WO2021179625A1

WO2021179625A1 - Smart diagnosis guide method and apparatus, electronic device, and storage medium

Info

Publication number: WO2021179625A1
Application number: PCT/CN2020/124260
Authority: WO
Inventors: 李彦轩; 刘卓; 孙行智
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-09-27
Filing date: 2020-10-28
Publication date: 2021-09-16
Also published as: CN111951943A; CN111951943B

Abstract

A smart diagnosis guide method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining at least one symptom of a patient, said at least one symptom including the symptom obtained during the present round of symptom inquiry (S21); obtaining a department recommendation result corresponding to the patient according to the at least one symptom (S22); obtaining a feedback of the department recommendation result (S23); according to the feedback, determining the symptom obtained during the present round of symptom inquiry as the target symptom (S24); performing a next round of symptom inquiry on the basis of the target symptom and repeating multiple rounds of symptom inquiries until a recommended department can be responded to the patient (S25). The present method is conducive to reducing the number of rounds of interaction with a patient during diagnosis guide, and is conducive to enhancing the accuracy of diagnosis guide.

Description

Intelligent triage method, device, electronic equipment and storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on September 27, 2020, the application number is 202011031139.4, and the invention title is "intelligent triage method, device, electronic equipment and storage medium", the entire content of which is incorporated by reference Incorporated in this application.

Technical field

This application relates to the field of artificial intelligence technology, and in particular to an intelligent triage method, device, electronic equipment and storage medium.

Background technique

Patient triage is an important task in the daily work of the hospital. Fast and accurate triage can save patients' time and improve their experience. Traditional triage methods are mostly manual triage. Because of the different levels of hospitals in different regions, the level of medical staff is also uneven, so the accuracy of manual triage is not the same, and it requires a certain amount of manpower input. In this context , Intelligent triage was born. The inventor realizes that the current intelligent triage often uses a reinforcement learning model to determine the number of specific consultation rounds. The existing reinforcement learning methods do not distinguish between patient symptoms and symptoms that are not helpful for department recommendations. It is important to consider. In addition, setting both symptom inquiry and department recommendation as the learning goal of the reinforcement learning model makes the model learning task more complicated, which will affect the accuracy of triage.

Summary of the invention

In response to the above-mentioned problems, the present application provides an intelligent triage method, device, electronic equipment, and storage medium, which is beneficial to improve the accuracy of triage.

To achieve the foregoing objective, the first aspect of the embodiments of the present application provides an intelligent triage method, which includes:

Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

Obtaining a department recommendation result corresponding to the patient according to the at least one symptom;

Obtain feedback on the recommendation result of the department;

According to the feedback, determine the symptom obtained in this round of symptom inquiry as the target symptom;

The next round of symptom inquiry is performed based on the target symptom, and multiple rounds of symptom inquiry are repeated until the patient returns to the recommended department.

The second aspect of the embodiments of the present application provides an intelligent triage device, which includes:

The symptom acquisition module is used to acquire at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

The department recommendation module is used to obtain the department recommendation result corresponding to the patient according to the at least one symptom;

Feedback acquisition module, used to acquire feedback on the recommendation result of the department;

The symptom determination module determines the symptom obtained in this round of symptom inquiry as the target symptom according to the feedback;

The department output module is configured to perform the next round of symptom inquiry based on the target symptom, and repeatedly execute multiple rounds of symptom inquiry until returning to the recommended department to the patient.

The third aspect of the embodiments of the present application provides an electronic device. The electronic device includes an input device and an output device, and also includes a processor, adapted to implement one or more instructions; and, a computer storage medium, which stores There are one or more instructions, and the one or more instructions are suitable for being loaded by the processor and executing the following steps:

Obtain feedback on the recommendation result of the department;

The fourth aspect of the embodiments of the present application provides a computer storage medium, the computer storage medium stores one or more instructions, and the one or more instructions are suitable for being loaded by a processor and executing the following steps:

Obtain feedback on the recommendation result of the department;

This application not only helps reduce the number of rounds of symptom inquiry, but also improves the accuracy of triage.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

FIG. 1 is a schematic diagram of a network system architecture provided by an embodiment of this application;

2 is a schematic flowchart of an intelligent triage method provided by an embodiment of this application;

FIG. 3 is an example diagram of a hierarchy of a reinforcement learning model provided by an embodiment of this application;

FIG. 4 is an example diagram of interaction between a reinforcement learning model and an environment provided by an embodiment of this application;

FIG. 5 is a schematic flowchart of another intelligent triage method provided by an embodiment of the application;

FIG. 6 is a schematic structural diagram of an intelligent triage device provided by an embodiment of the application;

FIG. 7 is a schematic structural diagram of an electronic device provided by an embodiment of the application.

