cost and other information of each patient.

Step S20, read the concurrency type information and the medication type information in each of the concurrency characteristic data, and classify each of the concurrency characteristic data according to each of the concurrency type information and the medication type information to form each Correspondence between the concurrency type information and each of the medication type information and the medical expenses in each of the concurrency characteristic data;

Further, because a piece of concurrent characteristic data characterizes the medical data of a patient visiting a medical institution for a predetermined type of chronic disease complication, that is, the concurrent characteristic data is related to the preset chronic disease complication, and the concurrent type information represents the Set the type of chronic disease complication, and the medication type information characterizes the medication type and dosage for such preset chronic complications. The type of the preset chronic disease complication and the type of medication used are different between the different patients, which makes the concurrent type information and medication type information in each piece of concurrent characteristic data different. In order to characterize the characteristics of various preset chronic disease complications through each concurrent characteristic data, the concurrent type information and medication type information in each concurrent characteristic data are read, and the concurrent characteristics are based on the concurrent type information and medication type information Classify the data; divide the concurrent feature data with the same type of concurrent type information and medication type information into the same category, so that a type characterizes a preset chronic disease complication and the medication type of the preset chronic disease complication, Through various types of classification to reflect the various treatment options of various preset chronic disease complications. Specifically, the steps for classifying each piece of concurrent feature data according to each type of concurrent information and medication type information include:

Step S21: Compare each of the concurrency type information, and divide each piece of the concurrency characteristic data having the same concurrency type information into the same group category;

Understandably, for different types of preset chronic disease complications, there are great differences in medication. When classifying each concurrent feature data, first classify the same type of preset chronic disease complications according to the concurrency type information. The same group is divided into different medication type information of concurrent feature data in each group. Specifically, the concurrent type information of each concurrent feature data is compared, the same concurrent type information is selected, and then the concurrent feature data having the same concurrent type information is divided into the same group. In the process of comparing the various types of concurrency information, read the types of concurrency information one by one, and determine whether the types of concurrency information previously read are consistent with the types of concurrency information currently read; if there are consistent Concurrent type information, the currently read concurrent type information is classified into this consistent concurrent type information; if there is no consistent concurrent type information, the currently read concurrent type information is regarded as a new type of concurrent type information , To facilitate subsequent reading judgment. For three concurrent feature data A, B, C, A includes [a1, a2, a3], B includes [b1, b2, b3], C includes [c1, c2, c3]; where a1, b1, c1 characterize Concurrent type information, a2, b2, and c2 represent medication type information, and a3, b3, and c3 represent medical expenses. When comparing, first read a1 and after judging that there is no concurrency type information consistent with it, use a1 as a new type of concurrency type information; then read b1 and determine whether it is the same concurrency type information as a1 , If it is the same, classify b1 into the concurrency type information of a1; then read c1, and judge whether it is the same concurrency type information as a1, if not, use c1 as a new type of concurrency type information, in order to It is used to judge the concurrency type information read subsequently. After the comparison of each concurrency type information is completed, each concurrency characteristic data is divided into each group category; each divided group category is a group category corresponding to various preset chronic disease complications, and one group category corresponds to one category preset Complications of chronic diseases.

Step S22, comparing the medication type information possessed by each piece of the concurrent characteristic data in each of the group categories, and comparing each piece of the concurrent characteristic in each of the group classes with the same medication type information The data is divided into sub-groups in each of the group categories.

Further, due to the same type of preset chronic disease complications, when the medication regimen used is different, there will be different medical expenses, so after forming a group corresponding to each type of preset chronic disease complications, continue to target each group category The concomitant feature data in is used to classify the medication regimen. Because the medication plan is reflected by medication type and medication amount, when re-classifying the divided groups, the medication type information with concurrent feature data in each group is used. Compare the medication type information of each concurrent feature data in each group category, and filter out the same medication type information, and then divide each concurrent feature data with the same medication type information in each group category into the same sub-group category, forming each The subgroup class in the group class. Similarly, in the process of comparing the information of each medication type, according to the division limit of each group, each medication type information in each group is read one by one, and it is judged whether there is any The currently read medication type information is consistent with the medication type information; if there is consistent medication type information, the currently read medication type information is classified into this consistent medication type information; if there is no consistent medication type information, Then, the currently read medication type information is used as a new type of medication type information to facilitate subsequent reading judgment. For the above-mentioned concurrent feature data A, B, and C, because the concurrent feature information a1 and b1 are the same, A and B are divided into the same group; read the medication type information a2 in the group and judge it does not exist After the medication type information that is consistent with it, a2 is regarded as a new type of medication type information; then b2 is read, and it is judged whether it is the same medication type information as a2, and if it is the same, b2 is classified into the medication type information of a2 in. After the comparison of the medication type information in the current group is completed, the medication type information of the next group is read for comparison, until all the groups classified according to the concurrent type information are compared with the medication type information.

