CN107103201B

CN107103201B - Medical navigation path generation method and device and medical path navigation method

Info

Publication number: CN107103201B
Application number: CN201710325176.8A
Authority: CN
Inventors: 邓侃; 李丕勋; 宫海天; 刘旭涛; 李柏松
Original assignee: Beijing RxThinking Ltd
Current assignee: Beijing RxThinking Ltd
Priority date: 2017-05-10
Filing date: 2017-05-10
Publication date: 2020-11-24
Anticipated expiration: 2037-05-10
Also published as: CN107103201A

Abstract

The invention discloses a medical navigation path generation method and device and a medical path navigation method. The generation method of the medical navigation path comprises the following steps: obtaining descriptions of the disease conditions in the historical medical record samples and action schemes related to the descriptions of the disease conditions; determining the dependency relationship between the disease descriptions and the action schemes according to the historical medical record samples; and constructing a medical navigation model according to the dependency relationship, and determining a medical navigation path according to the medical navigation model. According to the technical scheme, the medical navigation model can effectively summarize the diagnosis experience of medical staff in the medical diagnosis process recorded in the historical case samples, summarize various historical diagnosis paths corresponding to various disease descriptions, construct the medical navigation model, and determine the medical navigation path according to the medical navigation model, so that the medical experience can be summarized from the historical case history samples, the medical navigation path can be provided for assisting medical diagnosis, the medical experience can be truly shared, and medical progress is promoted.

Description

Medical navigation path generation method and device and medical path navigation method

Technical Field

The embodiment of the invention relates to the technical field of data processing, in particular to a medical navigation path generation method and device and a medical path navigation method.

Background

The medical record is the literal record of the medical staff in the course of the medical activities of examination, diagnosis, treatment, etc. for the occurrence, development and outcome of the disease of the patient. The medical record is the summary of clinical practice, is the legal basis for exploring disease laws and dealing with medical disputes, and is a precious wealth of the country.

In clinical medicine, the medical records are effectively sorted, and the clinical medical experience of doctors is mined, so that the method has great significance for medical progress. In actual diagnosis and treatment, due to differences in knowledge storage, clinical experience and the like among medical staff, diagnosis modes, medication habits and the like of different medical staff aiming at the same disease or symptom are different, and some medical staff have obvious effect but have little effect. The communication of treatment experience through medical staff in the organization industry not only needs a large amount of manpower and material resources, but also has no real-time property and universal sharing property. Therefore, how to effectively sort out medical experience from medical records and how to realize medical knowledge sharing is very important.

Disclosure of Invention

The embodiment of the invention provides a medical navigation path generation method, a medical navigation path generation device and a medical navigation path generation method, which are used for effectively sorting medical experience from medical records, generating a medical navigation path and realizing sharing of the medical experience.

In a first aspect, an embodiment of the present invention provides a method for generating a medical navigation path, where the method includes:

obtaining descriptions of the disease conditions in the historical medical record samples and action schemes related to the descriptions of the disease conditions;

determining the dependency relationship between the disease descriptions and the action schemes according to the historical medical record samples;

and constructing a medical navigation model according to the dependency relationship, and determining a medical navigation path according to the medical navigation model.

In a second aspect, an embodiment of the present invention further provides an apparatus for generating a medical navigation path, where the apparatus includes:

the sample acquisition module is used for acquiring all disease descriptions in the historical medical record samples and all action schemes related to all disease descriptions;

the dependency relationship determining module is used for determining the dependency relationship between each disease condition description and each action scheme according to the historical medical record samples;

and the medical navigation path determining module is used for constructing a medical navigation model according to the dependency relationship and determining a medical navigation path according to the medical navigation model.

In a third aspect, an embodiment of the present invention further provides a medical treatment path navigation method, where the method includes:

acquiring patient data input by a user; wherein the patient data comprises a current condition description of the patient;

processing the currently input patient data by adopting the medical navigation path established by the medical navigation path generation method according to any embodiment of the invention, and outputting a path navigation result corresponding to the currently input patient data for display; wherein the path navigation result comprises an action scheme or a disease type corresponding to the current disease description of the patient.

In a fourth aspect, an embodiment of the present invention provides a terminal, where the terminal includes:

one or more processors;

a storage device for storing one or more programs,

when the one or more programs are executed by the one or more processors, the one or more processors implement the method for generating a medical navigation path according to any embodiment of the present invention.

In a fifth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the method for generating a medical navigation path according to any embodiment of the present invention.

According to the technical scheme of the embodiment of the invention, medical experience in each historical medical record sample is summarized by acquiring each disease description in the historical medical record sample and each action scheme associated with each disease description, then determining the dependency relationship between each disease description and each action scheme, summarizing the diagnosis experience of medical staff in the medical diagnosis process recorded in the historical medical record sample, then constructing a medical navigation model according to the dependency relationship, summarizing each historical diagnosis path corresponding to each disease description, constructing the medical navigation model, further determining the medical navigation path according to the medical navigation model, effectively collating clinical medical experience from the historical medical record sample, solving the problems that the existing medical experience exchange wastes time and energy, well summarizing the medical experience, providing medical navigation path for assisting medical diagnosis, and really realizing the sharing of the medical experience, promoting the medical progress.

Drawings

In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, a brief description is given below of the drawings used in describing the embodiments. It should be clear that the described figures are only views of some of the embodiments of the invention to be described, not all, and that for a person skilled in the art, other figures can be derived from these figures without inventive effort.

Fig. 1 is a flowchart of a method for generating a medical navigation path according to an embodiment of the present invention;

fig. 2 is a flowchart of a method for generating a medical navigation path according to a second embodiment of the present invention;

fig. 3 is a structural diagram of a medical navigation path generation apparatus according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of a terminal according to a fourth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a flowchart of a method for generating a medical navigation path according to an embodiment of the present invention, where the method may be performed by a device for generating a medical navigation path, where the device may be implemented by hardware and/or software, and may generally be configured in a server independently or a terminal and a server cooperate to implement the method according to this embodiment.

