CN113160954A

Movatterモバイル変換

Info

Publication number: CN113160954A
Application number: CN202110371213.5A
Authority: CN
Inventors: 张金凤
Original assignee: Taikang Insurance Group Co Ltd
Current assignee: Taikang Insurance Group Co Ltd
Priority date: 2021-04-07
Filing date: 2021-04-07
Publication date: 2021-07-23
Anticipated expiration: 2041-04-07
Also published as: CN113160954B

Abstract

The disclosure relates to a medical resource allocation method and device, a storage medium and electronic equipment, and relates to the technical field of machine learning, wherein the method comprises the following steps: acquiring target attribute information, a target insurance portrait and a target health portrait of a target user according to user identification information included in the medical consultation request; inputting preset medical practice portrait, target attribute information, target insurance portrait and target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result; inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result; and matching the corresponding target medical resource for the target user according to the first medical resource distribution result and the second medical resource distribution result. The accuracy of the allocation result of the medical resource is improved.

Description

Medical resource allocation method and device, storage medium and electronic equipment

Technical Field

The embodiment of the invention relates to the technical field of machine learning, in particular to a medical resource allocation method, a medical resource allocation device, a computer-readable storage medium and electronic equipment.

Background

At present, when a user activates online health interests through some specific application programs, the application programs can provide specific signed doctor services for the user, namely, a signed doctor is matched for the user, and the signed doctor can provide long-term stable health consultation inquiry and management services for the user.

Specifically, when the user activates the health benefits, the assigned contracting doctors are mainly general doctors, and when special problems are encountered, the contracting doctors can pull the special doctors into a group to solve the problems.

However, the above allocation scheme has the following disadvantages: on one hand, when medical resources are allocated to the user, the health portrait and the insurance portrait of the user are not considered, and the accuracy of the allocation result of the medical resources is low; on the other hand, the corresponding medical resources cannot be directly matched for the user according to the actual requirements of the user, and the problem of medical resource waste exists.

Therefore, it is desirable to provide a new medical resource allocation method and apparatus.

It is to be noted that the information invented in the above background section is only for enhancing the understanding of the background of the present invention, and therefore, may include information that does not constitute prior art known to those of ordinary skill in the art.

Disclosure of Invention

The present invention is directed to a medical resource allocation method, a medical resource allocation apparatus, a computer-readable storage medium, and an electronic device, which overcome at least some of the problems of low accuracy of medical resource allocation results due to limitations and disadvantages of the related art.

According to an aspect of the present disclosure, there is provided a medical resource allocation method including:

receiving a medical consultation request sent by a target user through terminal equipment, responding to the medical consultation request, and acquiring target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request;

inputting a preset doctor practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result;

acquiring target disease information corresponding to the target insurance portrait, and inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result;

and matching corresponding target medical resources for the target user according to the first medical resource allocation result and the second medical resource allocation result, and sending the target medical resources to the terminal equipment.

In an exemplary embodiment of the present disclosure, the first medical resource allocation model includes a first fully-connected layer composed of a plurality of fully-connected units connected in parallel, a first vector splicing layer, and a first distance calculation layer;

wherein, will predetermine doctor's medical practice portrait, target attribute information, target insurance portrait and target health portrait input to first medical resource distribution model in, obtain first medical resource distribution result, include:

encoding the target attribute information and the target health portrait by using a first full connection unit in the first full connection layer to obtain a first user information code, and encoding the target insurance portrait by using a second full connection unit in the first full connection layer to obtain a first user portrait code;

encoding the preset medical practice portrait by using a third full-connection unit in the first full-connection layer to obtain a first medical practice portrait code;

splicing the first user information code and the first user portrait code by using the first vector splicing layer to obtain a first target user code;

and calculating a first distance between the first target user code and the first practice portrait code by using the first distance calculation layer, and obtaining a first medical resource distribution result according to a first distance calculation result.

In an exemplary embodiment of the present disclosure, the second medical resource allocation model includes a second fully-connected layer composed of a plurality of fully-connected units connected in parallel, a second vector splice layer, and a second distance calculation layer;

inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and the preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result, wherein the method comprises the following steps:

encoding the target attribute information and the target health portrait by using a fourth full-connection unit in the second full-connection layer to obtain a second user information code, and encoding the target insurance portrait by using a fifth full-connection unit in the second full-connection layer to obtain a second user portrait code;

encoding the preset doctor practice portrait by using a sixth full-connection unit in a second full-connection layer to obtain a second practice portrait code, and encoding the target disease information by using a seventh full-connection unit in the second full-connection layer to obtain a disease information code;

splicing the second user information code and the second user portrait code by using the second vector splicing layer to obtain a second target user code, and splicing the disease information code and the second professional portrait code to obtain a medical information code;

and calculating a second distance between the second target user code and the medical information code by using the second distance calculation layer, and obtaining a second medical resource distribution result according to a second distance calculation result.