After comparing the medication type information in each group class, the concurrent characteristic data in each group class is divided into sub-group classes in each group class according to the medication type information; the sub-group class in each divided group class That is, it is a sub-group category corresponding to various medication plans for preset chronic disease complications, and one sub-group category corresponds to a medication plan for a preset preset chronic disease complication.

In addition, because each concomitant feature data includes the medical expenses of the visiting patient, the concomitant feature data located in each sub-group category corresponds to the medical expenses; the concurrency type information in each sub-group category and the medication type information and medical expenses Correspondence is formed between them, which represents the medical cost corresponding to the treatment of various types of preset chronic disease complications with various medication regimens. Among them, the steps of forming the correspondence between each type of concurrency information and each type of medication information and the medical expenses in each piece of concurrent characteristic data include:

Step S23: Read the number of data copies with the concurrent feature data in each of the sub-group categories, and the medical expenses corresponding to each of the concurrent feature data;

Understandably, although the sub-groups in the group characterize the concurrent characteristic data of each medication plan that the same preset chronic disease complication type has, the medical expenses corresponding to the concurrent characteristic data are still different, but The difference is small; in order to more accurately reflect the predetermined chronic disease complication type and medical expenses corresponding to the medication plan characterized by the subgroup, the medical expenses in the subgroup are integrated and the medical expenses are Averaging treatment to reflect the medical expenses that may be required for various medication programs through the average medical expenses. Specifically, the number of copies of concurrent feature data in each sub-group category is different. Read the number of data copies with concurrent feature data in each sub-group category and the corresponding number of concurrent feature data in each sub-group category. For medical expenses, the medical expenses of each sub-group are averaged according to the number of data and the medical expenses in each part.

Step S24, according to the number of data copies and the medical expenses corresponding to each of the concurrent characteristic data, determine the average medical expenses of each sub-group to form each of the concurrent type information and each of the medication type information and Correspondence between medical expenses in each of the concurrent characteristic data.

Further, the medical expenses of the concurrent feature data in each sub-group category are added, and the result of the addition is the total value of the medical expenses of the concurrent feature data in each sub-group category. The ratio is the number, and the result of the ratio is the average medical cost of each subgroup; in the process of addition and ratio, each subgroup is used as a unit; that is, for each concurrent feature data in the same subgroup The medical expenses are added together, and at the same time, the total value of medical expenses in the same sub-group category is compared with the number of data copies, until each sub-group category in each group category obtains the average value of medical expenses. Form a correspondence between group categories, sub-group categories, and the average value of medical expenses, where group categories include multiple sub-group categories, and sub-group categories correspond to medical expenses one-to-one. Groups are divided according to concurrency type information to characterize each preset chronic disease complication, while subgroups are divided according to medication type information to characterize each preset chronic disease complication medication plan; thus each group, subgroup and medical expenses The corresponding relationship between the essence is the corresponding relationship between each concurrency type information, each medication type information, and the medical expenses in each concurrent characteristic data; and the medical expenses of each concurrent characteristic data used to form the corresponding relationship are the concurrent type information After being classified with the medication type information, the medical expenses of the concurrent feature data located in each category exist, and the medical expenses of the concurrent feature data located in each category exist in the form of mean medical expenses in the corresponding relationship.

In step S30, when receiving the diagnosis information of the visiting patient, the diagnosis information and the corresponding relationship are compared to predict the target medical expenses corresponding to the diagnosis information.

Furthermore, when a corresponding relationship is formed between the medical expenses corresponding to the various medication schemes for the preset chronic disease complications, when the visiting patient needs to predict the medical expenses of a certain type of chronic disease complications during the medical treatment, Send a prediction request to the server, and upload the diagnostic information and prediction request that need to be predicted to the server; the diagnostic information is the information of the chronic disease complications of the patient diagnosed by the medical institution. When the server receives the diagnosis information of the visiting patient, the diagnosis information is compared with the corresponding relationship; the diagnosis information includes the type information of the chronic disease complication of the visiting patient, and the corresponding relationship involves multiple preset chronic diseases Complications are treated by various medications. When comparing the diagnostic information and the corresponding relationship, find the group corresponding to the type information of the chronic disease complications represented in the diagnostic information from the corresponding relationship, and then by The sub-group in this group predicts the target medical cost required for the treatment of complications of chronic diseases suffered by patients who come from the diagnosis information with various medication schemes. The corresponding medication type information and the average medical cost in each sub-group category represent the target medical cost that the visiting patient needs to spend when treating with this type of medication type.