Detailed ways

In order to enable those skilled in the art to better understand the solutions of the application, the technical solutions in the embodiments of the application will be clearly and completely described below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only These are a part of the embodiments of this application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work should fall within the protection scope of this application.

The terms "including" and "having" appearing in the specification, claims, and drawings of this application and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes unlisted steps or units, or optionally also includes Other steps or units inherent to these processes, methods, products or equipment. In addition, the terms "first", "second", "third", etc. are used to distinguish different objects, but not to describe a specific sequence.

The technical solution of this application can be applied to the fields of artificial intelligence, digital medical care, smart city, blockchain and/or big data technology to improve the accuracy of triage and realize smart medical care.

The embodiment of the application provides an intelligent triage method, which can be implemented based on the network system architecture shown in FIG. 1. Please refer to FIG. 1. The network system architecture includes terminals and electronic devices. The terminals and electronic devices pass through a wired or wireless network For communication connection, the terminal is the terminal device used by the patient, which can be the patient’s mobile phone, tablet, computer, personal digital assistant (PDA), etc., or the intelligent terminal device set up by the hospital for triage. The patient passes The terminal interacts with the electronic device, for example: submitting a description of the symptoms to the electronic device. The electronic equipment includes at least an enhanced learning module, a triage module, and a communication module. The communication module is integrated with a digital protocol interface. The communication module obtains the information input by the patient through the terminal through the digital protocol interface. After receiving the information submitted by the terminal, the electronic device will The description of symptoms is sorted into structured symptom information. The reinforcement learning module collects and updates the patient’s symptom information after multiple rounds of interaction with the patient. The triage module predicts the probability of recommending departments based on the patient’s symptom information. The reinforcement learning module can The triage module obtains the feedback (reward) for performing the department recommendation task, and based on the value of the feedback, determines the distinguished symptoms, and gives priority to the symptom inquiry for the distinguished symptoms, until the collected information can output the final recommendation to the patient Department. The electronic device separates the department recommendation task from the reinforcement learning module, and is executed by the triage module recommended by the department alone, which simplifies the overall task of the reinforcement learning module. In addition, it is effective to determine the distinguished symptoms from the numerous symptoms. Accurate triage can still be completed while reducing the number of rounds of interaction with the patient.

Based on the network system architecture shown in FIG. 1, the intelligent triage method provided by the embodiments of the present application will be described in detail below in conjunction with other drawings.

Please refer to FIG. 2. FIG. 2 is a schematic flowchart of an intelligent triage method provided by an embodiment of the application. The method is applied to an electronic device. As shown in FIG. 2, it includes steps S21-S23:

S21: Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry.

In the specific embodiment of this application, a hierarchical reinforcement learning method is used to simplify the action space at each decision. As shown in Figure 3, the agent of the reinforcement learning model is designed into three levels, and the first level will obtain The historical symptoms obtained by the historical symptom inquiry, based on the historical symptoms obtained by executing the historical symptom inquiry task, determine whether to continue the symptom inquiry or directly return to the recommended department to end the interaction with the patient. The historical symptom inquiry refers to the symptoms before the current symptom inquiry Inquiry, the historical symptoms are the known symptoms of the patient. If the currently collected historical symptoms cannot be more accurately recommended for the department, the second level selects a target symptom subset from the preset multiple symptom subsets. Optionally, based on the symptom description obtained by interacting with the patient, the target symptom subset is a symptom subset that has a certain degree of relevance to the symptom description, rather than arbitrarily selected. For example, the information entered by the patient indicates that the patient has symptom A, Among the preset multiple symptom subsets [x1,x2,...,xn], x1, x5, xn-1 are symptom subsets that have a certain degree of relevance to symptom A, so choose one from these symptom subsets As a subset of target symptoms. At the same time, since each of the preset multiple symptom subsets includes multiple symptoms, the third level needs to select a symptom to be asked from the target symptom subset. It is understandable that dividing all symptoms into multiple symptom subsets reduces the size of the feature space corresponding to each symptom subset. In addition, selecting a symptom to be asked from the target symptom subset reduces the total number of symptom inquiries, so The model is easier to converge during training.

Please refer to Figure 4, the environment of reinforcement learning (Environment) is composed of patients and triage models. Compared with the training direction of supervised learning to learn models from tags, reinforcement learning is to train the model based on the feedback obtained from the interaction with the environment. When the reinforcement learning model makes a certain action in a certain state, the environment will give a feedback to measure the quality of the action. After the symptom to be asked is selected at the third level, the reinforcement learning model executes the cost based on the symptom to be asked. The symptom questioning task is to ask the patient about symptoms. If the feedback given by the environment for this round of symptom questioning is the third feedback (such as 1), it means that the patient does have the symptom to be questioned. This round of symptom questioning hits the patient’s true symptoms. That is, the symptom to be asked can be determined as the symptom obtained from the current round of symptom questioning, and at least one of the above symptoms is composed of the historical symptoms and the symptoms obtained from the current round of symptom questioning. On the contrary, if the feedback given by the environment for the current round of needle-shaped inquiry is the fourth feedback (such as -1), it means that the patient does not have the symptom to be asked, and it is not executed to determine the symptom to be asked for the current round of symptom inquiry. Symptom operation. Optionally, the reinforcement learning model used in this application is the Actor-Critic model.