In an exemplary embodiment of the present disclosure, the medical resource allocation method further includes:

acquiring basic attribute information, a historical insurance portrait, a historical health portrait, a first doctor portrait distributed for the historical user and a first matching degree between the first doctor portrait and the historical user;

generating first sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait and the first doctor portrait, and generating a first sample pair according to the first sample information and the first matching degree;

and training a first network model to be trained by using the first sample to obtain the first medical resource allocation model.

acquiring a second doctor portrait distributed for the historical user, historical disease information corresponding to the historical insurance portrait and a second matching degree between the second doctor portrait and the historical user;

generating second sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait, the historical disease information and the second doctor portrait, and generating a second sample pair according to the first sample information and the second matching degree;

and training a second network model to be trained by using the second sample pair to obtain the second medical resource allocation model.

In an exemplary embodiment of the present disclosure, acquiring target disease information corresponding to the target insurance image includes:

acquiring historical filed data, and extracting disease names and disease symptoms of filed diseases from the filed data;

and constructing an easily-out disease database according to the disease name and the disease symptom of the claim disease, and matching target disease information corresponding to the target insurance picture from the easily-out disease database.

In an exemplary embodiment of the present disclosure, acquiring target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request includes:

acquiring target attribute information of the target user from a user information database according to user identification information included in the medical consultation request; the target attribute information comprises a plurality of names, sexes, ages, cities, practices, industries and annual incomes of the target users;

acquiring the target health portrait from a medical information database according to the user identification information; wherein the target health portrait comprises a plurality of health labels, medical records, current medical history and family history;

acquiring the target insurance image from an insurance information database according to the user identification information; the target insurance portrait comprises a plurality of insurance states, insurance types, guarantee ranges, insurance standard, customer grades, claim settlement times, appearance reasons and claim payment amount.

According to an aspect of the present disclosure, there is provided a medical resource allocation apparatus including:

the system comprises a first information acquisition module, a second information acquisition module and a third information acquisition module, wherein the first information acquisition module is used for receiving a medical consultation request sent by a target user through terminal equipment, responding to the medical consultation request, and acquiring target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request;

the first medical resource allocation module is used for inputting a preset doctor practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result;

the second medical resource allocation module is used for acquiring target disease information corresponding to the target insurance portrait, and inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result;

and the target medical resource matching module is used for matching corresponding target medical resources for the target user according to the first medical resource distribution result and the second medical resource distribution result and sending the target medical resources to the terminal equipment.

In an exemplary embodiment of the present disclosure, the medical resource allocation apparatus may further include:

the second information acquisition module can be used for acquiring basic attribute information, a historical insurance portrait, a historical health portrait, a first doctor portrait distributed for the historical user and a first matching degree between the first doctor portrait and the historical user of the historical user;

a first sample pair generating module, configured to generate first sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait and the first doctor portrait, and generate a first sample pair according to the first sample information and the first matching degree;

the first model training module may be configured to train a first to-be-trained network model by using the first sample, so as to obtain the first medical resource allocation model.

In an exemplary embodiment of the present disclosure, the medical resource allocation apparatus further includes:

a third information acquisition module, configured to acquire a second doctor image assigned to the historical user, historical disease information corresponding to the historical insurance image, and a second matching degree between the second doctor image and the historical user;

the second sample pair generation module can be used for generating second sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait, the historical disease information and the second doctor portrait and generating a second sample pair according to the first sample information and the second matching degree;

the second model training module may be configured to train a second network model to be trained by using the second sample pair, so as to obtain the second medical resource allocation model.

According to an aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a medical resource allocation method as in any one of the above.

According to an aspect of the present disclosure, there is provided an electronic device including:

a processor; and

a memory for storing executable instructions of the processor;

wherein the processor is configured to perform any one of the medical resource allocation methods described above via execution of the executable instructions.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.

Fig. 1 schematically illustrates a flow chart of a medical resource allocation method according to an example embodiment of the present disclosure.

Fig. 2 is a flowchart schematically illustrating a method for inputting a preset medical practice representation, the target attribute information, a target insurance representation, and a target health representation into a first medical resource allocation model to obtain a first medical resource allocation result according to an exemplary embodiment of the present disclosure.

Fig. 3 schematically illustrates a structural example diagram of a first medical resource allocation model according to an example embodiment of the present disclosure.

Fig. 4 is a flowchart schematically illustrating a method for inputting the target disease information, the target attribute information, the target insurance representation, the target health representation and the predetermined medical practice representation into a second medical resource allocation model to obtain a second medical resource allocation result according to an exemplary embodiment of the present disclosure.

Fig. 5 schematically illustrates a structural example diagram of a second medical resource allocation model according to an example embodiment of the present disclosure.