Further, in another embodiment of the medical expense prediction method of the present application, the step of comparing the diagnostic information and the corresponding relationship to predict the target medical expense corresponding to the diagnostic information includes:

Step S31: Read the concurrent type identifier in the diagnostic information, and compare the concurrent type identifier with each of the concurrent type information in the corresponding relationship to determine the target group category corresponding to the diagnostic information;

Further, when the medical institution diagnoses the visiting patient and generates the diagnosis information of the visiting patient, it will add an identifier representing the type of chronic disease complication suffered by the visiting patient to the diagnosis information, and use this identifier as the concurrent type identifier. When comparing the diagnostic information and the corresponding relationship, read the concurrent type identifier in the diagnostic information to determine the type of the chronic disease complication of the patient; the preset concurrent representation of the read concurrent type identifier and the corresponding relationship Contrast type information of chronic disease complication types is compared, and from each corresponding type information of the corresponding relationship, the concurrent type information consistent with the concurrent type is determined. Since the preset chronic disease complication type corresponding to the group class divided according to the consistent concurrency type information is consistent with the chronic disease complication type characterized by the diagnosis information, this group class is regarded as the target group class corresponding to the diagnosis information, The medical expenses of the patients from the source of diagnosis information are predicted by the average medical expenses of each sub-group in the target group.

Step S32: Read the target sub-group category in the target group category and the average target medical cost corresponding to each target sub-type, and determine each average target medical cost to correspond to the diagnosis information Target medical expenses.

Understandably, because the concomitant feature data in the target group category will be divided into sub-group categories according to the medication type information, and each sub-group category corresponds to the average medical cost, which represents the preset chronic disease complications corresponding to the target group category. The cost of various types of drugs for treatment. However, patients suffering from preset chronic disease complications represented by the target group may be treated with any type of drugs characterized by each sub-group in the target group, so that their medical expenses are related to the type of drugs they use. Therefore, each sub-group category in the target group category is first read as the target sub-group category, and then the average medical cost corresponding to each target sub-group category is read as the average target medical cost. The read target subgroup categories and the corresponding target medical cost averages are the possible medication plans and corresponding medical costs for the visiting patient, so that the average target medical cost is taken as the target medical cost corresponding to the diagnostic information, In order to characterize the corresponding cost of the treatment of patients with chronic disease complications using various medication regimens, the corresponding costs may be required. At the same time, each target subgroup category and the corresponding target medical cost average value can also be output to the mobile phone, tablet computer and other terminals held by the visiting patient to inform the visiting patient of the cost of treatment with various medication schemes to facilitate Make reasonable choices for patients.

In addition, considering that the patient's other diseases caused by chronic diseases involve multiple types, that is, there are multiple chronic disease complications, such as diabetes with eye complications and leg complications. For such patients who are involved in a variety of chronic disease complications, the multiple chronic disease complications are reflected in the diagnosis information of the patient. The concurrent type identification in the diagnostic information includes multiple types. When comparing the concurrent type identification and the corresponding concurrent type information in the corresponding relationship, the determined target group class corresponding to the diagnostic information also involves multiple types and different types of target group classes The medical expenses corresponding to each chronic disease complication suffered by the visiting patient are different. In order to determine whether the visiting patient suffers from multiple chronic disease complications, the step of determining the average value of each target medical cost as the target medical cost corresponding to the diagnosis information in this embodiment includes:

Step S33, judging whether there are multiple types of the target group class, if there are multiple types of the target group class, counting the number of sub-group classes of the target sub-group class in each of the target group classes;

Determine whether there are multiple types of target groups determined by the concurrent type identification and the information of each concurrent type; if there are no multiple types of target groups, it means that the patient has no multiple chronic disease complications and only one type of chronic disease. The target medical cost can be predicted for this chronic disease complication, that is, the average target medical cost of each target subgroup in the target group category is used as the target medical cost of the patient. When there are multiple target groups, it means that the patient has a variety of chronic disease complications, and it is necessary to predict the target medical cost for multiple chronic disease complications. Because the chronic disease complications characterized by the target groups determined by comparison are the chronic disease complications of the visiting patients, the average medical cost corresponding to each sub-group category in each target group category is the concurrent of the chronic diseases suffered by the visiting patients. Disease, the various medical expenses required for treatment with different medication schemes, that is, the average target medical expenses of each target sub-group category in each target group category are the target medical expenses of the patient.