S22: Obtain a department recommendation result corresponding to the patient according to the at least one symptom.

In the specific embodiment of this application, the department recommendation task is separated from the reinforcement learning task and executed by an independent triage model trained by supervised learning, and the triage model is added to the environment, thereby simplifying the reinforcement The overall learning task of the learning model makes the reinforcement learning model only need to pay attention to whether to return the recommended department to the patient, rather than which one of the many departments to return to. As shown in Figure 4, when the reinforcement learning model decides that the patient needs to return to the recommended department, it only needs to send a signal to the triage model to return to the recommended department, and the triage model gives the specific recommended department.

Before returning the recommended department to the patient, the triage model will output the corresponding department recommendation results for the symptoms obtained in each round of symptom inquiry and the previous known symptoms. Specifically, after obtaining the above-mentioned at least one symptom, input it into a pre-trained triage model to extract key information, perform department recommendation prediction based on the extracted key information, and output department recommendation results. Among them, the key information can be key descriptions such as "fever" and "stomach pain", or can be a key feature extracted from the at least one symptom. The triage model can be a convolutional neural network, a long short-term memory network, etc., There is no limitation here.

S23: Obtain feedback on the recommendation result of the department.

In a specific embodiment of the present application, the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the patient's actual treatment department, for example: the triage model predicts The departments to be recommended are Department A, Department B, and Department C. The corresponding probabilities are 85%, 80%, and 50% as the first predicted probability. Among them, Department A, Department B, and Department C include the actual patient visits Department.

Optionally, the foregoing obtaining the feedback on the recommendation result of the department includes: obtaining the feedback on the recommendation result of the department based on the first predicted probability of the actual treatment department. Specifically, it is first necessary to obtain the patient's corresponding historical department recommendation results. The so-called historical department recommendation result is the department recommendation that the triage model predicts based on the collected historical symptoms (ie known symptoms) before this recommendation prediction As a result, similarly, the historical department recommendation results include the historical department to be recommended and the second predicted probability corresponding to the historical recommended department, and the historical department to be recommended also includes the patient's actual treatment department. For example, the historical department to be recommended are Department A, For Department B and Department D, the corresponding probabilities of 72%, 85%, and 60% are the second predicted probabilities. Department A, Department B, and Department D also include the patient's actual consultation department. In the case that the first predicted probability of the actual visiting department is greater than the second predicted probability of the actual visiting department, the first feedback for the recommended result of the department is obtained; the first predicted probability of the actual visiting department is less than the second predicted probability of the actual visiting department In the case of predicting the probability, a second feedback for the recommendation result of the department is obtained. Assuming that department A is the patient's actual treatment department, in the above example, the first predicted probability of department A is significantly greater than the second predicted probability, the environment will return a positive feedback for the recommended result of the department, that is, the first feedback. On the contrary, if the first predicted probability of department A is less than or equal to the second predicted probability, the environment will return a negative feedback for the recommended result of the department, that is, the second feedback.

Optionally, the present application may also obtain feedback on the recommended results of the departments based on a sorting method. In specific implementation, the recommended departments are sorted according to the first predicted probability to obtain the first sorting result of the actual treatment departments, assuming Department B is the patient's actual treatment department. Department A, Department B, and Department C are sorted according to the first predicted probability, and the first ranking result of Department B is second. In addition, obtain the second sorting result of the actual visiting department, that is, sorting the historical department to be recommended based on the second predicted probability of the historical department to be recommended, for example: according to the second predicted probability, the department A and department B are sorted. , Department D is sorted, and the second ranking result of department B is in the first place. In the case where the first sorting result has increased compared to the second sorting result, the environment will also return the first feedback. On the contrary, when the first sorting result is lower than the second sorting result, the environment will be the same The second feedback will be returned.

S24: Determine the symptom obtained in this round of symptom inquiry as the target symptom according to the feedback.

In the specific embodiment of this application, when the environment's feedback on the department recommendation result is the first feedback, the symptom obtained from the current round of symptom inquiry is determined as the target symptom, that is, the output given by the environment for the triage model is a positive number In the case of feedback, the symptoms obtained in this round of symptom inquiry are determined as distinguishable symptoms.