As shown in fig. 1, the method of this embodiment specifically includes:

s110, obtaining each disease condition description in the historical medical record sample and each action scheme related to each disease condition description.

Illustratively, the disease profile may include at least one of: basic information of the patient, vital signs, symptoms, laboratory indexes, examination markers and the like. Specifically, the basic information of the patient may include sex, age, nature of work, home address, and income of economy, etc.; patient vital signs may include height, weight, temperature, pulse, and blood pressure, among others; the patient's symptoms may include a subjective description of the patient's discomfort, which may be expressed in natural language such as dialogue, such as headache, nausea, dizziness, and loss of appetite in a sample of historical medical records; the patient's assay index may include low platelet count, high white blood cell count, high blood glucose concentration, high urine protein count, etc.; the examination markers of the patient can comprise electrocardiogram characteristics, electroencephalogram characteristics, X-ray pictures showing that a certain area is shaded, color ultrasonography showing that a certain area is low-echo nodules, CT showing cord-like shading, PET showing that a certain area is high in local metabolic rate and the like.

Each protocol of action corresponding to a disease description may also be understood as a method or medical means for obtaining a richer description of the disease. The action plan may specifically include questions asked by the doctor, such as "cough or not", etc., or the test and examination items (pathological examination, image examination, etc.) suggested to the patient.

It can be understood that the historical medical record samples include handwritten paper-based medical records, electronic historical medical records, and the like. The historical medical record samples are recorded with disease descriptions and one, two or more action schemes corresponding to the disease descriptions, and the diagnosis results corresponding to the disease descriptions. Wherein the diagnostic result comprises one, two or more disease types. For example, the type of disease associated with a disease description may include one disease, and may also include various complications resulting from the disease, or the patient may themselves suffer from multiple diseases.

In this embodiment, the obtaining of each disease description in the historical medical record samples and each action scheme associated with each disease description may specifically be obtaining each disease description in each historical medical record sample and each action scheme associated with each disease description, respectively, that is, obtaining each disease description in each historical medical record sample and each action scheme associated with each disease description in decibels by taking one historical medical record sample as a unit.

And S120, determining the dependency relationship between the disease descriptions and the action schemes according to the historical medical record samples.

In historical case history samples, the physician will typically give a reasonable treatment plan, i.e., there will be a responsive course of action, in conjunction with the patient's current description of the condition, until the physician can give a diagnosis. Therefore, it is necessary to establish a dependency between each disease description and each action plan. The disease description will be richer and richer along with the increase of the action scheme, and the doctor can give diagnosis results more and more favorably to a certain extent. However, once the current disease description enables the physician to give a reliable diagnosis, there is no action scheme for obtaining a richer disease description in the current disease description. Thus, there is an initial description of the condition in the historical medical record sample, as well as a description of the condition associated with the follow-up course of action.

Generally, in order to better diagnose a patient, a description of the condition of the same patient and an action plan corresponding to the description of the condition are often recorded in the same historical medical record sample. Moreover, there is often a great dependence between the disease description and the action plan of the same patient. Thus, the dependency between each disease description and each behavior plan can be determined from each historical medical record sample. Specifically, the dependency relationship between each disease description and each action plan in the same historical medical record sample can be determined according to the content recorded in each medical record sample. The dependency relationship can be understood as an association relationship between the current disease description, the next action scheme corresponding to the current disease description, and a further disease description obtained according to the action scheme. Establishing a dependency requires determining an association between each disease description and each action profile. This includes the initial patient's dictated disease description, the various movement schemes that go through the physician's diagnosis, and finally a disease description that assists the physician in making a diagnosis.

S130, constructing a medical navigation model according to the dependency relationship, and determining a medical navigation path according to the medical navigation model.

And a medical navigation model is established according to the dependency relationship, and the incidence relationship between each disease condition description and each action scheme can be recorded through the medical navigation model, so that the medical treatment experience in each historical medical record sample can be effectively summarized. The constructing of the medical navigation model according to the dependency relationship may specifically include: obtaining a current disease description; traversing the historical medical record samples according to the dependency relationship, and judging whether an action scheme associated with the current disease description exists; if an action scheme associated with the current disease description exists, acquiring a target disease description corresponding to the current disease description and the action scheme, and taking the target disease description as a next-level disease description of the current disease description; after the target disease description is used as a new current disease description, returning to execute the operation of traversing the historical medical record samples according to the dependency relationship and judging whether an action scheme associated with the current disease description exists or not until the new current disease description is not associated with any action scheme; if the action scheme associated with the current disease description does not exist, recording the disease descriptions of different levels and the action scheme corresponding to the associated upper and lower disease descriptions, and constructing a medical navigation model.

The medical navigation path may be understood as a follow-up course of action and diagnostic results associated with each disease description. It can be understood that the medical navigation model may include one or more navigation paths corresponding to each disease description, and the medical navigation path may be determined according to the medical navigation model, specifically, one or more diagnosis paths corresponding to the current disease description may be selected from the medical navigation model according to a set rule as the medical navigation path; or the medical navigation path in the medical navigation model can be sequenced and displayed according to a set rule, and the medical navigation path is determined according to a selected instruction input by a user.

In an embodiment of the present invention, determining the medical navigation path according to the medical navigation model specifically includes: determining the current action value corresponding to each action scheme according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description; and determining the current optimal action scheme as a current target action scheme according to the current action value, and determining a medical navigation path in the medical navigation model based on each current target action scheme and target income.

According to the technical scheme of the embodiment, medical experience in each historical medical record sample is summarized by acquiring each disease description in the historical medical record sample and each action scheme associated with each disease description, then determining the dependency relationship between each disease description and each action scheme, summarizing the diagnosis experience of medical staff in the medical diagnosis process recorded in the historical medical record sample, then constructing a medical navigation model according to the dependency relationship, summarizing each historical diagnosis path corresponding to each disease description, constructing the medical navigation model, further determining the medical navigation path according to the medical navigation model, so that clinical medical experience can be effectively collated from the historical medical record sample, the problems that the existing medical experience communication is time-consuming and labor-consuming and the like are solved, the summary of the medical experience is well realized, the medical navigation path is provided for assisting medical diagnosis, and the sharing of the medical experience is really realized, promoting the medical progress.

