CN114596947A

CN114596947A - Diagnosis and treatment department recommendation method and device, electronic equipment and storage medium

Info

Publication number: CN114596947A
Application number: CN202210220519.5A
Authority: CN
Inventors: 胡宇峰; 郑宇宏; 徐伟建; 郑烨翰; 彭卫华; 吕雅娟
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-03-08
Filing date: 2022-03-08
Publication date: 2022-06-07

Abstract

The disclosure provides a diagnosis and treatment department recommendation method and device, electronic equipment and a storage medium, relates to the technical field of artificial intelligence such as knowledge graphs and the like, and can be applied to scenes such as intelligent medical treatment and the like. The implementation scheme is as follows: a medical department recommending method comprises the following steps: acquiring voice information for describing current symptoms of a patient to be diagnosed; carrying out voice recognition on the voice information to obtain an original recognition result which is not decoded into characters in the voice information; extracting keywords for determining a diagnosis and treatment department of the patient from the original recognition result, wherein the keywords comprise current symptoms of the patient; and determining a diagnosis and treatment department for recommendation based on the keyword.

Description

Diagnosis and treatment department recommendation method and device, electronic equipment and storage medium

Technical Field

The disclosure relates to the technical field of artificial intelligence such as knowledge maps and the like, and can be applied to scenes such as intelligent medical treatment and the like. And more particularly, to a clinical department recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

Background

Artificial intelligence is the subject of research that makes computers simulate some human thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.), both at the hardware level and at the software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.

With the development of artificial intelligence technology, the artificial intelligence technology has been gradually applied in the medical field to assist in various important links involved in the medical process. Because the medical field has the characteristics of strong specialization, difficult access of medical data, scarce model training data and the like, how to better adapt to the artificial intelligence technology in the special field is still one of the research hotspots in the industry.

The approaches described in this section are not necessarily approaches that have been previously conceived or pursued. Unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section. Similarly, unless otherwise indicated, the problems mentioned in this section should not be considered as having been acknowledged in any prior art.

Disclosure of Invention

The present disclosure provides a diagnosis and treatment department recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

According to an aspect of the present disclosure, a medical department recommending method is provided. The method comprises the following steps: acquiring voice information for describing current symptoms of a patient to be diagnosed; carrying out voice recognition on the voice information to obtain an original recognition result which is not decoded into characters in the voice information; extracting keywords for determining a diagnosis and treatment department of the patient from the original recognition result, wherein the keywords comprise current symptoms of the patient; and determining a diagnosis and treatment department for recommendation based on the keyword.

According to another aspect of the present disclosure, a clinical department recommendation device is provided. The device comprises: the system comprises a voice acquisition module, a diagnosis module and a control module, wherein the voice acquisition module is configured to acquire voice information for describing current symptoms of a patient to be diagnosed; the voice recognition module is configured to perform voice recognition on the voice information to acquire an original recognition result which is not decoded into characters in the voice information; the characteristic extraction module is configured to extract keywords for determining the diagnosis and treatment department of the patient from the original recognition result, wherein the keywords comprise the current symptoms of the patient; and a result determination module configured to determine a diagnosis department for recommendation based on the keyword.

According to another aspect of the present disclosure, an electronic device is provided. The electronic device includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method according to the above.

According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method according to the above.

According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program realizes the method according to the above when executed by a processor.

According to one or more embodiments of the present disclosure, the accuracy of medical department recommendation can be improved.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the embodiments and, together with the description, serve to explain the exemplary implementations of the embodiments. The illustrated embodiments are for purposes of illustration only and do not limit the scope of the claims. Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements.

FIG. 1 illustrates a schematic diagram of an exemplary system in which various methods described herein may be implemented, according to an embodiment of the present disclosure;

fig. 2 illustrates a flow chart of a clinical department recommendation method according to one embodiment of the present disclosure;

fig. 3 illustrates a flow chart of a medical department recommendation method according to another embodiment of the present disclosure;

fig. 4 shows a schematic diagram of a clinical department recommendation method according to an embodiment of the present disclosure in an emergency scenario;

fig. 5 is a block diagram illustrating a configuration of a medical department recommending apparatus according to an embodiment of the present disclosure;

fig. 6 is a block diagram illustrating a configuration of a clinical department recommendation device according to another embodiment of the present disclosure;

FIG. 7 illustrates a block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the present disclosure, unless otherwise specified, the use of the terms "first", "second", and the like to describe various elements is not intended to limit the positional relationship, the temporal relationship, or the importance relationship of the elements, and such terms are used only to distinguish one element from another. In some examples, a first element and a second element may refer to the same instance of the element, and in some cases, based on the context, they may also refer to different instances.