Fig. 6 schematically illustrates a flow chart of a training method of a first medical resource allocation model according to an example embodiment of the present disclosure.

Fig. 7 schematically illustrates a flow chart of a training method of a second medical resource allocation model according to an example embodiment of the present disclosure.

Fig. 8 schematically illustrates a flow chart of another medical resource allocation method according to an example embodiment of the present disclosure.

Fig. 9 schematically illustrates a block diagram of a medical resource allocation apparatus according to an example embodiment of the present disclosure.

Fig. 10 schematically illustrates an electronic device for implementing the medical resource allocation method according to an example embodiment of the present disclosure.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the invention.

Furthermore, the drawings are merely schematic illustrations of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.

In some medical resource allocation schemes, the allocated contracting doctors are mainly general doctors, and when special problems are encountered, the contracting doctors can pull the special doctors into a group to solve the problems. However, general practitioners currently have limited resources, and the resources of specialist physicians are not well utilized.

In addition, the current allocation rule is that when the user activates the right card, allocation is performed according to online/offline, idle/busy and scheduling time of general practitioners; in addition, the current allocation rules do not comprehensively consider data such as insurance portraits and health portraits of clients, practice portraits of doctors, disease knowledge bases and the like, and simply allocate the data according to online/offline, idle/busy and shift time of general doctors, so that the following problems are caused:

on the one hand, the matching accuracy of the user and the doctor cannot be guaranteed. The following are specifically utilized: the quality of the user is unknown and it is also unknown which doctor is assigned to the user. In contrast to customers with very high premium for diabetes, a physician who is very experienced in treating diabetes and who is highly ranked, preferably in the same geographical area as the customer, should be assigned a priority in order to better serve the health of the customer.

On the other hand, the risk prevention capability is low, and the claim settlement factor is not taken into consideration. For example, other diseases can be predicted to be caused by the health condition of a certain patient, and the potentially induced diseases are within the guarantee range of application. If the health management process is not intervened in time, the risk is possibly brought out and the claim is settled.

On the other hand, general practitioner resources are not sufficient, and specialist physician resources are idle. When general practitioners encounter special problems, special doctors are required to enter groups, and service time is slow.

The example embodiment first provides a medical resource allocation method, which may be run on a server, a server cluster or a cloud server; of course, those skilled in the art may also operate the method of the present invention on other platforms as needed, and this is not particularly limited in this exemplary embodiment. Referring to fig. 1, the medical resource allocation method may include the steps of:

step S110, receiving a medical consultation request sent by a target user through terminal equipment, responding to the medical consultation request, and acquiring target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request;

s120, inputting a preset doctor practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result;

s130, acquiring target disease information corresponding to the target insurance portrait, and inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result;

step S140, matching corresponding target medical resources for the target user according to the first medical resource distribution result and the second medical resource distribution result, and sending the target medical resources to the terminal equipment.

Hereinafter, a medical resource allocation method according to an exemplary embodiment of the present disclosure will be explained and explained in detail with reference to the drawings.

First, the objects of the exemplary embodiments of the present disclosure are explained and illustrated. Particularly, the invention mainly solves the problems of dynamic combination and distribution of signed doctor teams provided for users, improves the matching degree of the users and the doctors, and is more suitable for the users, better serves the inquiry consultation and health management of the clients, improves the satisfaction degree of the clients and reduces the claims; meanwhile, doctors who sign the contract for the user are guaranteed to be distributed, good treatment and management can be carried out on the current health condition, potential health problems can be prevented and intervened, and the risk is avoided. Further, according to the treatment and prevention thought, when the client activates the health rights, a doctor who performs treatment and a doctor who performs prevention are dynamically combined from dimensions such as grade, region, expertise, service effect (speed and quality), bearing capacity and the like to form a contracting doctor team according to basic information, insurance portrait and health portrait of the client, medical practice portrait of the doctor and a disease knowledge base, and the contracting doctor team is matched with the client and contracts.

Next, in a medical resource allocation method of the present disclosure, referring to fig. 1:

in step S110, a medical consultation request sent by a target user through a terminal device is received, and target attribute information, a target insurance portrait and a target health portrait of the target user are obtained according to user identification information included in the medical consultation request in response to the medical consultation request.

Specifically, a target user may log in an application program (e.g., tai life) through a terminal device, and the application program may perform real-name authentication on the target user (e.g., requiring the user to fill in a real name, an identification number, a phone number, and the like); when a user carries out medical consultation through the interactive control in the application program, the terminal equipment can generate a medical consultation request according to the real name, the identity card number and the telephone number of the target user and send the medical consultation request to the server side; when the server receives the medical consultation request, the target attribute information, the shepherd insurance portrait and the target health portrait of the target user can be obtained according to the user identification information (such as an identity card number or a telephone number) included in the medical consultation request.