Example two

Fig. 2 is a flowchart of a method for generating a medical navigation path according to a second embodiment of the present invention, as shown in fig. 2, on the basis of the second embodiment, in this embodiment, optionally, the constructing a medical navigation model according to the dependency relationship includes: obtaining a current disease description; traversing the historical medical record samples according to the dependency relationship, and judging whether an action scheme associated with the current disease description exists; if an action scheme associated with the current disease description exists, acquiring a target disease description corresponding to the current disease description and the action scheme, and taking the target disease description as a next-level disease description of the current disease description; after the target disease description is used as a new current disease description, returning to execute the operation of traversing the historical medical record samples according to the dependency relationship and judging whether an action scheme associated with the current disease description exists or not until the new current disease description is not associated with any action scheme; if the action scheme associated with the current disease description does not exist, recording the disease descriptions of different levels and the action scheme corresponding to the associated upper and lower disease descriptions, and constructing a medical navigation model.

On the basis of the foregoing technical solution, further, the determining a medical navigation path according to the medical navigation model may include: determining the current action value corresponding to each action scheme according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description; and determining the current optimal action scheme as a current target action scheme according to the current action value, and determining a medical navigation path in the medical navigation model based on each current target action scheme and target income.

S210, obtaining each disease condition description in the historical medical record sample and each action scheme associated with each disease condition description.

And S220, determining the dependency relationship between the disease descriptions and the action schemes according to the historical medical record samples.

And S230, acquiring the current disease description.

Obtaining the current disease description includes: a description of a condition currently input by a user is received. Illustratively, the current disease description input by the user may be a disease description input by the doctor according to the patient's dictation, or may be a current disease description obtained by the doctor from a patient case sample according to the patient's previous medical record; but also a description of the disease after the physician has made a disease inquiry based on the patient's dictation, etc., which is not limited herein. It is understood that the current description of the medical condition may be the same as the description of the medical condition recorded in the historical medical record sample or may be different from the description of the medical condition obtained in the case sample.

S240, traversing the historical medical record samples according to the dependency relationship, judging whether an action scheme associated with the current disease description exists, and if so, executing S250; otherwise, S270 is executed.

According to the disease description and the action scheme related to the disease description, the disease description at each moment or each stage can be determined, the medical diagnosis path recorded in the historical medical record sample is determined, the medical description and the association relationship between each action scheme and the diagnosis result are summarized, the medical diagnosis experience can be summarized, and a foundation is laid for establishing the medical navigation path. In this embodiment, after the current disease description is obtained, the historical medical record sample where the current disease description is located may be traversed to determine whether there is an action scheme associated with the current disease description, so that the complete diagnosis scheme recorded in the historical medical record sample may be effectively sorted out, and medical experience knowledge may be better summarized

And S250, if an action scheme associated with the current disease description exists, acquiring a target disease description corresponding to the current disease description and the action scheme, and using the target disease description as a next-level disease description of the current disease description.

For example, the initially obtained current disease description may be used as a root node, each action plan associated with the current disease description may be used as each connecting edge of the root node, and each target disease description corresponding to each action plan may be determined as each leaf node of the root node. I.e., each node is a description of the condition and each link is an action plan. If the leaf node has no follow-up connection line, it means that there is no follow-up action scheme after the disease description corresponding to the leaf node.

The operation can determine the next level of disease description of the current disease description by the current disease description and the action scheme related to the current disease description. It is understood that the same medical condition description may exist in different historical medical record samples, and that the same or different action schemes may correspond to the same medical condition description, and thus, the current medical condition description may exist in one or more next levels of medical condition descriptions.

And S260, after the target disease description is used as a new current disease description, returning to execute S240.

In this embodiment, in order to summarize the complete medical treatment plan in the historical medical record sample, the same operation needs to be performed for the description of the disease condition at each level, i.e. it needs to be determined whether a subsequent action plan exists after the description of the disease condition. Therefore, after the target disease description is used as a new current disease description, the operations of traversing the historical medical record sample where the current disease description is located according to the dependency relationship between the disease description and the action plan, and determining whether the action plan associated with the current disease description exists are repeatedly performed to determine whether the current disease description is the last level disease description.

By adopting the method, the disease description of the last level can be gradually determined from the current disease description according to the current disease description and each action scheme, so that a set of complete medical diagnosis method is summarized, and medical experience knowledge in each historical medical record sample is effectively summarized.

S270, recording disease descriptions of different levels and action schemes corresponding to the related upper and lower layer disease descriptions to construct a medical navigation model.

Illustratively, a tree structure can be constructed according to the disease descriptions and the action schemes, and then the constructed tree structure is used as a medical navigation model. Specifically, each disease description may be used as each node of the tree structure, each action scheme may be used as a connecting edge of the tree structure, the initial disease description may be used as a root node, then, the disease description at the next level of the root node is determined as a leaf node of the root node according to each action scheme associated with the root node, the action schemes associated with the root node and each child node are used as connecting edges, and then, the leaf nodes of the leaf nodes are determined as the connecting edges, and so on, each node and connecting edge in the tree structure are determined step by step, and recording is performed, so that a complete tree structure is constructed according to each disease description and each action scheme in each historical medical record sample.

S280, determining the current action value corresponding to each action scheme according to the current income corresponding to each disease description and the future action value corresponding to each lower-layer disease description.