The terminology used in the description of the various described examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the elements may be one or more. Furthermore, the term "and/or" as used in this disclosure is intended to encompass any and all possible combinations of the listed items.

In the related art, it is often difficult to accurately acquire interesting information from speech related to the medical field through speech recognition due to inherent characteristics of strong professional in the medical field, difficult access of medical data, scarce model training data, and the like. Therefore, it is difficult to deploy or apply a clinical department recommendation system implementing a voice recognition function in a hospital or medical institution at present because it is difficult to ensure that relevant information for recommending a clinical department is accurately acquired from a disease description voice provided by a patient.

In addition, in an emergency scene in which a patient or his family members describe disease information of the patient by voice by dialing an emergency call or the like, it becomes important how to quickly and accurately acquire the individual of the patient and the disease information thereof from the voice. Meanwhile, in an emergency scene, the emergency doctor often performs screening on site after receiving the patient and/or consultation through the relevant department doctor, which may delay the treatment time and miss the emergency opportunity.

In view of at least one of the above technical problems, according to an aspect of the embodiments of the present disclosure, a method for recommending a medical department is provided. Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

Before describing the method of embodiments of the present disclosure in detail, an exemplary system in which the method of embodiments of the present disclosure may be implemented is first described in conjunction with fig. 1.

Fig. 1 illustrates a schematic diagram of an exemplary system 100 in which various methods and apparatus described herein may be implemented, according to an embodiment of the present disclosure. Referring to fig. 1, the system 100 includes one or

more client devices

101, 102, 103, 104, 105, and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120.

Client devices

101, 102, 103, 104, 105, and 106 may be configured to execute one or more applications.

In an embodiment of the present disclosure, server 120 may run one or more services or software applications that enable the execution of a medical department recommendation method in accordance with an embodiment of the present disclosure.

In some embodiments, the server 120 may also provide other services or software applications that may include non-virtual environments and virtual environments. In some embodiments, these services may be provided as web-based services or cloud services, for example, provided to users of

client devices

101, 102, 103, 104, 105, and/or 106 under a software as a service (SaaS) model.

In the configuration shown in fig. 1, server 120 may include one or more components that implement the functions performed by server 120. These components may include software components, hardware components, or a combination thereof, which may be executed by one or more processors. A user operating a

client device

101, 102, 103, 104, 105, and/or 106 may, in turn, utilize one or more client applications to interact with the server 120 to take advantage of the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from system 100. Accordingly, fig. 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

The user may use

client devices

101, 102, 103, 104, 105, and/or 106 to speak speech describing the patient's current condition. The client device may provide an interface that enables a user of the client device to interact with the client device. The client device may also output information to the user via the interface. Although fig. 1 depicts only six client devices, those skilled in the art will appreciate that any number of client devices may be supported by the present disclosure.

Client devices

101, 102, 103, 104, 105, and/or 106 may include various types of computer devices, such as portable handheld devices, general purpose computers (such as personal computers and laptops), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, gaming systems, thin clients, various messaging devices, sensors or other sensing devices, and so forth. These computer devices may run various types and versions of software applications and operating systems, such as MICROSOFT Windows, APPLE iOS, UNIX-like operating systems, Linux, or Linux-like operating systems (e.g., GOOGLE Chrome OS); or include various Mobile operating systems such as MICROSOFT Windows Mobile OS, iOS, Windows Phone, Android. Portable handheld devices may include cellular telephones, smart phones, tablets, Personal Digital Assistants (PDAs), and the like. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. The gaming system may include a variety of handheld gaming devices, internet-enabled gaming devices, and the like. The client device is capable of executing a variety of different applications, such as various Internet-related applications, communication applications (e.g., email applications), Short Message Service (SMS) applications, and may use a variety of communication protocols.

Network 110 may be any type of network known to those skilled in the art that may support data communications using any of a variety of available protocols, including but not limited to TCP/IP, SNA, IPX, etc. Merely by way of example, one or more networks 110 may be a Local Area Network (LAN), an ethernet-based network, a token ring, a Wide Area Network (WAN), the internet, a virtual network, a Virtual Private Network (VPN), an intranet, an extranet, a Public Switched Telephone Network (PSTN), an infrared network, a wireless network (e.g., bluetooth, WIFI), and/or any combination of these and/or other networks.

The server 120 may include one or more general purpose computers, special purpose server computers (e.g., PC (personal computer) servers, UNIX servers, mid-end servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and/or combination. The server 120 may include one or more virtual machines running a virtual operating system, or other computing architecture involving virtualization (e.g., one or more flexible pools of logical storage that may be virtualized to maintain virtual storage for the server). In various embodiments, the server 120 may run one or more services or software applications that provide the functionality described below.