Further, the target attribute information of the target user can be acquired from a user information database according to user identification information included in the medical consultation request; the target attribute information comprises a plurality of names, sexes, ages, cities, practices, industries and annual incomes of the target users; meanwhile, the target health portrait can be obtained from a medical information database according to the user identification information; wherein the target health portrait comprises a plurality of health labels, medical records, current medical history and family history; further, the target insurance picture can be obtained from an insurance information database according to the user identification information; the target insurance portrait comprises a plurality of insurance states, insurance types, guarantee ranges, insurance standard, customer grades, claim settlement times, appearance reasons and claim payment amount.

For example, the name, sex, age, city, occupation, industry, and annual income of the target user may be obtained from a user information database (for example, a customer management system) according to the identification number or telephone number of the target user;

moreover, the health label (health, sub-health, illness, etc.) of the target user, the main diagnosis result in the history online and offline medical record, the current medical history, the past medical history, the family history, and the like can be acquired from the medical information database (which may include the hospital information system and the internet medical system, or may be other systems, which is not particularly limited in this example) and the internet medical system according to the identification number or the telephone number;

further, the insurance status of the target user (insurance status may include non-insurance, under insurance, broken insurance, and settled insurance, etc.), insurance seed, insurance coverage, standard insurance premium (may include the total number of insurance premiums of the currently in-effect insurance policy), bowling, level (user level of the target user), whether the target user has the possibility of secondary development (specifically, may include high, general, low, no, etc.), number of settlement of claims, reason of insurance, amount of claims, and the like may be acquired from an insurance information database (which may be an insurance management system, for example) according to the identification number or the telephone number.

In step S120, a preset medical practice portrait, the target attribute information, the target insurance portrait, the target health portrait and the preset medical practice portrait are input into a first medical resource allocation model to obtain a first medical resource allocation result.

In this exemplary embodiment, first, a preset doctor practice portrait may be obtained from the doctor resource system, and specifically, the doctor practice portrait may include a doctor practice department, a doctor expertise, a doctor level (a specialist level, a general level, and the like), a first practice hospital of the doctor, a practice status, a doctor main diagnosis with the highest score in patient evaluation, a comprehensive evaluation, an average response market, whether online, busy, and shift time.

Furthermore, after the preset doctor practice portrait is obtained, the preset doctor practice portrait, the target attribute information, the target insurance portrait and the target can be input into the first medical resource allocation model, so that a first medical resource allocation result is obtained. The first medical resource allocation model comprises a first full connection layer, a first vector splicing layer and a first distance calculation layer, wherein the first full connection layer is composed of a plurality of full connection units which are connected in parallel.

Specifically, referring to fig. 2, the step of inputting a preset medical practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result may include the following steps:

step S210, encoding the target attribute information and the target health portrait by using a first full-connection unit in the first full-connection layer to obtain a first user information code, and encoding the target insurance portrait by using a second full-connection unit in the first full-connection layer to obtain a first user portrait code;

step S220, encoding the preset medical practice portrait by using a third full-connection unit in the first full-connection layer to obtain a first medical practice portrait code;

step S230, splicing the first user information code and the first user portrait code by using the first vector splicing layer to obtain a first target user code;

step S240, calculating a first distance between the first target user code and the first practice portrait code by using the first distance calculation layer, and obtaining a first medical resource allocation result according to a first distance calculation result.

Hereinafter, steps S210 to S240 will be explained and explained with reference to fig. 3.

First, referring to fig. 3, the first medical resource allocation model may include a first fully-connected layer 310 composed of afirst input layer 300, a plurality of fully-connected units connected in parallel (a first fully-connectedunit 311, a second fully-connectedunit 312, and a third fully-connected unit 313), a firstvector splicing layer 320, a firstdistance calculation layer 330, and afirst output layer 340. Each full-connection unit is connected with the first input layer and the first vector splicing layer respectively, and the first vector splicing layer is connected with the first output layer through the first distance calculation layer.

Based on this, it can be known that the specific calculation process of the first medical resource allocation result may include: firstly, in a user information full-connection layer (a first full-connection unit), each network unit can simultaneously process user basic information (namely target attribute information) and a target health picture of an input layer, and further obtain a first user information code; on the insurance portrait full connection layer, each network unit can simultaneously process the target insurance portrait of the input layer, and further obtain a first user portrait code (insurance information code); meanwhile, in a doctor information full-connection layer (a third full-connection unit), each network unit can process all information (doctor practice portrait) of a doctor so as to obtain a first practice portrait code; secondly, splicing the first user information code and the second user portrait code on a first splicing layer to obtain a first target user code; and then, in a first distance calculation layer, distance calculation is carried out on the first target user code and the first license portrait code in a full-connection mode, finally, a value between 0 and 1 is output by using a sigmoid activation function, and then the maximum value is selected as a first medical resource distribution result.