The current benefit can be understood as how well the diagnosis is given based on the current collected description of the condition. In the actual diagnosis process, the richer the disease description is, the better the diagnosis result is, but the more the disease description is, the less the diagnosis result is affected. Consider that the current revenue does not prompt the user what to do next, ask for more symptoms, or have the patient go to some test or examination. Therefore, in order to better determine the current action value corresponding to each action plan, the whole diagnosis process can be considered comprehensively, and not only the current benefit corresponding to each disease description but also the future action value corresponding to each lower-layer disease description corresponding to the disease description need to be considered. For example, an action cost function may be constructed for guiding the physician in the following optimal action plan. Specifically, an action value function may be constructed according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description, and then the current action value corresponding to each action plan may be calculated based on the action value function.

Optionally, the constructing an action cost function according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description includes: in the recorded results, starting from the disease description at the lowest layer, sequentially calculating current action value functions corresponding to the disease descriptions; if the current calculated disease description does not have a lower-layer disease description, determining a current revenue function corresponding to the current calculated disease description according to the current calculated disease description, and taking the current revenue function of the current calculated disease description as a current action value function of the current calculated disease description; and if the current disease description has a lower disease description, adding the current income corresponding to each disease description and the discount of the future action value of the action scheme as a current action value function of the currently calculated disease description. Wherein, the discount of the future action value can be the weighting of the action value corresponding to the next level disease description.

The disease description at the lowest level is recorded as the disease description at the last level, and is also the leaf node at the last level. It should be noted that there is no action plan and no underlying disease description at the leaf node of the last level. In the recorded result, the current action value function corresponding to each disease description is calculated in sequence from the disease description at the lowest layer, and it can be understood that the current action value is calculated from the leaf node at the lowest layer of the tree structure formed by each disease description and each action scheme, and the current action value of the father node is calculated by tracing up in sequence until the current action value of the root node is calculated.

If the current computed disease description does not have a lower disease description, the current computed disease description can be understood as having no action scheme subsequently, namely the current computed disease description is the leaf node of the last level of the current tree structure, the current revenue function corresponding to the current computed disease description is determined according to the current computed disease description, namely the current revenue function of the leaf node of the last level of the current tree structure can be computed, and the current revenue function is taken as the current action value function of the leaf node. Similarly, if the current computed disease description has a lower disease description, it can be understood that the current computed disease description has a subsequent action plan, i.e. a root node or an intermediate node of the tree structure, the current benefit corresponding to the node can be added to the discount of the future action value of the action plan as a function of the current action value of the node.

Illustratively, a description of the condition s at time t is not obtained_tAnd according to the current collected disease description s_tGo through all possible action schemes a corresponding to the disease description_tIs shown as

So as to obtain all possible disease descriptions s at the next moment, namely t +1_t+1Similarly, s_t+1Can be expressed as

Where n represents the total number of disease descriptions at time t + 1.

Suppose at time t-1, for s₁There are 4 action scenarios a₁Description of the disease s₁For "there is a lump in the neck", this time for s₁There are k action scenarios a₁Are respectively as

Such as, for example,

to "cough up", ask the patient for more symptoms;

to perform venous blood test, especially to check the white blood cell count, the test item is recommended;

for 'there is lump in neck, do biopsy, take pathological section', carry on pathological examination;

in order to examine whether the metabolism of the tumor is high or not by making PET on the neck, the image examination of positron emission computed tomography is performed. The diagnosing physician can then follow 4 protocols a₁To pick one to execute.

If, at time t-1, the doctor advises the patient to take an action plan

That is, when there is a lump in the neck, biopsy is taken, and at t 2, the biopsy is reportedCompletion of the disease, description of the disease₂May be added, in which case s₂May be one of the following three cases:

for "there is a lump in the neck, there is no abnormality in the neck biopsy";

for "there is a lump in the neck, there is a benign tumor in the lymphohematopoietic tissue";

it is indicated as "there is a lump in the neck and there is malignant tumor in the lymphohematopoietic tissue". Respectively calculate corresponding to

And

current profit of

And

if the biopsy is done, the result of the biopsy is "lymphohematopoietic malignancy", the entire diagnosis is complete and no follow-up protocol is applied, i.e., a₂Null. Then it can take

Value of (A) as a current disease description

Corresponding current action value

A value of (i), i.e

If the biopsy result is "no abnormality in neck biopsy", then the white blood cell count needs to be checked and verified, i.e. a₂To check the white blood cell count. There are three possibilities for examining the white blood cell count, "normal", "higher" and "lower". The corresponding disease descriptions and benefits are:

the neck part has lumps, the neck part has no abnormality, and the white blood cell count is normal;

it is "neck mass, no abnormality, high white blood cell count";

the disease is marked by the fact that the neck has a lump and no abnormality and the white blood cell count is low. Respectively calculate corresponding to

And

current profit of

And

if the result is "normal white blood cell count" after checking the white blood cell count, the diagnostic process is ended, i.e. a₃Null, then

It should be noted that only when a_tNull, i.e. when there is no follow-up course of action for the current condition, the current condition describes the corresponding current value of action Q(s)_t,a_t＝null)＝r_t。

If the current disease condition is describedWhen the follow-up action scheme is described, the current action value corresponding to the current disease description can be represented by the following formula:

wherein s is_tRepresenting a description of the condition at time t or a description of the t-th condition; a is_tRepresenting the description of the condition at time t or the description of the t-th condition s_tAn associated course of action; q(s)_t,a_t) Representing the current action value corresponding to the disease description at the time t or the t-th disease description; r(s)_t) Representing the description of the condition at time t or the description of the t-th condition s_tThe corresponding current profit; s_t+1Represents the disease description at the moment of t +1 or the description of the t +1 disease;

represents the disease description at the time of t +1 or the t +1 disease description s_t+1J is greater than or equal to 1 and less than or equal to s_tA positive integer for the total number of associated action plans;

represents the disease description at the time of t +1 or the t +1 disease description s_t+1And each action scheme

The maximum value of each corresponding current action value, gamma is

The weight of (a) may be a preset constant, and the specific value may be set according to an actual situation, which is not limited herein.