The computing units in server 120 may run one or more operating systems including any of the operating systems described above, as well as any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and/or middle tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, and the like.

In some implementations, the server 120 may include one or more applications to analyze and consolidate data feeds and/or event updates received from users of the

client devices

101, 102, 103, 104, 105, and/or 106. Server 120 may also include one or more applications to display data feeds and/or real-time events via one or more display devices of

client devices

101, 102, 103, 104, 105, and/or 106.

In some embodiments, the server 120 may be a server of a distributed system, or a server incorporating a blockchain. The server 120 may also be a cloud server, or a smart cloud computing server or a smart cloud host with artificial intelligence technology. The cloud Server is a host product in a cloud computing service system, and is used for solving the defects of high management difficulty and weak service expansibility in the traditional physical host and Virtual Private Server (VPS) service.

The system 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and video files. The database 130 may reside in various locations. For example, the database used by the server 120 may be local to the server 120, or may be remote from the server 120 and may communicate with the server 120 via a network-based or dedicated connection. The database 130 may be of different types. In certain embodiments, the database used by the server 120 may be, for example, a relational database. One or more of these databases may store, update, and retrieve data to and from the database in response to the command.

In some embodiments, one or more of the databases 130 may also be used by applications to store application data. The databases used by the application may be different types of databases, such as key-value stores, object stores, or regular stores supported by a file system.

The system 100 of fig. 1 may be configured and operated in various ways to enable application of the various methods and apparatus described in accordance with embodiments of the present disclosure.

Fig. 2 shows a flowchart of a medical department recommendation method according to an embodiment of the present disclosure. As shown in fig. 2, the method 200 may include the steps of:

in step S202, voice information describing a current symptom of a patient to be diagnosed is acquired.

In step S204, speech recognition is performed on the speech information to obtain an original recognition result that is not decoded into characters in the speech information.

In step S206, keywords for determining the medical department of the patient are extracted from the original recognition result, wherein the keywords include the current symptoms of the patient.

In step S208, a medical department for recommendation is determined based on the keyword.

According to the diagnosis and treatment department recommendation method disclosed by the embodiment of the disclosure, the speech recognition and the keyword extraction are cascaded, so that the interested keywords are directly extracted from the original recognition result under the condition that the speech recognition result is not decoded into characters, the speech recognition error caused by character decoding can be avoided, and the speech recognition effect specific to the medical field is improved. Therefore, the information related to the possibly-suffered disease of the patient at present can be accurately acquired from the disease description voice provided by the patient, and the accuracy of the recommendation of the medical department is improved.

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related patient information all accord with the regulations of related laws and regulations, and do not violate the common customs of public sequences.

One or more aspects of each step of the clinical department recommendation method according to an embodiment of the present disclosure will be described in detail below.

In step S202, in an example, the voice describing the current symptoms of the patient may be a disease description voice provided by the patient and/or his family members for medical consultation, registration, etc. in a general outpatient scene, or may be a disease description voice provided by the patient and/or his family members by dialing an emergency phone, etc. in an emergency scene. The condition-describing speech may include information that is descriptive of the patient's current condition.

According to some embodiments, the speech information acquired in step S202 may include mood information generated by the patient when speaking a speech, and the raw recognition result acquired in step S204 may include a result of recognizing the mood information.

In this way, diseases that the patient may potentially suffer from can be discovered by means of the mood information in the patient's voice, thereby contributing to the accuracy of the clinical department recommendation.

In an example, the mood information can include non-content information such as shortness of breath, slurred mouth, etc. in the speech information. Such non-content information may be included in the audio data of the speech information. If the patient experiences an ill-of-mouth condition in the voice of the emergency call that the patient places, then voice recognizing the voice information about the voice may also include recognizing the portion of mood information, and the result of the recognition may include information indicating that the patient may potentially have a stroke. This information may be included in the original recognition result obtained by performing speech recognition on the speech information. In another example, the mood information may also include non-content information such as intonation, mood, etc. in the voice information.

In an example, to realize the tone Recognition, an enhanced model with a tone Recognition function can be obtained by adding an adapter for tone Recognition at an output layer on the basis of a general Automatic Speech Recognition (ASR) model.

In step S204, the original recognition result of the speech that is not decoded into words is obtained to avoid the speech recognition error caused by word decoding. As described above, speech recognition applied to the medical field tends to have a low recognition rate, and recognition errors such as a large number of wrongly written words may be included in the recognition result. If the recognition result thus decoded into words is transmitted to a subsequent semantic understanding module for extracting keywords, a larger error may be generated. Thus, the original recognition result of the speech, which is not decoded into words, is obtained in the speech recognition step. That is, the speech is not decoded in the speech recognition step to the extent of the word vector, but rather its original recognition vector is utilized to directly extract keywords in subsequent language understanding.