S25: Perform the next round of symptom inquiry based on the target symptom, and repeat multiple rounds of symptom inquiry until returning to the recommended department to the patient.

In the specific embodiment of this application, after the symptoms obtained in the current round of symptom inquiry are determined as the target symptom, the target symptom is used to update the historical symptom, that is, the target symptom will also be determined as a known symptom or a historical symptom, and then return to the figure The starting point in 4 is to determine whether the current known symptoms are enough to return to the recommended department for the patient, if so, give a signal to the triage model to return to the recommended department, if otherwise, give priority to the target symptom to start the next round of symptom inquiry, and repeat the same procedure. Round symptom inquiry, until the known symptoms are sufficient to return the patient to the recommended department.

Further, the calculation formula of the feedback is:

r _s =α ₁ ·symptom_score+α ₂ ·dept_score+α ₃ ·tanh(τ·(Δp)), where r _s represents the calculated feedback, α ₁ , α ₂ and α ₃ are preset weights, and symptom_score represents In the symptom inquiry, whether the symptom to be asked hits the actual symptom of the patient. If it is, the value is 1, if it is otherwise, the value is -1. The dept_score represents the triage effect of the reinforcement learning model. The value is 1, if it is inconsistent, the value is -1, tanh(τ·(Δp)) represents the effect of each round of symptom inquiry on the results of the department recommendation, τ represents the preset constant, and Δp represents the symptoms of each round Ask about the changes in the order of the department recommended by the department after the new symptoms are obtained.

It can be seen that the embodiment of the present application obtains at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry; obtains the department recommendation result corresponding to the patient according to the at least one symptom; obtains Feedback on department recommendation results; determine the symptom obtained in the current round of symptom inquiry as the target symptom according to the feedback; perform the next round of symptom inquiry based on the target symptom, and repeat multiple rounds of symptom inquiry until the patient returns to the recommended department. In this way, the recommended department is predicted based on at least one symptom of the patient to obtain the department recommendation result, and then according to the feedback obtained for the department recommendation result, the symptom obtained in the current round of symptom inquiry is determined as the target symptom, that is, it is distinguishable and recommended to the department. For the symptom of help, the next round of symptom inquiry and triage prediction based on the target symptom will not only help reduce the number of symptom inquiry rounds, but also help improve the accuracy of triage.

In one embodiment of the present application, the solution of the present application can also be applied to the field of smart medical care. For example, through multiple rounds of interaction between electronic devices and patients, based on the symptom-related description input by the patient, the patient is intelligently triaged, Out of the departments that can be consulted by the patient. Since this application will determine the distinguished symptoms from the many symptoms of the patient based on the feedback given by the model environment, the symptom inquiry and triage based on the distinguished symptoms will help improve the accuracy of triage and not only reduce the medical care. The labor cost of personnel also improves the efficiency of patients' medical treatment, so that patients have a better medical experience.

Please refer to FIG. 5, which is a schematic flow chart of another intelligent triage method provided by the embodiment of the present application, which can also be implemented based on the network system architecture shown in FIG. 1, as shown in FIG. 5, including steps S51-S56:

S51: Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

S52: Input the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

S53: Perform department recommendation prediction based on the extracted key information, and output department recommendation results;

S54. Obtain feedback on the recommendation result of the department;

S55: Determine the symptom obtained in this round of symptom inquiry as the target symptom according to the feedback;

S56: Perform the next round of symptom inquiry based on the target symptom, and repeat multiple rounds of symptom inquiry until returning to the recommended department to the patient.

Among them, the specific implementation of steps S51-S56 has been described in the embodiment shown in FIG. 2 and can achieve the same or similar beneficial effects. In order to avoid repetition, the details are not repeated here.

Based on the description of the foregoing embodiment of the intelligent triage method, please refer to FIG. 6. FIG. 6 is a schematic structural diagram of an intelligent triage device provided by an embodiment of the application. As shown in FIG. 6, the device includes:

The symptom acquisition module 61 is configured to acquire at least one symptom of the patient; the at least one symptom includes the symptoms obtained in the current round of symptom inquiry;

The department recommendation module 62 is configured to obtain a department recommendation result corresponding to the patient according to the at least one symptom;

The feedback obtaining module 63 is configured to obtain feedback on the recommendation result of the department;

The symptom determination module 64 determines the symptom obtained in the current round of symptom inquiry as the target symptom according to the feedback;

The department output module 65 is configured to perform the next round of symptom inquiry based on the target symptom, and repeat multiple rounds of symptom inquiry until returning to the recommended department to the patient.

In a possible implementation manner, in terms of obtaining a department recommendation result corresponding to the patient according to the at least one symptom, the department recommendation module 62 is specifically configured to:

Inputting the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

Based on the extracted key information, a recommendation prediction of the department is performed, and the recommendation result of the department is output.