In step S130, target disease information corresponding to the target insurance portrait is obtained, and the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait are input into a second medical resource allocation model to obtain a second medical resource allocation result.

In the present exemplary embodiment, first, the disease information corresponding to the insurance image is acquired. The method specifically comprises the following steps: firstly, acquiring historical settled data, and extracting disease names and disease symptoms of settled diseases from the settled data; secondly, an easily-out disease database is constructed according to the disease name and the disease symptom of the claim disease, and target disease information corresponding to the target insurance picture is matched from the easily-out disease database. For example, the disease name of the claim disease and the corresponding symptom information can be extracted from the claim data in the last year or 3 years, and then the extracted disease name and the corresponding symptom information are collated to obtain the easy-to-go disease database; and further, corresponding target disease information can be matched from the easy-to-risk disease database according to the insurance type and the guarantee range in the target insurance portrait. It should be noted that, the "health problem prone to occur" of the client is determined by comparing the potential diseases of the client with the parameters of the insurance type, the guarantee range, the reason for occurrence and the like in the insurance portrait, and the "health problem prone to occur" is matched with the parameters of the doctor's office of employment, the doctor's expertise, the condition of employment, the diagnosis of the patient with the highest score in the evaluation of the patient, the comprehensive evaluation and the like in the doctor's office of employment portrait, so as to find out the doctor who can prevent and intervene the potential health problem of the client, provide services for the client, track the health problem in time, prevent the delay, avoid occurrence of risks and reduce claims.

Secondly, after the target disease information is acquired, the target disease information, the target attribute information, the target insurance portrait, the target health portrait and the preset doctor practice portrait can be input into a second medical resource allocation model to obtain a second medical resource allocation result. The second medical resource allocation model comprises a second full-connection layer, a second vector splicing layer and a second distance calculation layer, wherein the second full-connection layer is composed of a plurality of full-connection units which are connected in parallel.

Specifically, referring to fig. 4, the step of inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and the preset medical practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result may include the following steps:

step S410, encoding the target attribute information and the target health portrait by using a fourth full-connection unit in the second full-connection layer to obtain a second user information code, and encoding the target insurance portrait by using a fifth full-connection unit in the second full-connection layer to obtain a second user portrait code;

step S420, encoding the preset doctor practice portrait by using a sixth full connection unit in a second full connection layer to obtain a second practice portrait code, and encoding the target disease information by using a seventh full connection unit in the second full connection layer to obtain a disease information code;

step S430, splicing the second user information code and the second user portrait code by using the second vector splicing layer to obtain a second target user code, and splicing the disease information code and the second professional portrait code to obtain a medical information code;

step S440, calculating a second distance between the second target user code and the medical information code by using the second distance calculation layer, and obtaining a second medical resource allocation result according to a second distance calculation result.

Hereinafter, steps S410 to S440 will be explained and explained with reference to fig. 5.

First, referring to fig. 5, the second medical resource allocation model may include a second fully-connected layer 510 composed of asecond input layer 500, a plurality of fully-connected units connected in parallel (a fourth fully-connectedunit 511, a fifth fully-connectedunit 512, a sixth fully-connectedunit 513, and a seventh fully-connected unit 514), a secondvector splicing layer 520, a seconddistance calculation layer 530, and asecond output layer 540. Each full-connection unit is connected with the second input layer and the second vector splicing layer respectively, and the second vector splicing layer is connected with the second output layer through the second distance calculation layer.

Based on this, it can be known that the specific calculation process of the second medical resource allocation result may include: firstly, in a client information full-connection layer (a fourth full-connection unit), each network unit can simultaneously process target attribute information and a target health portrait of an input layer so as to obtain a second user information input code; in the insurance portrait full-link layer (fifth full-link unit), each network unit can simultaneously process the target insurance portrait of the input layer, and then a second user portrait code is obtained; in the doctor information full-connection layer (sixth full-connection unit), each network unit can process all information (doctor practice portrait) of the doctor, and then second practice portrait codes are obtained; in a disease library full-connection layer (a seventh full-connection unit), each network unit can simultaneously process and process target disease information of an input layer so as to obtain a disease information code; secondly, splicing a second user information code and a second user portrait code together on a second splicing layer to obtain a second target user code, and splicing a disease information code and a second professional portrait code together to obtain a medical information code; further, in a second distance calculation layer, distance calculation is carried out on a second target user code and a medical information code which are connected together in a full connection mode, a sigmoid activation function is finally used for outputting a value between 0 and 1, and then the maximum value is selected as a second medical resource distribution result.

It should be noted that the difference between the first medical resource allocation model and the second medical resource allocation model is that one more fully connected unit is added to the second medical resource allocation model, and the first medical resource allocation model and the second medical resource allocation model have different purposes, so the included parameters are different; therefore, the encoding result obtained by encoding each input information by the first medical resource allocation model is different from the encoding result obtained by encoding each input information by the second medical resource allocation model.