In step S206, since the result of the previous speech recognition step is the original recognition result not decoded into text, it can provide a higher accuracy than the recognition result decoded into text, and thus the keyword of interest can be extracted more accurately.

In other words, by cascading speech recognition with keyword extraction such that a keyword of interest is directly extracted from an original recognition vector without decoding a speech recognition result into text, it is possible to avoid speech recognition errors due to text decoding, thereby improving a speech recognition effect specific to the medical field.

In an example, the keyword extraction step in step S206 may use a known Natural Language Understanding (NLU) technique. The keywords may further include personal information of the patient, such as name, age, etc., which may be used to obtain a past history of the patient. It should be noted that, in the technical solution of the present disclosure, the acquisition, storage, application, etc. of the related patient information all conform to the regulations of the relevant laws and regulations, and do not violate the good custom of the public order. The keywords may also include other clinical elements in addition to the current condition, such as a surgical history.

According to some embodiments, at least one of the speech recognition step in step S204 and the keyword extraction step in step S204 may use a pre-trained model, wherein the pre-trained model may be a model obtained by adding an adapter for a specific purpose to an output layer for training on the basis of a general model. In the step of performing voice recognition on the voice information, the pre-trained model may be a model obtained by adding at least one of an adapter for medical field vocabulary and an adapter for mood recognition to an output layer in the general model for voice recognition to train. In the step of extracting keywords from the original recognition result, the pre-trained model may be a model obtained by adding an adapter for a keyword of interest to an output layer in a general model for semantic understanding in the medical field for training.

In this way, the amount of data used for model training may be reduced, and an enhanced model for downstream tasks may be provided.

For speech recognition, as mentioned above, model training data in the medical field is often scarce, and a patient or family member may not use professional medical vocabulary but a popular language, thereby making it difficult to directly train out a sound model for speech recognition in the medical field. Thus, in accordance with embodiments of the present disclosure, an enhanced model may be obtained by training by adding an adapter for the medical domain vocabulary at the output layer, for example, on the basis of a generic ASR model. The parameters of the adapter are the main targets of training, i.e. the weights of the neural network. In an example, the generic model for speech recognition may be, for example, a transform (Transformer) based generic speech recognition model.

In addition, since the speech recognition may further include mood recognition, an adapter for mood recognition may also be added at the output layer.

Similarly, for semantic understanding (i.e., keyword extraction), an enhanced model may be obtained by training on the basis of, for example, a generic medical language model by adding adapters for one or more keywords (e.g., personal information of a patient, or clinical elements such as current symptoms) at the output layer. In an example, the generic model for semantic understanding may be, for example, an ERNIE-Health (medical pre-trained language model) based generic medical language model.

In step S206, determining a department for recommendation based on the keyword may be considered differently according to whether the past history of the patient can be obtained. An embodiment of this step will be further described below with reference to fig. 3.

Fig. 3 is a flowchart illustrating a medical department recommendation method according to another embodiment of the present disclosure. As shown in fig. 3, the method 300 may include steps S302, S304, and S306 similar to steps S202, S204, and S206 described with reference to fig. 2. Accordingly, details of various aspects of the above steps are not described herein.

As shown in fig. 3, the method 300 further includes a step S308 of determining a clinical department for recommendation based on the keyword. This step S308 may be performed differently depending on whether or not the past history of the patient can be obtained.

According to some embodiments, in the case that the past history of the patient cannot be obtained, step S308 may include step S3082, i.e., a medical department suitable for the patient may be acquired from a pre-acquired disease knowledge base based on the current medical condition included in the keyword.

In an example, the disease knowledge base may be constructed by a medical practitioner, such as a doctor or an expert in the medical field. The disease knowledge base may include information on, for example, signs, sequelae, associations between different diseases, etc. associated with a particular disease. For example, where a medical records repository containing a large number of medical records (e.g., from a particular hospital or medical facility) can be accessed, relevant a priori medical knowledge for building a disease knowledge base can be obtained from the medical records repository. Thus, the corresponding clinical department can be retrieved from the disease knowledge base based on the keywords describing the specific disease symptoms. For example, when the keyword includes "stomachache" or "stomachache", information on "digestive medicine" may be acquired from the disease knowledge base.

In this way, in the case where the past history of the patient cannot be acquired, the medical department suitable for the patient can be determined by the prior medical knowledge contained in the disease knowledge base.