However, the medical costs of different types of chronic disease complications in the treatment process are different. In the process of predicting the target medical cost of patients with various chronic disease complications, the dominant cost of the target medical cost can be predicted. The leading cost is the treatment of various types of chronic disease complications, accounting for the highest total amount of medical expenses required; for example, the chronic disease complications of patients include M and N, and it is predicted that M corresponds to The target medical cost is m, and the target medical cost corresponding to N is n. If m is greater than n, it means that the dominant cost in target medical costs m and n is m. Considering that the target medical costs of various types of chronic disease complications exist as the average of the corresponding medical costs in each target group category, the target medical costs of a type of chronic disease complications correspond to different medication schemes with multiple average medical costs. In order to reflect the leading cost of target medical expenses corresponding to various types of chronic disease complications, it is necessary to integrate the average of multiple target medical expenses in the target medical expenses corresponding to various types of chronic disease complications. Specifically, the number of target subgroup categories in each target group category is inconsistent, making the average number of target medical expenses in target medical expenses inconsistent; count the number of subgroup categories of target subgroup categories in each target group category In order to average the target medical cost in each target group based on the number of sub-groups, the average target medical cost can be used to reflect the medical costs that may be required for various chronic disease complications.

Step S34: Determine the average group cost of each target group according to the number of the subgroups and the average value of the target medical expenses of each target subgroup in each of the target groups;

Further, after the number of sub-groups of each target group is counted, the average value of the target medical cost of each target sub-group in each target group is read, and mean value of each target medical cost is added to obtain The result of the addition is the sum of the average target medical costs of the target subgroups of each target group class; the sum of the average target medical costs and the number of subgroup classes are used as a ratio, and the result of the ratio is the group of each target group class The average cost reflects the average medical cost required to treat the pre-defined chronic disease complications characterized by the target group. In the process of addition and ratio, each target group is used as a unit; that is, the target medical expenses in the same target group are added, and the average target medical expenses in the same target group are used. The sum is compared to the number of sub-group categories until each target group category generates an average group cost.

Step S35: Compare the average cost of each of the groups, determine the average cost of the target group with the largest value, and output the average cost of the target group to the terminal held by the visiting patient, taking the average cost of the target group as Remind the leading expenses in the target medical expenses.

Understandably, because the number of target subgroup categories in each target group category is different, and the average value of target medical expenses corresponding to each target subgroup category is also different, the average value of group costs formed by each target group category is also There are differences; in order to determine the leading cost in the target medical cost, the average cost of each group is compared, and the average cost of the group with the largest value is found from the average cost of each group as the average cost of the target group, which represents the average cost of the patient and the target group To characterize the treatment of chronic disease complications, the maximum medical cost required. Output the determined average cost of the target group and its corresponding chronic disease complication type to the terminal held by the visiting patient to inform the visiting patient in the process of treating various types of chronic disease complications they have, The most expensive medical expenses and the corresponding types of chronic disease complications.

Further, in another embodiment of the medical expense prediction method of the present application, after the step of forming the correspondence relationship between each of the concurrency type information and each of the medication type information and each of the concurrency characteristic data, include:

Step S25, when receiving the medical data to be tracked, reading the concurrent type field and the medication type field in the medical data to be tracked;

Understandably, due to the corresponding relationship between the concurrency type information, medication type information and medical expenses formed, it accurately reflects the medical expenses required for the treatment of various chronic disease complications with various medication schemes; This correspondence relationship judges the abnormality of medical expenses incurred by the clinic for each chronic disease complication. Specifically, the medical cost required for abnormality judgment, the type of chronic disease complication that generates the medical cost, and the medication plan are used as the data to be tracked, and the data to be tracked are transmitted to the server. When receiving the medical data to be tracked, the concurrent type field and the medication type field are read; the concurrent type field is used to characterize the type of chronic disease complications, and the medication type field is used to characterize the medication regimen, based on the concurrent type field and medication The type field determines the anomaly of medical expenses in the data to be tracked.

Step S26: Compare the concurrency type field with each of the concurrency type information in the corresponding relationship to determine a tracking group class corresponding to the data to be tracked.

Further, the correspondence relationship is divided into groups according to the concurrency type information, and the concurrency type information characterizes the type of chronic disease complications; thus comparing the read concurrency type field with each concurrency type information in the correspondence relationship, from the correspondence relationship Find the concurrency type information consistent with the concurrency type field, and the group class based on the consistent concurrency type information is the group class corresponding to the chronic disease complication represented by the concurrency type field; the corresponding group class is determined to be tracked The tracking group corresponding to the medical data is judged by the abnormality of the medical expenses in the medical data to be tracked by the tracking group.