In step S140, according to the first medical resource allocation result and the second medical resource allocation result, matching a corresponding target medical resource for the target user, and sending the target medical resource to the terminal device.

Specifically, after the first medical resource allocation result and the second medical resource allocation result are obtained, the first medical resource allocation result and the second medical resource allocation result can be integrated, and then the final target medical resource can be obtained, the target medical resource can be a doctor team, doctors in the doctor team can perform treatment management on the current medical consultation request of the target user, and can also perform prevention and intervention on the potential medical risk of the target user, so that the problem of avoiding danger is solved.

Hereinafter, a specific training procedure of the first medical resource allocation model and the second medical resource allocation model according to the exemplary embodiment of the present disclosure will be explained and explained with reference to fig. 6 and fig. 7.

First, referring to fig. 6, a specific training process of the first medical resource allocation model may include the following steps:

step S610, acquiring basic attribute information, a historical insurance portrait, a historical health portrait, a first doctor portrait distributed for the historical user, and a first matching degree between the first doctor portrait and the historical user of the historical user.

Step S620, generating first sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait and the first doctor portrait, and generating a first sample pair according to the first sample information and the first matching degree.

Step S630, training the first to-be-trained network model by using the first sample, so as to obtain the first medical resource allocation model.

Hereinafter, steps S610 to S630 will be explained and explained. First, data samples for training model 1 (first medical resource allocation model) are collected, each sample being a data pair (first sample pair) in the form of x, y. Wherein x is a multi-dimensional vector including basic attribute information of a certain user, health information (historical health portrait), historical insurance portrait of the historical user, and information (first doctor portrait) of a certain doctor assigned to the historical user; specifically, the first doctor portrait needs to select a doctor suitable for the current health condition and quality condition of the client (obtained by indexes such as insurance allowance of the insurance portrait); then, normalizing each dimension information of the historical user, each dimension information of the insurance portrait and each dimension information of the doctor, and converting the normalized information into a number between 0 and 1; where y is a real number, a value of 0 indicates that the client is not reasonably assigned to the doctor, and a value of 1 indicates that the client is reasonably assigned to the doctor.

Further, the model is trained based on the first sample until the accuracy of the first network model to be trained does not substantially change. In the training process, a first sample pair can be divided into a training set, a verification set and a test set, and the specific proportion can be 7:2: 1; moreover, binary cross entropy can be adopted as a loss function, and a random gradient descent method with momentum can be adopted as an optimization method; also, some regularization constraints may be added to avoid overfitting (such as L2 regularization); training on a training set, and gradually increasing the number of training rounds until the accuracy on the verification data set does not change basically; finally, the test is carried out on the test set, and the test result can be used as the real accuracy of the on-line service in the future.

Next, referring to fig. 7, a specific training process of the second medical resource allocation model may include the following steps:

step S710, acquiring a second doctor portrait distributed for the historical user, historical disease information corresponding to the historical insurance portrait and a second matching degree between the second doctor portrait and the historical user;

step S720, generating second sample information according to the basic attribute information, the historical insurance portrait, the historical health portrait, the historical disease information and the second doctor portrait, and generating a second sample pair according to the first sample information and the second matching degree;

step S730, training a second network model to be trained by using the second sample pair, so as to obtain the second medical resource allocation model.

Hereinafter, steps S710 to S730 will be explained and explained. Specifically, for a second sample pair of the training model 2 (a second medical resource allocation model), where x is a multi-dimensional vector including basic attribute information of a historical user, a historical health representation, a historical insurance representation, historical disease information, and a second doctor representation; wherein the matched second doctor profile is to select a doctor suitable for the potential health condition (such as vulnerable disease information) of the historical user; meanwhile, normalizing all dimension information of the historical user, all dimension information of the insurance portrait, all dimension information of the disease library and all dimension information of the second doctor portrait to be a number between 0 and 1; y is a real number, a value of 0 indicates that the historical user is not reasonably assigned to the doctor, and a value of 1 indicates that the historical user is reasonably assigned to the doctor.

Further, dividing the second sample pair into a training set, a verification set and a test set, wherein the specific proportion can be 7:2: 1; and then training the second network model to be trained by using the second sample until the accuracy of the model is basically not changed any more. Specifically, binary cross entropy may be used as a loss function, a stochastic gradient descent method with momentum may be used as an optimization method, and some regularization constraints may be added to avoid overfitting (such as L2 regularization). Training on a training set, and gradually increasing the number of training rounds until the accuracy on the verification data set does not change basically; finally, the test is carried out on the test set, and the test result can be used as the real accuracy of the on-line service in the future.