According to some embodiments, prior to step S308, the method 300 may further include step S307 for obtaining a past history of the patient and ranking the importance of the past history. Here, the keyword may further include personal information of the patient, and step S307 may include:

in step S3072, acquiring a past history of the patient from a pre-acquired medical record library based on the personal information included in the keyword, wherein the past history includes at least one past history entry, and each past history entry includes at least one of a disease history and a surgical history of the patient; and sorting the at least one past history item by importance degree to generate a sorted past history item at step S3074.

In this way, in the case where the past history of the patient can be acquired, it is possible to contribute to providing more accurate clinical department recommendations by taking the past history into consideration in combination. Meanwhile, the past history is sorted according to the importance degree, so that the accuracy of the recommendation of the diagnosis department can be further improved.

In an example, at step S3072, the pre-fetched medical records repository may be a medical records repository of a hospital or medical institution in which the patient is located. In an example, the medical records repository can include an Electronic Medical Records (EMR) repository and a Hospital Information System (HIS) repository. A large number of medical records contained in the medical records repository can be filtered to screen out poor quality medical records (e.g., medical records that are of poor form and/or connotation), thereby ensuring the quality of the medical records repository.

In step S3074, the past history is ranked based on such considerations: different diseases or surgeries may have different effects on the patient's subsequent life and treatment. For example, some diseases or surgeries may have serious sequelae that affect the patient's follow-up life. Some diseases may be chronic and require long-term treatment. Thus, the past history is ranked in importance to indicate different degrees of influence of different past histories on the patient.

In an example, each past history, i.e., each past history entry, can have a corresponding importance index to rank. The importance index may be determined based on medical attributes such as time, severity, etc. of the disease or surgery. In other words, the earlier the ranking, the greater the effect on the patient of the disease or procedure to which the prior history entry corresponds. For example, the past history of the patient may include a history of hypertension and a surgical history of tonsillectomy, and the past history entry corresponding to hypertension may be ranked more forward than the past history entry corresponding to tonsillectomy.

According to some embodiments, obtaining the past history of the patient from the pre-fetched medical records repository in step S3072 may include the steps of: collecting medical records of a patient from a medical record repository; and extracting a specified field indicating the past history in the medical record to acquire information on the at least one past history entry, wherein the specified field is subjected to normalization processing for normalization before being extracted.

In this way, normalization of the extracted prior history can be ensured by normalizing the fields of interest in cases where the medical records repository contains medical records across multiple different regions or medical records with different specifications.

In an example, designated fields indicating a disease history and a surgical history can be normalized according to relevant specifications in accordance with national standards.

According to some embodiments, in a case where the past history of the patient can be obtained, step S308 may include step S3084 of determining a list of at least one department applicable to the patient based on the current medical condition included in the keyword and the sorted past history entries, wherein the at least one department in the list is sorted by the degree of importance.

In this way, a potentially most likely disease history or surgical history associated with a patient's current condition may be mined in conjunction with the past history of the patient to improve the accuracy of determining the clinical department. In addition, in an emergency scene, the method can help to reduce the screening time required by emergency doctors and/or the consultation time of doctors in related departments, thereby avoiding missing the emergency opportunity.

In an example, a current potential illness may be retrieved from a knowledge base of illnesses as described above based on current signs, and at least one clinical department applicable to the patient may be determined based on the current potential illness and the illness reflected by past history. For example, a current condition of a patient may relate to gastric pain, however, in the case of knowing that the patient has a long-term history of cardiac disease from the past history of the patient, it may be that the gastric pain currently perceived by the patient is not associated with gastric disease, but rather with a long-term history of cardiac disease, as conditions caused by such a history of cardiac disease are sometimes similar to gastric pain and may be perceived by the patient as gastric pain. Therefore, by considering the past history of the patient, the disease which the patient is most likely to suffer from can be maximally discovered, and the corresponding medical department can be determined.

According to some embodiments, the method 300 may further include step S310 of generating a manifest for guiding the patient to emergency treatment based on the past history and the current symptoms included in the keywords.

In this way, more targeted emergency treatment recommendations may be provided to the patient by taking into account the past history of the patient. This may be particularly beneficial in emergency scenarios.

In an example, the generated manifest may be sent directly to the patient or his family members via a mobile device, such as a cell phone. Alternatively, the generated list may be first sent to the relevant emergency medical practitioner for screening or supplementing, etc., and then sent to the patient or family members. In the latter case, the operations for generating the checklist may have self-learning capabilities that may receive feedback from the emergency physician and adjust the algorithm used to generate the checklist based on the feedback.

As described above, according to the clinical department recommendation method of the embodiment of the present disclosure, by cascading the voice recognition step and the keyword extraction step so as to directly extract the interested keyword from the original recognition result without decoding the voice recognition result into the text, the voice recognition error due to the text decoding can be avoided, thereby improving the voice recognition effect specific to the medical field. Therefore, the information related to the possibly-existing disease of the patient can be accurately acquired from the disease description voice provided by the patient, and the accuracy of the recommendation of the medical department is improved.