For a disease description at time t, at the next time t +1, there may be multiple associated action scenarios a_t+1Wherein

In order to ensure that the best diagnostic result is achieved, it is necessary toIn various action schemes

Select the best action scheme so that

Maximization, which can be expressed as

γ is a predetermined constant for expressing the discount rate of future profit. Therefore, the temperature of the molten metal is controlled,

indicating how much the future action value at the next time is equivalent to the current revenue. Current action value corresponding to current disease description

The current action value can be simply expressed as the present income plus the future action value.

In the actual diagnosis process, the economic cost, waiting time and the like of different test items and examination items are different. To better fit the patient's needs, finding the best treatment plan within the patient's acceptable range, an action cost factor may be introduced in calculating the action value. Specifically, constructing the action cost function according to the current income corresponding to each disease condition description and the future action cost corresponding to each lower-layer disease condition description may include: obtaining action cost of current disease description, and constructing a psychological utility function according to the action cost; and adding the current income corresponding to each disease description and the discount of the future action value corresponding to each lower-layer disease description, and subtracting the psychological utility function to construct a current action value function.

For example, it is assumed that the cost of each action is fixed and can be preset. For ease of calculation, the cost of action can be scaled from about 0 to 1. It is not necessarily linear due to the psychological perception of cost to the patient. Thus, a psycho-utility function of action costs can be constructed with a curve shaped like a sigmoid function to more closely depict the patient's psycho-perception of costs. Suppose that when the cost is close to 0, the patient is insensitive to changes in cost; when the cost exceeds a psychological threshold, approaching 1, patient complaints about cost are large. The specific value of the psychological threshold may be set according to the urban area and the situation, and may be 0.5, for example. At this point, it may be stated that at a cost around 0.5, the patient may be most sensitive to changes in cost, while the cost continues to rise and the patient may already be numb.

Illustratively, the psychologic Utility function based on action cost in the invention can be specifically expressed by the following formula:

where α and β are preset constants, Cost (a)_t) And representing the action cost corresponding to the action scheme at the time t. After adding the psychological utility factor of the action cost, the corresponding current action value can be expressed as

Wherein gamma is the current action value at the moment of t +1

The specific value of the weighted value γ may be set according to actual conditions, and is not limited herein.

Determining a current revenue function corresponding to the currently calculated disease condition description according to the currently calculated disease condition description may specifically include: constructing a medical experience summary model according to the disease descriptions in the historical medical record samples and the disease diagnosis results corresponding to the disease descriptions; determining a disease diagnosis result corresponding to the currently calculated disease description based on the medical experience summary model, and calculating a probability distribution of at least one disease; according to the probability distribution of the disease, calculating the entropy of the diagnosis result corresponding to the current calculated disease description, and constructing a current income function according to the entropy.

Specifically, according to each disease description in the historical medical record samples and the disease diagnosis result corresponding to the disease description, the medical experience summary model may be constructed by taking each disease description in the historical medical record samples as input, taking the disease diagnosis result corresponding to the disease description in the historical medical record samples and the probability distribution of each disease included in the diagnosis result as output, training a preset generated confrontation network (GAN), obtaining a constructed medical experience summary model, and summarizing the clinical diagnosis decision experience of a qualified doctor. The generation of the countermeasure network (GAN) includes a Generative model and a discriminant model.

Illustratively, constructing a medical empirical summary model from the disease descriptions in the historical medical record samples and the disease diagnosis corresponding to the disease descriptions may include: training a preset generative model according to target data in the historical medical record samples; generating first forged data according to the trained generative model, and training a preset identification model according to the target data and the first forged data; and generating second forged data according to the trained generative model, adjusting parameters of the generative model according to the discrimination result of the trained discriminative model on the second forged data, and taking the adjusted generative model meeting preset balance conditions as a medical experience summary model. Wherein the target data comprises a description of each condition and a disease diagnosis corresponding to the description of the condition; the first counterfeit data comprises a randomly generated first counterfeit condition description and at least one first counterfeit disease type corresponding to the first counterfeit condition description; the second counterfeit data includes a randomly generated second counterfeit condition description and at least one second counterfeit disease type corresponding to the second counterfeit condition description.

The disease diagnosis result corresponding to the currently calculated disease description is determined based on the medical experience summary model, and the probability distribution of at least one disease is calculated, which may be that the current disease description is obtained first, and the disease description is input into the trained medical experience summary model, the currently input disease description is processed by the medical experience summary, the disease diagnosis result corresponding to the currently calculated disease description is output, and the probability distribution of at least one disease is calculated and displayed.

In this embodiment, the current profit r_tIt is only described whether the diagnosis is focused on a few diseases, but it can also be said that the diagnosis is given according to how well the current collected description of the disease is. Assuming that the smaller the number or the kind of the diseases included in the diagnosis result, the higher the probability of suffering from each disease, the more confident the current diagnosis result is, the current profit r_tThe higher. The embodiment of the invention adopts entropy to express the current income r_t. The entropy corresponding to each disease type in the diagnosis result can be calculated by the following formula:

wherein s is_tRepresenting the disease description collected at the time t; d_tRepresenting the diagnosis result corresponding to the disease description collected at the time t; the diagnostic result may comprise one or more diseases,

representing the jth disease in the diagnosis result corresponding to the disease description collected at the time t;

representing the probability of suffering from the jth disease in the diagnosis; encopy (d)_tSt) represents the entropy of the diagnosis corresponding to the description of the condition collected at time t.

The outputted diagnosis shows that the patient may suffer from five disease types, for example, according to the limited disease description that has been collected, wherein the probability of each disease type is 20%. The Entropy of this diagnosis is Encopy (d)₁|s₁) -1.0 (5 × 0.20 × log0.20) ═ 2.32, where the log is base 2. Subsequently, if more descriptions of the condition are collected, the diagnosis is updated to indicate that the patient is likely to be suffering from the conditionTwo disease types were observed, the probability of each disease type was 20% and 80%, respectively, and the Entropy of the diagnosis was Encopy (d)₂|s₂) -1.0 (0.20 log0.20+0.80 log0.80) to 0.72. In extreme cases, the diagnosis corresponding to the input description of the condition may have only one disease type with a probability of 100%, when the entropy corresponding to the disease type is 0. Therefore, the more the diagnosis result is concentrated on a few disease types, the lower the entropy value is. If the diagnosis result has only one disease, the larger the confidence of the diagnosis result is, the lowest the entropy value is, which is 0.