In a possible implementation manner, the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the actual treatment department of the patient; Regarding the feedback of department recommendation results, the feedback obtaining module 63 is specifically used to:

Obtain feedback on the recommendation result of the department based on the first predicted probability of the actual visiting department.

In a possible implementation manner, in terms of obtaining feedback on the recommendation result of the department based on the first predicted probability of the actual visiting department, the feedback obtaining module 63 is specifically configured to:

Obtaining the recommended result of the historical department corresponding to the patient; the recommended result of the historical department includes the historical department to be recommended and the second predicted probability corresponding to the historical department to be recommended, and the historical department to be recommended includes the actual department;

Obtaining the second predicted probability of the actual clinic visit from the recommendation result of the historical department;

In the case that the first predicted probability of the actual medical department is greater than the second predicted probability of the actual medical department, the first feedback for the recommended result of the department is obtained; the first predicted probability of the actual medical department is less than In the case of being equal to the second predicted probability of the actual treatment department, obtaining a second feedback for the recommendation result of the department;

or,

Sorting the departments to be recommended based on the first predicted probability of the departments to be recommended to obtain the first ranking result of the actual visiting departments;

Acquiring a second sorting result of the actual treatment department; the second sorting result is obtained by sorting the historical department to be recommended based on the second predicted probability of the historical department to be recommended;

In the case where the first ranking result is higher than the second ranking result, a first feedback for the department recommendation result is obtained; when the first ranking result is compared with the second ranking result In the case of a decrease in the result, a second feedback for the recommended result of the department is obtained.

In a possible implementation manner, in terms of determining the symptom obtained in the current round of symptom inquiry as the target symptom according to the feedback, the symptom determining module 64 is specifically configured to:

In the case where the feedback is the first feedback for the recommendation result of the department, the symptom obtained in the current round of symptom inquiry is determined as the target symptom.

In a possible implementation manner, in terms of obtaining at least one symptom of the patient, the symptom obtaining module 61 is specifically configured to:

Get historical symptoms and ask about historical symptoms;

Judging whether to return to the recommended department to the patient based on the historical symptoms;

If otherwise, the target symptom subset is determined from the preset multiple symptom subsets, and the symptom to be asked is determined from the target symptom subset;

Use the symptom to be asked to execute the symptom questioning task of the current round, and in the case of obtaining the third feedback for the symptom questioning of the current round, determine the symptom to be questioned as the symptoms obtained from the symptom questioning of the current round; the third The feedback indicates that the patient has the symptom to be asked;

The at least one symptom is composed of the historical symptoms and the symptoms obtained from the current round of symptom inquiry.

According to an embodiment of the present application, the units of the intelligent triage device shown in FIG. 6 can be separately or completely combined into one or several other units to form, or some of the units can also be split. It is composed of multiple units with smaller functions, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In practical applications, the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit. In other embodiments of the present application, the intelligent triage-based device may also include other units. In practical applications, these functions may also be implemented with the assistance of other units, and may be implemented by multiple units in cooperation.

According to another embodiment of the present application, a general-purpose computing device such as a computer including a central processing unit (CPU), a random access storage medium (RAM), a read-only storage medium (ROM) and other processing elements and storage elements can be used Run a computer program (including program code) that can execute the steps involved in the corresponding method shown in FIG. 2 or FIG. 5 to construct the intelligent triage device as shown in FIG. 6 and to implement the implementation of this application The intelligent triage method of cases. The computer program may be recorded on, for example, a computer-readable recording medium, and loaded into the above-mentioned computing device through the computer-readable recording medium, and run in it.

Based on the description of the foregoing method embodiment and device embodiment, an embodiment of the present application also provides an electronic device. Referring to FIG. 7, the electronic device includes at least a processor 71, an input device 72, an output device 73 and a computer storage medium 74. Among them, the processor 71, the input device 72, the output device 73, and the computer storage medium 74 in the electronic device may be connected by a bus or other means.

The computer storage medium 74 may be stored in the memory of the electronic device. The computer storage medium 74 is used to store a computer program. The computer program includes program instructions. The processor 71 is used to execute the program stored in the computer storage medium 74. instruction. The processor 71 (or CPU (Central Processing Unit, central processing unit)) is the computing core and control core of an electronic device. It is suitable for implementing one or more instructions, and specifically for loading and executing one or more instructions to achieve Corresponding method flow or corresponding function.