Step S27: Compare the medication type field with the medication type information of each of the sub-group classes in the tracking group class to determine the tracking sub-group class corresponding to the medical data to be tracked;

Furthermore, the tracking group is divided into multiple sub-groups with different medication schemes according to the medication type information, and different sub-groups correspond to different medical expenses; the medication type field and tracking of the medication plan in the medical data to be tracked Comparing the medication type information of each sub-group in the group category, finding the medication type information consistent with the medication type field from each medication type information, and the sub-group category divided according to the consistent medication type information is characterized by the medication type field The sub-group category corresponding to the medication plan; determine the corresponding sub-group category as the tracking sub-group category corresponding to the medical data to be tracked, so that the tracking sub-group category can determine the abnormality of the medical expenses in the tracking data.

Step S28: Read the average value of the tracking medical expenses corresponding to the tracking subgroup and the medical expenses to be tracked in the medical data to be tracked, and compare the average value of the tracking medical expenses to the medical expenses to be tracked To determine the abnormality of the medical expenses to be tracked.

Understandably, each subgroup in the corresponding relationship has a corresponding average medical cost. The average medical cost corresponding to the tracking subgroup is read as the average medical cost of tracking; at the same time, the data in the medical data to be tracked is read. Medical expenses are regarded as medical expenses to be tracked which need to be judged for abnormality. Compare the average of the tracked medical expenses and the medical expenses to be tracked to determine whether the average medical expenses to be tracked and the average medical expenses to be tracked are consistent. If they are consistent, the medical expenses to be tracked are normal, and if they are not consistent, the medical expenses to be tracked are abnormal. Consistency can be characterized by the floating range of the average tracking medical cost, such as setting the floating range of the average tracking medical cost to be plus or minus 10; that is, based on the average tracking medical cost when comparing, the medical cost to be tracked is reduced by the average tracking medical cost Between the range of 10 and the increase of 10, it is normal; otherwise it is abnormal to ensure that the medical expenses to be tracked are accurately judged by abnormality.

Further, in another embodiment of the medical expense prediction method of the present application, the step of predicting the target medical expense corresponding to the diagnosis information includes:

Step S40, according to the preset association relationship between each of the preset chronic disease complications, predict the change trend of the disease corresponding to the diagnosis information;

Understandably, for various diseases caused by chronic diseases, after a certain concurrent disease occurs, this concurrent disease may develop into other concurrent diseases, that is, there is a correlation between various types of chronic disease complications, such as diabetic retinopathy may develop into Diabetic nephropathy, dominant diabetic nephropathy may develop into end-stage renal failure. Such an association relationship is calculated from historical medical data of various chronic disease complications, and the association relationship is set in the server as a preset association relationship in advance. After receiving the diagnosis information of the visiting patient and predicting the target medical expenses corresponding to the diagnosis information; further predicting the change trend of the disease corresponding to the diagnosis information by the preset association relationship, that is, the treating patient reflected in the diagnosis information suffers from The possible complications of chronic disease complications. If the chronic disease complication P in the preset association relationship changes with time to K, then to S, and then to w; if the chronic disease complication characterized by the diagnosis information of the patient has changed from P to K, it can be predicted The next symptom change trend is S.

Step S50: Compare the disease change trend and the corresponding relationship, determine the cost change trend corresponding to the disease change trend, and send the disease change trend and the cost change trend to the patient There are terminals.

Because the change in the disease is the type of chronic disease complications, and the different medication treatments used for different types of chronic disease complications, the medical costs are different. After determining the disease change trend of the diagnosis information, that is, the type of possible chronic disease complications, compare the disease change trend with the corresponding relationship, and find out from the corresponding relationship the concurrent type information that is consistent with the chronic disease complication type characterized by the disease change trend, The group divided according to the consistent concurrency type information is the medical cost required to treat the possible chronic disease complications with each medication plan, and the required medical cost is the target medical corresponding to the diagnostic information Medical expenses that may change in expenses, and use them as the trend of expenses changes. Send the predicted disease change trend and cost change trend to the terminal held by the visiting patient for display to inform the visiting patient of the possible changes in chronic disease complications based on the chronic disease complications reflected in the current diagnosis information, And the medical expenses required for the treatment of this changing chronic disease complication with various medication schemes; and remind the patients of the current chronic disease complications of attention to avoid deterioration.