Hereinafter, the medical resource allocation method according to the exemplary embodiment of the present disclosure will be further explained and explained with reference to fig. 8. Referring to fig. 8, the medical resource allocation method may include the steps of:

step S801, inquiring basic information (name, gender, age and city) of a user;

step S802, inquiring insurance portrait (insurance state, standard insurance fee, bowling, level) of user;

step S803, inquiring the user health portrait (health label, main diagnosis in the historical online and offline medical record, current medical history in the health file and family history);

step S804, inquiring the doctor practice portrait (doctor practice department, doctor specialty, doctor diagnosis of the doctor with the highest score in patient evaluation, doctor level, first practice hospital of the doctor and practice status), finding out the doctor with higher matching degree with the user, and then distributing according to the online/offline, idle/busy and shift time of the doctor;

step S805, after the dimensionalities of the user basic information, the health information, the user insurance portrait information and the doctor information are coded, inputting a first medical resource allocation model, and acquiring a doctor with the highest matching degree as a doctor 1 matched with the current health condition of the user;

step S806, after the dimensionalities of the user basic information, the health information, the user insurance portrait information, the disease information and the doctor information are coded, inputting a second medical resource allocation model, and acquiring a doctor with the highest matching degree as a doctor 2 matched with the potential health condition of the user;

in step S807, doctor 1 and doctor 2 are regarded as a contracted doctor group for the user.

Specifically, for example, a user YY, a man, a 40 year old, a city in which the user is located, mart and han, the IT industry, a general manager of a private company, the annual income of the user is about 80 ten thousand, in the middle of the life, the insurance of health insurance (115 serious diseases and 60 mild diseases in the guarantee range), the long-term accident insurance, the life insurance and the like are ensured, the standard insurance fee is 60 ten thousand, the bowling is 10 years, the grade is honored, the secondary development possibility is high, the settlement times is 0, the reason for the insurance is not available, and the payment amount is 0.

The health label is allergic, the tobacco age is five years, allergic rhinitis ranks first in the historical diagnosis of the clinic, allergic rhinitis exists in the current medical history, and father has asthma in the family history.

When the user carries out real-name authentication in Thai life and activates health rights and interests, the system combines comprehensive calculation in a first medical resource allocation model according to the basic information of the user, the health portrait and the insurance portrait of the user and the practice portrait of a doctor to screen out a doctor in King. (the doctor in king has the highest degree of matching with the user in model 1, the degree of matching is 0.92, and the highest degree is 1).

The Wang doctor specializes in treating allergic rhinitis, the patient with the highest score in the evaluation is also the treatment of allergic rhinitis, the Wang doctor is the chief physician, the first executive hospital is Taikang Hospital (Wuhan), the department of the department is the department of otorhinolaryngology, and the state of the department is in use. And comprehensively evaluating five stars. The average response time is 30 seconds, when YY activates the health right, the king doctor is online and shows idle, and the shift is scheduled on the day.

The system screens out a certain doctor in the model 2 according to the basic information of the user, the health image and the insurance image of the user, the practice image of the doctor and the disease knowledge base. (Zhang a doctor in model 2 with the highest match to the user, the match was 0.9 and the highest was 1).

One doctor is a subsidiary principal physician of respiratory medicine in Taikang Tongji (Wuhan) hospital, because severe asthma exists in the scope of guarantee in the severe risk purchased by the user YY, asthma exists in the family history of the user YY, the user YY is allergic constitution and smokes for 5 years, although allergic rhinitis exists, the possibility of inducing asthma is high, and an asthma specialist is combined with the system to join a contract group. The health change condition of the user can be known in time, and the health guidance and intervention can be performed on the user in time, so that the induction of asthma is avoided; finally, a doctor of king and a doctor of Zhang form a contracting doctor team, and match and contract with the user.

The medical resource allocation model provided by the embodiment of the disclosure not only ensures doctors who sign with the user allocation, but also can perform good treatment management on the current health condition, improve the health service quality and improve the user satisfaction; moreover, doctors who sign with the user can be guaranteed to prevent and intervene potential health problems, and risks are avoided, so that claim settlement is reduced, and the purpose of fully integrating and optimally configuring doctor resources is achieved; furthermore, as the insurance portrait of the user is increased, the user quality can be judged according to the insurance portrait, the insurance state, the standard insurance premium, the bowling, the grade, the secondary development possibility, the claim settlement times and the claim amount parameters, and the user quality is judged to act on the selection and interpretation of the grade and the service effect dimension of the doctor, so that the high-quality user can be matched with the high-quality doctor, the health service quality is ensured, and the user experience is further improved.

The embodiment of the disclosure also provides a medical resource distribution device. Referring to fig. 9, the medical resource allocation apparatus may include a firstinformation acquisition module 910, a first medicalresource allocation module 920, a second medicalresource allocation module 930, and a target medicalresource matching module 940.