In an embodiment, the processor 71 of the electronic device provided in the embodiment of the present application may be used to perform a series of intelligent triage processing:

Obtain feedback on the recommendation result of the department;

In another embodiment, the execution of the processor 71 to obtain the department recommendation result corresponding to the patient according to the at least one symptom includes:

In still another embodiment, the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the patient's actual treatment department; the processor 71 executes the acquisition target The feedback of the recommended results of the department includes:

In another embodiment, the processor 71 executing the first predicted probability based on the actual department to obtain feedback on the recommendation result of the department includes:

or,

In the case where the first ranking result is higher than the second ranking result, a first feedback for the department recommendation result is obtained; when the first ranking result is compared to the second ranking result In the case that the result has declined, a second feedback for the recommended result of the department is obtained.

In another embodiment, the processor 71 executes the determination of the symptom obtained in the current round of symptom inquiry as the target symptom according to the feedback, including:

In still another embodiment, the processor 71 executing the acquiring at least one symptom of the patient includes:

Get historical symptoms and ask about historical symptoms;

Exemplarily, the foregoing electronic device may be a server, a cloud server, a computer host, a server cluster, etc. The electronic device includes but is not limited to a processor 71, an input device 72, an output device 73, and a computer storage medium 74. Those skilled in the art can understand that the schematic diagram is only an example of the electronic device, and does not constitute a limitation on the electronic device, and may include more or fewer components than those shown in the figure, or a combination of certain components, or different components.

It should be noted that, since the processor 71 of the electronic device executes the computer program to implement the steps in the above-mentioned intelligent triage method, the embodiments of the above-mentioned intelligent triage method are all applicable to the electronic device, and can achieve the same or similar The beneficial effects.

The embodiment of the present application also provides a computer storage medium (Memory). The computer storage medium is a memory device in an electronic device for storing programs and data. It can be understood that the computer storage medium herein may include a built-in storage medium in the terminal, and of course, may also include an extended storage medium supported by the terminal. The computer storage medium provides storage space, and the storage space stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by the processor 71 are also stored in the storage space, and these instructions may be one or more computer programs (including program codes). It should be noted that the computer storage medium here can be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory; optionally, it can also be at least one located far away from the aforementioned processing. The computer storage medium of the device 71. Optionally, the computer storage medium involved in this application may be a computer-readable storage medium, and the computer storage medium may be non-volatile or volatile.

In one embodiment, the processor 71 can load and execute one or more instructions stored in the computer storage medium to implement the corresponding steps of the above-mentioned intelligent triage method; in a specific implementation, one or more instructions in the computer storage medium The instructions are loaded by the processor 71 and execute the following steps:

Obtain feedback on the recommendation result of the department;

In another example, when one or more instructions in the computer storage medium are loaded by the processor 71, the following steps are executed:

or,

Get historical symptoms and ask about historical symptoms;

Exemplarily, the computer program in the computer storage medium includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc.

It should be noted that, since the computer program of the computer storage medium is executed by the processor to realize the steps in the above-mentioned intelligent triage method, all the embodiments of the above-mentioned intelligent triage method are applicable to the computer storage medium and can achieve The same or similar beneficial effects.

The embodiments of the application are described in detail above, and specific examples are used in this article to illustrate the principles and implementation of the application. The descriptions of the above embodiments are only used to help understand the methods and core ideas of the application; at the same time, for Those of ordinary skill in the art, based on the idea of the application, will have changes in the specific implementation and the scope of application. In summary, the content of this specification should not be construed as a limitation to the application.

Claims

An intelligent triage method, wherein the method includes:

Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

Obtaining a department recommendation result corresponding to the patient according to the at least one symptom;

Obtain feedback on the recommendation result of the department;

According to the feedback, determine the symptom obtained in this round of symptom inquiry as the target symptom;

The next round of symptom inquiry is performed based on the target symptom, and multiple rounds of symptom inquiry are repeated until the patient returns to the recommended department.
The method according to claim 1, wherein the obtaining a department recommendation result corresponding to the patient according to the at least one symptom comprises:

Inputting the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

Based on the extracted key information, a recommendation prediction of the department is performed, and the recommendation result of the department is output.
The method according to claim 1 or 2, wherein the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the actual department of the patient; The description of obtaining feedback on the recommended results of the department includes:

Obtain feedback on the recommendation result of the department based on the first predicted probability of the actual visiting department.
The method according to claim 3, wherein the obtaining feedback on a recommendation result of the department based on the first predicted probability of the actual department visits a doctor comprises:

Obtaining the recommended result of the historical department corresponding to the patient; the recommended result of the historical department includes the historical department to be recommended and the second predicted probability corresponding to the historical department to be recommended, and the historical department to be recommended includes the actual department;

Obtaining the second predicted probability of the actual clinic visit from the recommendation result of the historical department;