In addition, referring to FIG. 2, this application provides a medical expense prediction device. In the first embodiment of the medical expense prediction device of this application, the medical expense prediction device includes:

The obtaining module 10 is used to obtain multiple pieces of historical consultation data from a medical institution, compare each piece of the historical consultation data with a preset identifier one by one, and carry the preset identifier in each piece of the historical consultation data The historical consultation data of is screened into multiple pieces of concurrent characteristic data corresponding to the preset chronic disease complications, wherein each piece of the concurrent characteristic data includes at least information about the type of concurrency, medication type and medical expenses;

The classification module 20 is configured to read the concurrent type information and medication type information in each of the concurrent characteristic data, and classify each of the concurrent characteristic data according to each of the concurrent type information and the medication type information , Forming a correspondence between each of the concurrency type information and each of the medication type information and the medical expenses in each of the concurrency characteristic data;

The prediction module 30 is configured to compare the diagnosis information and the corresponding relationship when receiving the diagnosis information of the visiting patient, and predict the target medical expenses corresponding to the diagnosis information.

In the medical expense prediction apparatus of this embodiment, the acquisition module 10 grabs multiple pieces of concurrent characteristic data corresponding to preset chronic disease complications from a large amount of historical visit data to characterize the process of patients with various chronic disease complications during the visit Information; the classification module 20 then classifies the concurrent feature data based on the concurrent type information and medication type information read from the concurrent feature data; because each concurrent feature data also involves the medical expenses of the patient during the consultation, the classification is formed Correspondence between each type of concurrency information, medication type information and medical expenses; the subsequent prediction module 30 compares the received diagnostic information characterizing the disease and the corresponding relationship between the received patients to determine the corresponding relationship with the diagnosis The medical cost of the disease represented by the information on various medication programs has realized the prediction of the target medical cost corresponding to the diagnostic information. Because the concurrency feature data comes from a large amount of real and effective historical visit data, the corresponding relationship formed has a high accuracy, thereby improving the accuracy of predicting the medical expenses of patients with various chronic disease complications.

Wherein, each virtual function module of the above medical expense prediction apparatus is stored in the memory 1005 of the medical expense prediction device shown in FIG. 3, and when the processor 1001 executes the medical expense prediction program, the functions of each module in the embodiment shown in FIG. 2 are realized.

It should be noted that, those of ordinary skill in the art may understand that all or part of the steps to implement the above embodiments may be completed by hardware, or may be completed by a program instructing related hardware. The program may be stored in a computer-readable In the storage medium, the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk.

Referring to FIG. 3, FIG. 3 is a schematic structural diagram of a device in a hardware operating environment involved in a method according to an embodiment of the present application.

The medical expense prediction device in the embodiment of the present application may be a PC (personal computer, personal computer ), or terminal devices such as smart phones, tablet computers, e-book readers, and portable computers.

As shown in FIG. 3, the medical expense prediction device may include: a processor 1001, such as a CPU (Central Processing) Unit, central processing unit), memory 1005, communication bus 1002. Among them, the communication bus 1002 is used to implement connection communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM (random access memory, random access memory), can also be a stable memory (non-volatile memory), such as disk storage. The memory 1005 may optionally be a storage device independent of the foregoing processor 1001.

Those skilled in the art can understand that the structure of the medical expense prediction device shown in FIG. 3 does not constitute a limitation on the medical expense prediction device, and may include more or less components than the illustration, or a combination of certain components, or different Parts arrangement.

As shown in FIG. 3, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, and a medical expense prediction program. The operating system is a program that manages and controls the hardware and software resources of the medical expense prediction equipment, and supports the operation of the medical expense prediction program and other software and/or programs. The network communication module is used to implement communication between various components within the memory 1005, and to communicate with other hardware and software in the medical expense prediction device.

In the medical expense prediction apparatus shown in FIG. 3, the processor 1001 is used to execute a medical expense prediction program stored in the memory 1005 to implement the steps in each embodiment of the above medical expense prediction method.

The present application provides a computer-readable storage medium. The computer-readable storage medium is preferably a computer-readable computer-readable storage medium. The computer-readable computer-readable storage medium stores one or more programs. One or more programs may also be executed by one or more processors to implement the steps in the above embodiments of the medical expense prediction method.

Through the description of the above embodiments, those skilled in the art can clearly understand that the methods in the above embodiments can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware, but in many cases the former is better Implementation. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product in essence or a part that contributes to the existing technology, and the computer software product is stored in a computer-readable storage medium (such as The ROM/RAM, magnetic disk, and optical disk include several instructions to enable a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to perform the methods described in the embodiments of the present application.

The above are only the preferred embodiments of the present application, and therefore do not limit the patent scope of the present application. Any equivalent structural transformations made by using the description and drawings of this application under the concept of this application, or directly/indirectly used in Other related technical fields are included in the patent protection scope of this application.