Fig. 4 shows a schematic diagram of a medical department recommendation method according to an embodiment of the present disclosure in an emergency scenario.

As shown in fig. 4, in an emergency scenario, a patient may place an emergency call via a mobile device, such as a cell phone, to send out an emergency voice 405 to a hospital or other medical facility. The emergency speech 405 may contain information describing the patient's current signs.

After the emergency speech 405 is acquired, speech recognition 410 may be performed on the emergency speech 405. Speech recognition 410 may include both speech recognition and mood recognition. The language identification may include identifying content information in the emergency speech 405, i.e., identifying speech information of the emergency speech itself. Mood recognition may recognize non-content information in the rescue voice 405. For example, the patient has a slurried mouth condition during speaking of the emergency speech 405, and thus the recognition results may include information indicating that the patient may potentially have a stroke.

Speech recognition 410 may use pre-trained models. Training may be performed on the basis of the generic ASR model 412 by adding an adapter 414 for speech recognition of medical domain vocabulary and an adapter 416 for mood recognition at the output layer. Thereby, an enhanced model suitable for speech recognition and mood recognition in the medical field can be obtained.

Speech recognition 410 may obtain raw recognition results 418 of emergency speech 405 that are not decoded into text. Here, obtaining the original recognition result that is not decoded into text aims to avoid a speech recognition error due to text decoding.

In other words, the results output via speech recognition 410 may be fed directly downstream to perform language understanding 420. In language understanding 420, keyword extraction 422 may be performed to extract keywords of interest directly from raw recognition results 418. Here, the keywords of interest may include the patient's current symptoms 424 as well as the patient's personal information 426.

Based on the patient's personal information 426, where a medical records repository 430 associated with the patient (e.g., of a hospital or medical facility in which the patient is located) is accessible, a past history of the patient, such as a disease history, surgical history, etc., can be retrieved from the medical records repository 430. Retrieving the past history of the patient can include collecting the patient's medical record from the medical records repository 430 and extracting a specified field in the medical record indicating the past history.

If the past history of the patient is not retrieved in the medical records repository 430, a list of clinical departments 450 applicable to the patient may be determined from the disease knowledge base 440 based on the current symptoms 424. A priori medical knowledge in the disease knowledge base 440 can also be obtained from the medical records base 430.

If the past history of the patient is retrieved in the medical records repository 430, the related past histories can also be ranked by importance. In this case, the current potential illness retrieved from the illness knowledge base 440 based on the current symptoms 424 and the past history of the patient retrieved in the medical records base 430 can be considered together, thereby determining the list of medical departments 450 suitable for the patient. Here, the accuracy of determining the clinical department list 450 may be improved because the potentially most likely disease history or surgical history associated with the patient's current condition 424 can be mined through past history.

Information about the clinical department list 450 may be recommended to the emergency doctor to assist in the emergency treatment. In addition, a list for guiding the patient to emergency treatment may also be generated based on past history and current symptoms 424.

In the emergency scene as described above, thanks to the clinical department recommendation method according to the embodiment of the present disclosure, information related to current disease symptoms can be accurately extracted from emergency voices of patients for emergency treatment, and it is possible to help reduce the screening time required by emergency doctors and/or the time for consultation by relevant department doctors, thereby avoiding missing emergency opportunities.

According to another aspect of the disclosure, a diagnosis and treatment department recommending device is also provided.

Fig. 5 is a block diagram illustrating a structure of a medical department recommending apparatus 500 according to an embodiment of the present disclosure.

As shown in FIG. 5, the apparatus 500 includes a speech acquisition module 502, a speech recognition module 504, a feature extraction module 506, and a result determination module 508.

The voice acquisition module 502 is configured to acquire voice information describing a current condition of a patient to be diagnosed.

The speech recognition module 504 is configured to perform speech recognition on the speech information to obtain an original recognition result that is not decoded into words in the speech information.

The feature extraction module 506 is configured to extract keywords from the raw recognition results for determining the clinical department of the patient, wherein the keywords include the current signs of the patient.

The result determination module 508 is configured to determine a clinical department for recommendation based on the keywords.

According to some embodiments, the speech information may include mood information generated by the patient when speaking the speech, and the raw recognition results may include results of recognizing the mood information.

According to some embodiments, at least one of the speech recognition module 504 and the feature extraction module 506 includes a pre-trained model, wherein in the speech recognition module 504, the pre-trained model is a model obtained by adding at least one of an adapter for a medical field vocabulary and an adapter for mood recognition to an output layer in a general model for speech recognition for training, and in the feature extraction module 506, the pre-trained model is a model obtained by adding an adapter for a keyword of interest to an output layer in a general model for medical field semantic understanding for training.