Wherein:

the firstinformation obtaining module 910 may be configured to receive a medical consultation request sent by a target user through a terminal device, and obtain target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request in response to the medical consultation request;

the first medicalresource allocation module 920 may be configured to input a preset doctor practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result;

the second medicalresource allocation module 930 may be configured to obtain target disease information corresponding to the target insurance portrait, and input the target disease information, the target attribute information, the target insurance portrait, the target health portrait, and a preset doctor practice portrait into the second medical resource allocation model to obtain a second medical resource allocation result;

the target medicalresource matching module 940 may be configured to match a corresponding target medical resource for the target user according to the first medical resource allocation result and the second medical resource allocation result, and send the target medical resource to the terminal device.

The specific details of each module in the medical resource allocation apparatus have been described in detail in the corresponding medical resource allocation method, and therefore are not described herein again.

It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the invention. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.

Moreover, although the steps of the methods of the present invention are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.

In an exemplary embodiment of the present invention, there is also provided an electronic device capable of implementing the above method.

As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.

An electronic device 600 according to this embodiment of the invention is described below with reference to fig. 10. Theelectronic device 1000 shown in fig. 10 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 10, theelectronic device 1000 is embodied in the form of a general purpose computing device. The components of theelectronic device 1000 may include, but are not limited to: the at least oneprocessing unit 1010, the at least onememory unit 1020, abus 1030 connecting different system components (including thememory unit 1020 and the processing unit 1010), and adisplay unit 1040.

Wherein the storage unit stores program code that is executable by theprocessing unit 1010 to cause theprocessing unit 1010 to perform steps according to various exemplary embodiments of the present invention as described in the "exemplary methods" section above in this specification. For example, theprocessing unit 1010 may execute step S110 as shown in fig. 1: receiving a medical consultation request sent by a target user through terminal equipment, responding to the medical consultation request, and acquiring target attribute information, a target insurance portrait and a target health portrait of the target user according to user identification information included in the medical consultation request; step S120: inputting a preset doctor practice portrait, the target attribute information, a target insurance portrait and a target health portrait into a first medical resource allocation model to obtain a first medical resource allocation result; step S130: acquiring target disease information corresponding to the target insurance portrait, and inputting the target disease information, the target attribute information, the target insurance portrait, the target health portrait and a preset doctor practice portrait into a second medical resource allocation model to obtain a second medical resource allocation result; step S140: and matching corresponding target medical resources for the target user according to the first medical resource allocation result and the second medical resource allocation result, and sending the target medical resources to the terminal equipment.

Thestorage unit 1020 may include readable media in the form of volatile memory units, such as a random access memory unit (RAM)10201 and/or acache memory unit 10202, and may further include a read-only memory unit (ROM) 10203.

Thememory unit 1020 may also include a program/utility 10204 having a set (at least one) ofprogram modules 10205,such program modules 10205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

Bus 1030 may be any one or more of several types of bus structures including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, and a local bus using any of a variety of bus architectures.

Theelectronic device 1000 may also communicate with one or more external devices 1100 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with theelectronic device 1000, and/or with any devices (e.g., router, modem, etc.) that enable theelectronic device 1000 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interfaces 1050. Also, theelectronic device 1000 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the internet) via thenetwork adapter 1060. As shown, thenetwork adapter 1060 communicates with the other modules of theelectronic device 1000 over thebus 1030. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with theelectronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to make a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) execute the method according to the embodiment of the present invention.

In an exemplary embodiment of the present invention, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above section "exemplary methods" of the present description, when said program product is run on the terminal device.

According to the program product for realizing the method, the portable compact disc read only memory (CD-ROM) can be adopted, the program code is included, and the program product can be operated on terminal equipment, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).

Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.

Claims

1. A medical resource allocation method, comprising:

2. The medical resource allocation method according to claim 1, wherein the first medical resource allocation model includes a first fully-connected layer composed of a plurality of fully-connected units connected in parallel, a first vector splicing layer, and a first distance calculation layer;

3. The medical resource allocation method according to claim 1, wherein the second medical resource allocation model includes a second fully-connected layer composed of a plurality of fully-connected units connected in parallel, a second vector splice layer, and a second distance calculation layer;

4. The medical resource allocation method according to claim 1, characterized in that the medical resource allocation method further comprises:

5. The medical resource allocation method according to claim 4, characterized in that the medical resource allocation method further comprises:

6. The medical resource allocation method according to claim 1, wherein acquiring target disease information corresponding to the target insurance image includes:

7. The medical resource allocation method according to claim 1, wherein acquiring target attribute information, a target insurance representation and a target health representation of the target user according to user identification information included in the medical consultation request comprises:

8. A medical resource allocation apparatus, comprising:

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the medical resource allocation method according to any one of claims 1 to 7.

10. An electronic device, comprising:

a processor; and

a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the medical resource allocation method of any one of claims 1-7 via execution of the executable instructions.