Claims

A medical expense prediction method, characterized in that the medical expense prediction method includes the following steps:

Obtain multiple copies of historical consultation data from medical institutions, compare each of the historical consultation data with preset identifiers one by one, and filter the historical consultation data carrying the preset identifier in each of the historical consultation data into Multiple pieces of concurrent characteristic data corresponding to preset chronic disease complications, wherein each piece of concurrent characteristic data at least includes information on the type of concurrency, medication type, and medical expenses;

Read the concurrency type information and the medication type information in each of the concurrency characteristic data, and classify each of the concurrency characteristic data according to each of the concurrency type information and the medication type information to form each of the concurrency Correspondence between type information and each type of medication information and medical expenses in each piece of concurrent characteristic data;

When receiving the diagnosis information of the visiting patient, the diagnosis information and the corresponding relationship are compared, and a target medical cost corresponding to the diagnosis information is predicted.
The medical cost forecasting method according to claim 1, wherein each of the concurrency characteristic data is classified according to each of the concurrency type information and the medication type information to form each of the concurrency type information And the step of the correspondence between each of the medication type information and the medical expenses in each of the concurrent characteristic data includes:

Comparing each of the concurrency type information, and dividing each piece of the concurrency characteristic data having the same concurrency type information into the same group;

Comparing the medication type information possessed by each piece of the concurrent feature data in each of the group categories, and dividing each piece of the concurrent feature data having the same medication type information in each of the group categories into Sub-group category in each of the group categories;

Reading the number of data copies having the concurrent feature data in each of the sub-group categories, and the medical expenses corresponding to the concurrent feature data of each share;

According to the number of data copies and the medical expenses corresponding to each of the concurrent characteristic data, determine the average medical expenses of each of the sub-groups to form each of the concurrency type information and each of the medication type information and each The corresponding relationship of medical expenses in the concurrent characteristic data is described.
The method for predicting medical expenses according to claim 2, wherein the step of comparing the diagnosis information and the corresponding relationship to predict the target medical expenses corresponding to the diagnosis information includes:

Reading the concurrent type identifier in the diagnostic information, and comparing the concurrent type identifier with each of the concurrent type information in the corresponding relationship, and determining a target group category corresponding to the diagnostic information;

Reading the target sub-group class in the target group class and the average value of the target medical expenses corresponding to each target sub-type, and determining each average value of the target medical expenses as the target medical treatment corresponding to the diagnosis information cost.
The medical cost prediction method according to claim 3, wherein after the step of determining the average value of each target medical cost as the target medical cost corresponding to the diagnosis information includes:

Judging whether there are multiple classes of the target group class, if there are multiple classes of the target group class, counting the number of sub-group classes of the target sub-group class in each of the target group classes;

Determine the average group cost of each target group according to the number of subgroups and the average value of the target medical expenses of each target subgroup in each of the target groups;

Compare the average cost of each of the groups, determine the average value of the target group with the largest value, and output the average value of the target group to the terminal held by the visiting patient, with the average value of the target group as the target Remind the leading expenses in medical expenses.
The medical cost prediction method according to claim 2, wherein the step of forming a correspondence between each of the concurrency type information and each of the medication type information and each of the concurrency characteristic data includes :

When receiving the medical data to be tracked, the concurrent type field and the medication type field in the medical data to be tracked are read;

Comparing the concurrency type field with each of the concurrency type information in the corresponding relationship, and determining a tracking group class corresponding to the data to be tracked;

Comparing the medication type field with the medication type information of each of the sub-group classes in the tracking group class to determine the tracking sub-group class corresponding to the medical data to be tracked;

Reading the average value of the tracking medical expenses corresponding to the tracking subgroup and the medical expenses to be tracked in the medical data to be tracked, and comparing the average value of the tracking medical expenses to the medical expenses to be tracked to determine Describe the anomalies of medical expenses to be tracked.
The method for predicting medical expenses according to claim 1, wherein after the step of predicting the target medical expenses corresponding to the diagnosis information includes:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The method for predicting medical expenses according to claim 2, wherein the step of predicting the target medical expenses corresponding to the diagnostic information includes:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The medical cost prediction method according to claim 3, wherein after the step of predicting the target medical cost corresponding to the diagnosis information includes:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The method for predicting medical expenses according to claim 4, wherein the step of predicting the target medical expenses corresponding to the diagnosis information includes:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
A medical expense prediction device, characterized in that the medical expense prediction device includes:

The obtaining module is used to obtain multiple historical consultation data from the medical institution, and compare each of the historical consultation data with the preset identifier one by one, and compare each historical consultation data with the preset identifier The historical visit data is screened into multiple pieces of concurrent characteristic data corresponding to the preset chronic disease complications, wherein each of the concurrent characteristic data includes at least information about the type of concurrency, medication type, and medical expenses;

A classification module, used to read the concurrency type information and the medication type information in each of the concurrent feature data, and classify each of the concurrent feature data according to each of the concurrency type information and the medication type information, Forming a correspondence between each of the concurrency type information and each of the medication type information and the medical expenses in each of the concurrency characteristic data;

The prediction module is configured to compare the diagnosis information and the corresponding relationship when receiving the diagnosis information of the visiting patient, and predict the target medical expenses corresponding to the diagnosis information.
The medical expense prediction apparatus according to claim 10, wherein the classification module includes:

A comparison unit, configured to compare each of the concurrency type information, and divide each piece of the concurrency characteristic data having the same concurrency type information into the same group category;

A dividing unit is used to compare the medication type information possessed by each piece of the concurrent characteristic data in each of the group categories, and to compare each piece of medication information having the same medication type information in each group category Concurrent feature data is divided into sub-groups in each of the groups;

A reading unit, configured to read the number of data copies having the concurrent feature data in each of the sub-group categories, and the medical expenses corresponding to each of the concurrent feature data;

A forming unit, configured to determine the average medical cost of each subgroup according to the number of data copies and the medical cost corresponding to each of the concurrent characteristic data, to form each of the concurrent type information and each of the medication types Correspondence between the information and the medical expenses in each of the concurrent characteristic data.
The medical expense prediction apparatus according to claim 11, wherein the prediction module further includes a determination unit, the determination unit is configured to:

Reading the concurrent type identifier in the diagnostic information, and comparing the concurrent type identifier with each of the concurrent type information in the corresponding relationship, and determining a target group category corresponding to the diagnostic information;

Reading the target sub-group class in the target group class and the average value of the target medical expenses corresponding to each target sub-type, and determining each average value of the target medical expenses as the target medical treatment corresponding to the diagnosis information cost.
The medical expense prediction apparatus according to claim 12, wherein the prediction module further comprises:

The judging unit is used to judge whether there are multiple types of the target group class, if there are multiple types of the target group class, the number of sub-group classes of the target sub-group class in each of the target group classes is counted;

The determining unit is further configured to determine the average group cost of each target group class based on the number of the sub-group classes and the average value of the target medical expenses of each target sub-group class in each of the target group classes;

The output unit is used to compare the average cost of each group, determine the average value of the target group cost with the largest value, and output the average value of the target group cost to the terminal held by the visiting patient, so that the target group cost The average value is used as a reminder of the leading cost in the target medical cost.
The medical expense prediction apparatus according to claim 11, wherein the classification module is further used to:

When receiving the medical data to be tracked, the concurrent type field and the medication type field in the medical data to be tracked are read;

Comparing the concurrency type field with each of the concurrency type information in the corresponding relationship, and determining a tracking group class corresponding to the data to be tracked;

Comparing the medication type field with the medication type information of each of the sub-group classes in the tracking group class to determine the tracking sub-group class corresponding to the medical data to be tracked;

Reading the average value of the tracking medical expenses corresponding to the tracking subgroup and the medical expenses to be tracked in the medical data to be tracked, and comparing the average value of the tracking medical expenses to the medical expenses to be tracked to determine Describe the anomalies of medical expenses to be tracked.
The medical expense prediction apparatus according to claim 10, wherein the prediction module is further used to:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The medical expense prediction apparatus according to claim 11, wherein the prediction module is further used to:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The medical expense prediction apparatus according to claim 12, wherein the prediction module is further used to:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
The medical expense prediction apparatus according to claim 13, wherein the prediction module is further used for:

Predict the change trend of the disease corresponding to the diagnosis information according to the preset association relationship between each of the preset chronic disease complications;

Comparing the disease change trend and the corresponding relationship, determining the cost change trend corresponding to the disease change trend, and sending the disease change trend and the cost change trend to the terminal held by the visiting patient.
A medical expense prediction device, characterized in that the medical expense prediction device includes: a memory, a processor, a communication bus, and a medical expense prediction program stored on the memory;

The communication bus is used to implement connection communication between the processor and the memory;

The processor is used to execute the medical expense prediction program to implement the steps of the medical expense prediction method according to any one of claims 1 to 9.
A computer-readable storage medium, characterized in that a medical expense prediction program is stored on the computer-readable storage medium, and when the medical expense prediction program is executed by a processor, the program according to any one of claims 1 to 9 is realized Describe the steps of the medical expense prediction method. The