The operations performed by the above-mentioned modules 502 to 508 may correspond to the steps S202 to S208 described with reference to fig. 2, and therefore, the details of the various aspects are not repeated here.

Fig. 6 shows a block diagram of a clinical department recommendation device 600 according to another embodiment of the present disclosure. Modules 602 through 608 shown in fig. 6 may correspond to modules 502 through 508, respectively, shown in fig. 5. In addition, the apparatus 600 and one or more of the modules 602 to 608 may further include further sub-functional modules, as will be described in detail below.

According to some embodiments, the result determination module 608 may include: a first result determination unit 6080 configured to acquire a medical department applicable to the patient from a pre-acquired disease knowledge base based on the current symptom included in the keyword.

According to some embodiments, the keywords further comprise personal information of the patient, and wherein the apparatus 600 further comprises: a past history acquisition module 610 configured to acquire a past history of the patient from a pre-acquired medical record repository based on the personal information included in the keyword, wherein the past history includes at least one past history entry, and each past history entry includes at least one of a disease history and a surgical history of the patient; and a ranking module 612 configured to rank the at least one past history entry by degree of importance to generate a ranked past history entry.

According to some embodiments, the past history acquisition module 610 may include: a collection unit 6100 configured to collect medical records of the patient from a medical records repository; and an extraction unit 6102 configured to extract a specified field indicating a past history in the medical record to acquire information on at least one past history entry, wherein the specified field is subjected to normalization processing for normalization before being extracted.

According to some embodiments, the result determination module 608 may include: a second result determination unit 6082 configured to determine a list of at least one clinical department applicable to the patient based on the current symptom included in the keyword and the sorted past history entry, wherein the at least one clinical department in the list is sorted by the degree of importance.

According to some embodiments, the apparatus 600 may further comprise: a list generation module 614 configured to generate a list for guiding the patient to emergency treatment based on the past history and the current symptoms included in the keywords.

According to another aspect of the present disclosure, there is also provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform a method according to an embodiment of the disclosure.

According to another aspect of the present disclosure, there is also provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method according to the embodiments of the present disclosure.

According to another aspect of the present disclosure, there is also provided a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method according to an embodiment of the present disclosure.

Referring to fig. 7, a block diagram of a structure of an electronic device 700, which may be a server or a client of the present disclosure, which is an example of a hardware device that may be applied to aspects of the present disclosure, will now be described. Electronic device is intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the electronic device 700 includes a computing unit 701, which may perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data required for the operation of the electronic device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

A plurality of components in the electronic device 700 are connected to the I/O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 may be any type of device capable of inputting information to the electronic device 700, and the input unit 706 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device, and may include, but is not limited to, a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and/or a remote controller. Output unit 707 may be any type of device capable of presenting information and may include, but is not limited to, a display, speakers, a video/audio output terminal, a vibrator, and/or a printer. Storage unit 708 may include, but is not limited to, magnetic or optical disks. The communication unit 709 allows the electronic device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers and/or chipsets, such as bluetooth (TM) devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and/or the like.

Computing unit 701 may be a variety of general purpose and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The computing unit 701 performs the various methods and processes described above, such as clinical department recommendations. For example, in some embodiments, the clinical department recommendation method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 700 via the ROM 702 and/or the communication unit 709. When loaded into RAM703 and executed by the computing unit 701, may perform one or more of the steps of the clinical department recommendation described above. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the clinical department recommendation method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server combining a blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be performed in parallel, sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related patient information all accord with the regulations of related laws and regulations, and do not violate the good custom of the public order.

While embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it is to be understood that the above-described methods, systems and apparatus are merely illustrative embodiments or examples and that the scope of the invention is not to be limited by these embodiments or examples, but only by the claims as issued and their equivalents. Various elements in the embodiments or examples may be omitted or may be replaced with equivalents thereof. Further, the steps may be performed in an order different from that described in the present disclosure. Further, various elements in the embodiments or examples may be combined in various ways. It is important that as technology evolves, many of the elements described herein may be replaced with equivalent elements that appear after the present disclosure.

Claims

1. A medical department recommendation method comprises the following steps:

acquiring voice information for describing current symptoms of a patient to be diagnosed;

carrying out voice recognition on the voice information to obtain an original recognition result which is not decoded into characters in the voice information;

extracting keywords for determining a medical department of the patient from the original recognition result, wherein the keywords comprise the current symptoms of the patient; and

determining the medical department for recommendation based on the keyword.

2. The method according to claim 1, wherein the speech information includes mood information generated by the patient when uttering speech, and the raw recognition result includes a result of recognizing the mood information.