If the same condition is addressed, we have collected a large number of medical records. Extracting the disease description and the diagnosis result from the medical records, then counting the probability distribution of various related diseases in the diagnosis result of the medical records based on a medical experience summary model, and calculating the entropy of the diagnosis result aiming at the disease description. Wherein, the medical experience summary model outputs the description s of the disease_tProbability pdf (d) of each disease in corresponding diagnosis_t|s_t) Is available

Expressed, further according to pdf (d)_t|s_t) The entropy of the diagnosis can be calculated as the current profit r_t。

It should be noted that the meaning of entropy expression is how well the output diagnosis result is for the disease description. In other words, if the diagnosis results in the historical medical record samples include a plurality of diseases for the same medical condition description, or the diagnosis results in different historical medical record samples are very different, it indicates that the diagnosis of the medical condition description has a large divergence of the diagnosis results and a low confidence in the diagnosis results. But the size of the entropy value does not indicate whether the diagnostic result itself is correct. That is, even if the doctor has a high degree of confidence in the diagnosis result, it cannot be guaranteed that the diagnosis result is always correct.

The value range of the entropy is from 0 to + ∞, and the smaller the entropy is, the larger the profit is. To get r to_tIs limited to a range from 0 to 1Enclose so that r_tThe larger the value is, the larger the current profit is, and r can be set_tConstructed as a function of:

illustratively, a clinical medical treatment path navigation table can be set, and the navigation table can be divided into three columns, wherein the content of each column is s_i、a_iAnd Q(s)_i,a_i) Namely, the disease condition description, the action plan, and the current action value corresponding to the disease condition description and the action plan. The navigation table has a plurality of rows, each corresponding to an s_iAnd a_iCombinations of (a) and (b). Corresponding to the same disease condition description s_iThere may be multiple action schemes a_iSpecifically, a can be_iExpressed as:

where k is the total number of actions (quantity),

representing the kth action scheme corresponding to the ith disease description; in a similar manner to that described above,

the corresponding jth action plan of the ith disease description is shown. Therefore, strictly speaking, the content of each row in the navigation table is s_i、

After a certain amount of historical medical record sample data is collected, the medical experience summary model can be trained firstly based on the historical medical record sample, and the medical navigation table can be filled after the medical experience summary model is trained. Specifically, the disease condition description s can be collated from a medical record_tAnd the diagnostic result d_t. Wherein the diagnosis result d_tIncluding one or more disease types. Recording T as the end point of the diagnosis path, i.e. the current action value Q(s)_T,a_T＝null)＝r(s_T) Wherein a is_TNull means that the whole diagnostic process is over, and there is no follow-up course of action after the corresponding disease is described; the current profit can be formulated as:

then, the calculation formula of the current action value determined by the current income corresponding to each disease description and the future action value of the action scheme can be used

Or considering the psychological factors of the patient, and adding the calculation formula of the current action value after the psychological utility function

Reversely deducing the value of the action at the previous moment from the end point T moment, e.g. end point T moment

Gradually obtaining the current action value corresponding to each disease description until the current action value Q(s) corresponding to the initial disease description is calculated₁,a₁)。

The same treatment is carried out on each historical case sample, the first historical case sample is processed, then the second historical case sample is processed, and Q(s) in the navigation table is updated_t,a_t) Until all historical case samples have been processed. It can be understood that with the development of medicine and the accumulation of clinical medical experience, new historical medical record samples can be continuously obtained to update the medical navigation table.

S290, determining the current optimal action scheme as the current target action scheme according to the current action value, and determining the medical navigation path in the medical navigation model based on each current target action scheme and the target income.

In the embodiment of the present invention, the optimal action plan may be understood as the most appropriate action plan, for example, the action plan may be an action plan that can quickly and accurately determine the diagnosis result by performing the minimum examination, or an action plan that is determined by combining the user's own conditions and is suitable for the user.

In this embodiment, the current action value may be used as a reference, and the currently best action plan may be determined according to the calculated current action value. Specifically, the determining a current optimal action scheme as a current target action scheme according to the current action value may include: and taking the action scheme corresponding to the maximum action value in the calculated current action values as the current target action scheme. I.e. corresponding to the same disease description s_tThere are various action schemes

Each action corresponds to a current action value

Taking each corresponding to multiple action schemes

Maximum of (1), corresponding action scheme

Is the best course of action.

If the current income of the disease condition description of the next level obtained after the action scheme of the next level is added after a plurality of current disease condition descriptions continue is calculated, and the obtained current income is not changed greatly, the influence of the disease condition description of the next level obtained after the action scheme is added on the diagnosis result is probably not large. Therefore, before determining the medical navigation path based on each condition description, the action scheme associated with each condition description and the target benefit, the method may further comprise: calculating current earnings corresponding to the disease descriptions, and taking the current earnings as target earnings if the variation amplitude of the current earnings corresponding to the disease descriptions in a preset number is smaller than a preset threshold value; or calculating the current income corresponding to each disease description, and taking the current income as the target income if the current income corresponding to each disease description with the continuous preset number is converged.

The preset number is to be understood that the number of descriptions of the disease condition may be preset according to actual conditions, and may be, for example, 3, 4, or 5, and is not limited herein. Meanwhile, the preset threshold value may also be set by default according to experience or by a user according to an actual situation, which is not limited herein.