In the case that the first predicted probability of the actual medical department is greater than the second predicted probability of the actual medical department, the first feedback for the recommended result of the department is obtained; the first predicted probability of the actual medical department is less than In the case of being equal to the second predicted probability of the actual treatment department, obtaining a second feedback for the recommendation result of the department;

or,

Sorting the departments to be recommended based on the first predicted probability of the departments to be recommended to obtain the first ranking result of the actual visiting departments;

Acquiring a second sorting result of the actual treatment department; the second sorting result is obtained by sorting the historical department to be recommended based on the second predicted probability of the historical department to be recommended;

In the case where the first ranking result is higher than the second ranking result, a first feedback for the department recommendation result is obtained; when the first ranking result is compared with the second ranking result In the case of a decrease in the result, a second feedback for the recommended result of the department is obtained.
The method according to claim 4, wherein the determining the symptom obtained in the current round of symptom inquiry as the target symptom according to the feedback comprises:

In the case where the feedback is the first feedback for the recommendation result of the department, the symptom obtained in the current round of symptom inquiry is determined as the target symptom.
The method according to claim 1, wherein said obtaining at least one symptom of the patient comprises:

Get historical symptoms and ask about historical symptoms;

Judging whether to return to the recommended department to the patient based on the historical symptoms;

If otherwise, the target symptom subset is determined from the preset multiple symptom subsets, and the symptom to be asked is determined from the target symptom subset;

Use the symptom to be asked to execute the symptom questioning task of the current round, and in the case of obtaining the third feedback for the symptom questioning of the current round, determine the symptom to be questioned as the symptoms obtained from the symptom questioning of the current round; the third The feedback indicates that the patient has the symptom to be asked;

The at least one symptom is composed of the historical symptoms and the symptoms obtained from the current round of symptom inquiry.
An intelligent triage device, wherein the device includes:

The symptom acquisition module is used to acquire at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

The department recommendation module is used to obtain the department recommendation result corresponding to the patient according to the at least one symptom;

Feedback acquisition module, used to acquire feedback on the recommendation result of the department;

The symptom determination module determines the symptom obtained in this round of symptom inquiry as the target symptom according to the feedback;

The department output module is configured to perform the next round of symptom inquiry based on the target symptom, and repeatedly execute multiple rounds of symptom inquiry until returning to the recommended department to the patient.
8. The device according to claim 7, wherein, in terms of obtaining a department recommendation result corresponding to the patient according to the at least one symptom, the department recommendation module is specifically configured to:

Inputting the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

Based on the extracted key information, a recommendation prediction of the department is performed, and the recommendation result of the department is output.
An electronic device, including an input device and an output device, which also includes:

Processor, suitable for implementing one or more instructions; and,

A computer storage medium storing one or more instructions, and the one or more instructions are suitable for being loaded by the processor and executing the following methods:

Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

Obtaining a department recommendation result corresponding to the patient according to the at least one symptom;

Obtain feedback on the recommendation result of the department;

According to the feedback, determine the symptom obtained in this round of symptom inquiry as the target symptom;

The next round of symptom inquiry is performed based on the target symptom, and multiple rounds of symptom inquiry are repeated until the patient returns to the recommended department.
The electronic device according to claim 9, wherein when the department recommendation result corresponding to the patient is obtained according to the at least one symptom, the following is specifically executed:

Inputting the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

Based on the extracted key information, a recommendation prediction of the department is performed, and the recommendation result of the department is output.
The electronic device according to claim 9 or 10, wherein the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the actual department of the patient; When the feedback on the recommendation result of the department is obtained, the following is specifically executed:

Obtain feedback on the recommendation result of the department based on the first predicted probability of the actual visiting department.
The electronic device according to claim 11, wherein, when the feedback on the recommendation result of the department is obtained based on the first predicted probability of the actual treatment department, the following is specifically executed:

Obtaining the recommended result of the historical department corresponding to the patient; the recommended result of the historical department includes the historical department to be recommended and the second predicted probability corresponding to the historical department to be recommended, and the historical department to be recommended includes the actual department;

Obtaining the second predicted probability of the actual clinic visit from the recommendation result of the historical department;

In the case that the first predicted probability of the actual medical department is greater than the second predicted probability of the actual medical department, the first feedback for the recommended result of the department is obtained; the first predicted probability of the actual medical department is less than In the case of being equal to the second predicted probability of the actual treatment department, obtaining a second feedback for the recommendation result of the department;

or,

Sorting the departments to be recommended based on the first predicted probability of the departments to be recommended to obtain the first ranking result of the actual visiting departments;

Acquiring a second sorting result of the actual treatment department; the second sorting result is obtained by sorting the historical department to be recommended based on the second predicted probability of the historical department to be recommended;

In the case where the first ranking result is higher than the second ranking result, a first feedback for the department recommendation result is obtained; when the first ranking result is compared with the second ranking result In the case of a decrease in the result, a second feedback for the recommended result of the department is obtained.
The electronic device according to claim 12, wherein when the symptom obtained in the current round of symptom inquiry is determined as the target symptom according to the feedback, the following is specifically executed:

In the case where the feedback is the first feedback for the recommendation result of the department, the symptom obtained in the current round of symptom inquiry is determined as the target symptom.
The electronic device according to claim 9, wherein when the at least one symptom of the patient is obtained, the following is specifically executed:

Get historical symptoms and ask about historical symptoms;

Judging whether to return to the recommended department to the patient based on the historical symptoms;

If otherwise, the target symptom subset is determined from the preset multiple symptom subsets, and the symptom to be asked is determined from the target symptom subset;

Use the symptom to be asked to execute the symptom questioning task of the current round, and in the case of obtaining the third feedback for the symptom questioning of the current round, determine the symptom to be questioned as the symptoms obtained from the symptom questioning of the current round; the third The feedback indicates that the patient has the symptom to be asked;

The at least one symptom is composed of the historical symptoms and the symptoms obtained from the current round of symptom inquiry.
A computer storage medium, wherein the computer storage medium stores one or more instructions, and the one or more instructions are suitable for being loaded by a processor and executing the following methods:

Obtain at least one symptom of the patient; the at least one symptom includes the symptom obtained in the current round of symptom inquiry;

Obtaining a department recommendation result corresponding to the patient according to the at least one symptom;

Obtain feedback on the recommendation result of the department;

According to the feedback, determine the symptom obtained in this round of symptom inquiry as the target symptom;

The next round of symptom inquiry is performed based on the target symptom, and multiple rounds of symptom inquiry are repeated until the patient returns to the recommended department.
15. The computer storage medium according to claim 15, wherein when the department recommendation result corresponding to the patient is obtained according to the at least one symptom, the following is specifically executed:

Inputting the at least one symptom into a pre-trained triage model to extract key information; the triage model is an independent supervised learning model used to perform department recommendation tasks;

Based on the extracted key information, a recommendation prediction of the department is performed, and the recommendation result of the department is output.
The computer storage medium according to claim 15 or 16, wherein the department recommendation result includes the department to be recommended and the first predicted probability corresponding to the department to be recommended, and the department to be recommended includes the actual department of the patient. ; When obtaining the feedback on the recommendation result of the department, specifically execute:

Obtain feedback on the recommendation result of the department based on the first predicted probability of the actual visiting department.
18. The computer storage medium according to claim 17, wherein, when the feedback on the recommendation result of the department is obtained based on the first predicted probability of the actual treatment department, the following is specifically executed:

Obtaining the recommended result of the historical department corresponding to the patient; the recommended result of the historical department includes the historical department to be recommended and the second predicted probability corresponding to the historical department to be recommended, and the historical department to be recommended includes the actual department;

Obtaining the second predicted probability of the actual clinic visit from the recommendation result of the historical department;

In the case that the first predicted probability of the actual medical department is greater than the second predicted probability of the actual medical department, a first feedback for the recommended result of the department is obtained; the first predicted probability of the actual medical department is less than In the case of being equal to the second predicted probability of the actual treatment department, obtaining a second feedback for the recommendation result of the department;

or,

Sorting the departments to be recommended based on the first predicted probability of the departments to be recommended to obtain the first ranking result of the actual visiting departments;

Acquiring a second sorting result of the actual treatment department; the second sorting result is obtained by sorting the historical department to be recommended based on the second predicted probability of the historical department to be recommended;

In the case where the first ranking result is higher than the second ranking result, a first feedback for the department recommendation result is obtained; when the first ranking result is compared with the second ranking result In the case of a decrease in the result, a second feedback for the recommended result of the department is obtained.
18. The computer storage medium according to claim 18, wherein when the symptom obtained from the current round of symptom inquiry is determined as the target symptom according to the feedback, the following is specifically executed:

In the case where the feedback is the first feedback for the recommendation result of the department, the symptom obtained in the current round of symptom inquiry is determined as the target symptom.
The computer storage medium according to claim 15, wherein when the at least one symptom of the patient is obtained, the following is specifically executed:

Get historical symptoms and ask about historical symptoms;

Judging whether to return to the recommended department to the patient based on the historical symptoms;

If otherwise, the target symptom subset is determined from the preset multiple symptom subsets, and the symptom to be asked is determined from the target symptom subset;

Use the symptom to be asked to execute the symptom questioning task of the current round, and in the case of obtaining the third feedback for the symptom questioning of the current round, determine the symptom to be questioned as the symptoms obtained from the symptom questioning of the current round; the third The feedback indicates that the patient has the symptom to be asked;

The at least one symptom is composed of the historical symptoms and the symptoms obtained from the current round of symptom inquiry.