3. The method according to claim 1 or 2, wherein at least one of the step of speech recognizing the speech information and the step of extracting keywords from the raw recognition result uses a pre-trained model, wherein,

in the step of performing speech recognition on the speech information, the pre-trained model is a model obtained by adding at least one of an adapter for medical field vocabulary and an adapter for mood recognition to an output layer in a general model for speech recognition to train,

in the step of extracting keywords from the original recognition result, the pre-trained model is a model obtained by adding an adapter for a keyword of interest to an output layer in a general model for semantic understanding in the medical field for training.

4. The method of any one of claims 1 to 3, wherein determining the clinical department for recommendation based on the keyword comprises:

acquiring the diagnosis and treatment department suitable for the patient from a pre-acquired disease knowledge base based on the current symptoms included in the keywords.

5. The method of any of claims 1-3, wherein the keyword further includes personal information of the patient, and wherein the method further comprises:

acquiring a past history of the patient from a pre-acquired medical record library based on the personal information included in the keyword, wherein the past history comprises at least one past history item, and each past history item comprises at least one of a disease history and a surgical history of the patient; and

the at least one past history item is ranked by importance to generate ranked past history items.

6. The method of claim 5, wherein obtaining the past history of the patient from a pre-acquired medical record repository comprises:

collecting medical records of the patient from the medical records repository; and

extracting a specified field indicating the past history in the medical record to acquire information about the at least one past history entry, wherein the specified field is subjected to normalization processing for normalization before being extracted.

7. The method of claim 5 or 6, determining the medical department for recommendation based on the keyword comprises:

determining a list of at least one clinical department applicable to the patient based on the current signs and the sorted past history items included in the keywords, wherein the at least one clinical department in the list is sorted by importance degree.

8. The method of any of claims 5 to 7, further comprising:

generating a list for guiding the patient to emergency treatment based on the past history and the current symptoms included in the keywords.

9. A clinical department recommendation device comprising:

the system comprises a voice acquisition module, a diagnosis module and a control module, wherein the voice acquisition module is configured to acquire voice information for describing current symptoms of a patient to be diagnosed;

the voice recognition module is configured to perform voice recognition on the voice information to acquire an original recognition result which is not decoded into characters in the voice information;

a feature extraction module configured to extract keywords for determining a diagnosis and treatment department of the patient from the raw recognition result, wherein the keywords include the current symptom of the patient; and

a result determination module configured to determine the medical department for recommendation based on the keyword.

10. The apparatus according to claim 9, wherein the speech information includes mood information generated by the patient when uttering speech, and the raw recognition result includes a result of recognizing the mood information.

11. The apparatus of claim 9 or 10, wherein at least one of the speech recognition module and the feature extraction module comprises a pre-trained model, wherein,

in the speech recognition module, the pre-trained model is a model obtained by adding at least one of an adapter for medical field vocabulary and an adapter for mood recognition to an output layer in a general model for speech recognition to train,

in the feature extraction module, the pre-trained model is a model obtained by adding an adapter for a keyword of interest to an output layer in a general model for semantic understanding in the medical field for training.

12. The apparatus of any of claims 9 to 11, wherein the result determination module comprises:

a first result determination unit configured to acquire the medical department applicable to the patient from a pre-acquired disease knowledge base based on the current symptom included in the keyword.

13. The apparatus of any of claims 9 to 11, wherein the keyword further comprises personal information of the patient, and wherein the apparatus further comprises:

a past history acquisition module configured to acquire a past history of the patient from a pre-acquired medical record library based on the personal information included in the keyword, wherein the past history includes at least one past history entry, each past history entry including at least one of a disease history and a surgical history of the patient; and

a ranking module configured to rank the at least one past history entry by degree of importance to generate a ranked past history entry.

14. The apparatus of claim 13, wherein the past history acquisition module comprises:

a collection unit configured to collect medical records of the patient from the medical records repository; and

an extracting unit configured to extract a specified field indicating the past history in the medical record to acquire information on the at least one past history entry, wherein the specified field is subjected to normalization processing for normalization before being extracted.

15. The apparatus of claim 13 or 14, the result determination module comprising:

a second result determination unit configured to determine a list of at least one clinical department applicable to the patient based on the current symptom included in the keyword and the sorted past history entry, wherein the at least one clinical department in the list is sorted by degree of importance.

16. The apparatus of any of claims 13 to 15, further comprising:

a list generation module configured to generate a list for guiding the patient to emergency treatment based on the past history and the current symptoms included in the keywords.

17. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-8.

19. A computer program product comprising a computer program, wherein the computer program realizes the method of any one of claims 1-8 when executed by a processor.