For example, assuming that the initially inputted disease description is denoted as the starting time, T is 1, the final disease description obtained after each action plan is denoted as the diagnosis ending time, T is T, and the current profit r corresponding to each time is calculated during the whole diagnosis process_tSpecifically, the calculation can be performed by the following formula:

wherein, Encopy (d)_t|s_t) Indicating the entropy of each disease type corresponding to the description of the condition at time t, if r_t，r_t+1，…，r_t+ΔIs continuously smaller than a predetermined threshold value, e.g. a predetermined constant τ, the diagnostic process is terminated, at which time T is T + Δ. Where Δ is a time window length of the investigation period, and may be a constant preset according to an actual situation.

It should be noted that the descriptions of "time t" and "time t + 1" in the embodiments of the present invention are only used to distinguish different disease descriptions and action schemes associated with the disease descriptions, and do not represent that the disease descriptions or action schemes may change depending on time, and may be understood as the t-th or t + 1-th disease descriptions.

If the current disease description is different from the disease description recorded in the historical medical record sample, the fact that the disease description does not appear in the historical medical record sample is shown, namely, the past experience can not be referred to, at the moment, the diagnosis result corresponding to the current disease description and the probability of each disease in the diagnosis result can be estimated through a medical experience summary model, and then the current action value corresponding to the current disease description is estimated.

According to the technical scheme of the embodiment, each disease description is divided into different levels through the dependency relationship between each disease description and each action scheme, the disease descriptions of different levels are recorded, the action schemes corresponding to the associated upper-layer disease descriptions and lower-layer disease descriptions are established, a medical navigation model is established, medical experiences in historical medical record samples can be clearly, accurately and orderly summarized, the current income corresponding to each disease description and the future action value corresponding to each lower-layer disease description are combined, the current action value corresponding to each action scheme is determined, each disease description and each associated action scheme are taken as an integral scheme for consideration, the current action value is taken as a reference, the current optimal action scheme is determined, and the optimal medical navigation path is determined.

EXAMPLE III

Fig. 3 is a block diagram of a medical navigation path generation apparatus according to a third embodiment of the present invention. The device can be implemented by means of hardware and/or software, and can be generally configured in a server independently or the terminal and the server cooperate to implement the method of the embodiment. As shown in fig. 3, the medical navigation path generation device of the present embodiment includes: a sample acquisition module 310, a dependency determination module 320, and a medical navigation path determination module 330.

The sample acquiring module 310 is configured to acquire each medical condition description in the historical medical record sample and each action scheme associated with each medical condition description; a dependency relationship determining module 320, configured to determine a dependency relationship between each disease description and each action scheme according to the historical medical record samples; and the medical navigation path determining module 330 is configured to construct a medical navigation model according to the dependency relationship, and determine a medical navigation path according to the medical navigation model.

On the basis of the technical scheme, the medical navigation path determining module can be used for:

obtaining a current disease description;

traversing the historical medical record samples according to the dependency relationship, and judging whether an action scheme associated with the current disease description exists;

if an action scheme associated with the current disease description exists, acquiring a target disease description corresponding to the current disease description and the action scheme, and taking the target disease description as a next-level disease description of the current disease description;

after the target disease description is used as a new current disease description, returning to execute the operation of traversing the historical medical record samples according to the dependency relationship and judging whether an action scheme associated with the current disease description exists or not until the new current disease description is not associated with any action scheme;

if the action scheme associated with the current disease description does not exist, recording the disease descriptions of different levels and the action scheme corresponding to the associated upper and lower disease descriptions, and constructing a medical navigation model.

On the basis of the above technical solutions, the medical navigation path determining module may include a current action value determining unit and a medical navigation path determining unit.

The current action value determining unit is used for determining the current action value corresponding to each action scheme according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description; and the medical navigation path determining unit is used for determining the current optimal action scheme as the current target action scheme according to the current action value and determining the medical navigation path in the medical navigation model based on each current target action scheme and target income.

On the basis of the above technical solutions, the current action value determining unit may include an action value function construction subunit and a current action value operator unit.

The action value function constructing subunit is used for constructing an action value function according to the current income corresponding to each disease description and the future action value corresponding to each lower-layer disease description; and the current action value operator unit is used for calculating the current action value corresponding to each action scheme based on the action value function.

On the basis of the above technical solutions, the action cost function constructing subunit is specifically configured to:

in the recorded results, starting from the disease description at the lowest layer, sequentially calculating current action value functions corresponding to the disease descriptions;

if the current calculated disease description does not have a lower-layer disease description, determining a current revenue function corresponding to the current calculated disease description according to the current calculated disease description, and taking the current revenue function of the current calculated disease description as a current action value function of the current calculated disease description; and if the current disease description has a lower disease description, adding the current income corresponding to each disease description and the discount of the future action value of the action scheme as a current action value function of the currently calculated disease description.

On the basis of the above technical solutions, the determining a current revenue function corresponding to the currently calculated disease description according to the currently calculated disease description includes:

constructing a medical experience summary model according to the disease descriptions in the historical medical record samples and the disease diagnosis results corresponding to the disease descriptions;

determining a disease diagnosis result corresponding to the currently calculated disease description based on the medical experience summary model, and calculating a probability distribution of at least one disease;

according to the probability distribution of the disease, calculating the entropy of the diagnosis result corresponding to the current calculated disease description, and constructing a current income function according to the entropy.

On the basis of the above technical solutions, the medical navigation path determining unit may be specifically configured to:

and taking the action scheme corresponding to the maximum action value in the calculated current action values as the current target action scheme.

On the basis of the above technical solutions, the action cost function constructing subunit may further be configured to:

obtaining action cost of current disease description, and constructing a psychological utility function according to the action cost;

and adding the current income corresponding to each disease description and the discount of the future action value corresponding to each lower-layer disease description, and subtracting the psychological utility function to construct a current action value function.

On the basis of the above technical solutions, the method further comprises:

and the target income determining module is used for calculating the current income corresponding to each disease description before determining a medical navigation path based on each disease description, the action scheme associated with each disease description and the target income, and taking the current income as the target income if the variation amplitude of the current income corresponding to a preset number of disease descriptions is less than a preset threshold value.

The device can execute the methods provided by the first embodiment and the second embodiment of the invention, and has corresponding functional modules and beneficial effects for executing the methods. For details of the technology that are not described in detail in this embodiment, reference may be made to the methods provided in the first embodiment and the second embodiment of the present invention.

In addition, the embodiment also provides a medical treatment path navigation method, which comprises the following steps:

According to the technical scheme, the medical navigation path established by the medical navigation path generation method can output the path navigation result according to the patient data input by the user, can provide a diagnosis path for the user according to the current disease description of the patient and the medical experience, assists the user in diagnosing, fully shares the medical experience, and promotes the medical development.

Example four

Fig. 4 is a schematic structural diagram of a terminal according to a fourth embodiment of the present invention. Fig. 4 illustrates a block diagram of an exemplary terminal 412 suitable for use in implementing embodiments of the present invention. The terminal 412 shown in fig. 4 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. 4, terminal 412 is in the form of a general purpose computing device. The components of the terminal 412 may include, but are not limited to: one or more processors or processors 416; and a storage device 428 for storing one or more programs, the bus 418 connecting the various system components, including the storage device 428 and the processor 416.

Bus 418 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Terminal 412 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by terminal 412 and includes both volatile and nonvolatile media, removable and non-removable media.

Storage 428 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)430 and/or cache memory 432. The terminal 412 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 434 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 4, commonly referred to as a "hard drive"). Although not shown in FIG. 4, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 418 by one or more data media interfaces. Memory 428 can include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 440 having a set (at least one) of program modules 442 may be stored, for instance, in memory 428, such program modules 442 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 442 generally perform the functions and/or methodologies of the described embodiments of the invention.

The terminal 412 may also communicate with one or more external devices 414 (e.g., keyboard, pointing device, display 424, etc.), one or more devices that enable a user to interact with the terminal 412, and/or any devices (e.g., network card, modem, etc.) that enable the terminal 412 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 422. Also, the terminal 412 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) through the network adapter 420. As shown, the network adapter 420 communicates with the other modules of the terminal 412 over a bus 418. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the terminal 412, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processor 416 executes various functional applications and data processing by executing programs stored in the storage device 428, for example, implementing the method for generating a medical navigation path provided by the embodiment of the present invention.

In addition, an embodiment of the present invention further provides a storage medium readable by a computer, having a computer program stored thereon, where the program is used to execute a method for generating a medical navigation path when executed by a processor, and the method includes:

Optionally, the computer executable instructions, when executed by the computer processor, may be further configured to implement a technical solution of the method for generating a medical navigation path according to any embodiment of the present invention.

The logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable storage medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable storage medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer readable storage medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and stored in a computer memory.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware hardware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are well known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a programmable gate array PGA, a field programmable gate array FPGA, or the like.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for generating a medical navigation path, comprising:

obtaining descriptions of medical conditions in a historical medical record sample and various action schemes associated with the descriptions of medical conditions;

constructing a medical navigation model according to the dependency relationship, and determining a medical navigation path according to the medical navigation model;

constructing a medical experience summary model according to the disease descriptions in the historical medical record samples and the disease diagnosis results corresponding to the disease descriptions; determining a disease diagnosis result corresponding to the currently calculated disease description based on the medical experience summary model, and calculating a probability distribution of at least one disease; according to the probability distribution of the disease, calculating the entropy of a diagnosis result corresponding to the current calculated disease description, and constructing a current income function according to the entropy;

the entropy corresponding to each disease type in the diagnosis result is calculated by the following formula:

wherein s is_tRepresenting the disease description collected at the time t; d_tRepresenting the diagnosis result corresponding to the disease description collected at the time t; the diagnostic result comprises one or more diseases,

representing the probability of suffering from the jth disease in the diagnosis; encopy (d)_t,s_t) Representing the entropy of the diagnosis corresponding to the description of the condition collected at time t.

2. The method of claim 1, wherein constructing a medical navigation model from the dependencies comprises:

obtaining a current disease description;

3. The method of claim 2, wherein said determining a medical navigation path according to said medical navigation model comprises:

determining the current action value corresponding to each action scheme according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description;

and determining the current optimal action scheme as a current target action scheme according to the current action value, and determining a medical navigation path in the medical navigation model based on each current target action scheme and target income.

4. The method of claim 3, wherein determining the current action value for each action plan based on the current benefit for each condition description and the future action value for each underlying condition description comprises:

constructing an action value function according to the current income corresponding to each disease condition description and the future action value corresponding to each lower-layer disease condition description;

and calculating the current action value corresponding to each action scheme based on the action value function.

5. The method of claim 4, wherein constructing a cost-of-action function based on the current benefit for each condition description and the future cost-of-action for each underlying condition description comprises:

6. The method according to claim 3, wherein the determining a current optimal action plan as a current target action plan according to the current action value comprises:

7. The method of claim 4, wherein constructing a cost-of-action function based on the current benefit for each condition description and the future cost-of-action for each underlying condition description further comprises:

8. The method of any of claims 3-7, further comprising, prior to said determining a medical navigation path based on each condition description, an action plan associated with each said condition description, and a target benefit:

and calculating the current income corresponding to each disease condition description, and if the variation amplitude of the current income corresponding to the preset number of disease condition descriptions is smaller than a preset threshold value, taking the current income as the target income.

9. An apparatus for generating a medical navigation path, comprising:

the medical navigation path determining module is used for constructing a medical navigation model according to the dependency relationship and determining a medical navigation path according to the medical navigation model;

the current revenue function determining module is used for constructing a medical experience summarizing model according to the disease descriptions in the historical medical record samples and the disease diagnosis results corresponding to the disease descriptions; determining a disease diagnosis result corresponding to the currently calculated disease description based on the medical experience summary model, and calculating a probability distribution of at least one disease; according to the probability distribution of the disease, calculating the entropy of a diagnosis result corresponding to the current calculated disease description, and constructing a current income function according to the entropy;