CN116959678A - First-aid guiding method, first-aid dispatching system, terminal equipment and storage medium - Google Patents

Abstract

The embodiment of the application discloses a first-aid guiding method, a first-aid dispatching system, terminal equipment and a storage medium, wherein the first-aid guiding method comprises the following steps: the first-aid call initiated by the terminal identification is connected, and the video call is initiated to the terminal identification; under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information; determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through video call.

Description

First-aid guiding method, first-aid dispatching system, terminal equipment and storage medium

Technical Field

The application relates to the technical field of intelligent medical treatment, in particular to a first-aid guiding method, a first-aid dispatching system, terminal equipment and a storage medium.

Background

In the prior art, for pre-hospital emergency, a traditional 120 emergency dispatching system is mostly adopted, after a receiver makes a call, emergency information is exchanged in a voice call mode, position information of a patient is determined, and the ambulance is dispatched to the scene for rescue according to the determined position information; the medical treatment and treatment scheme can be determined by evaluating the illness state according to the professional knowledge of an operator in the process of voice communication according to the medical treatment knowledge base provided by the medical treatment priority grading dispatch system (Medical Priority Dispatch System, MPDS). The two modes in the prior art for self-rescue or mutual rescue cannot be effectively mastered by operators in the specific situation of the emergency scene, and the modes in the prior art are seriously dependent on the professional knowledge and judgment of the operators, so that systematic medical guidance cannot be provided for the emergency scene before the ambulance arrives.

Disclosure of Invention

In view of the foregoing, it is desirable to provide an emergency guiding method, an emergency dispatching system, a terminal device and a storage medium, which can provide systematic medical guidance for an emergency scene before an ambulance arrives.

In order to achieve the above purpose, the technical scheme of the application is realized as follows:

in a first aspect, an embodiment of the present application provides a method for guiding first aid, the method including:

the first-aid call initiated by the terminal identification is connected, and the video call is initiated to the terminal identification;

under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information;

determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency;

and determining an emergency plan according to the emergency category, and outputting the emergency plan through the video call.

In a second aspect, an embodiment of the present application provides an emergency dispatch system, including:

the switching-on unit is used for switching on the first-aid call initiated by the terminal identifier and initiating a video call to the terminal identifier;

the system comprises an acquisition unit, a first-aid processing unit and a first-aid processing unit, wherein the acquisition unit is used for acquiring voice information of an emergency scene and image information of a patient to be emergency under the condition that a video call is received; extracting complaint symptom information from the voice information;

The determining unit is used for determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through the video call.

In a third aspect, an embodiment of the present application provides a terminal device, including: a processor, a memory, and a communication bus; the processor implements the first aid guiding method when executing the running program stored in the memory.

In a fourth aspect, an embodiment of the present application provides a storage medium having stored thereon a computer program which, when executed by a processor, implements the above-described emergency instruction method.

The embodiment of the application provides a first aid guiding method, terminal equipment and storage medium, wherein the method comprises the following steps: the first-aid call initiated by the terminal identification is connected, and the video call is initiated to the terminal identification; under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information; determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through video call. According to the implementation scheme, in the process of acquiring the emergency guiding method corresponding to the emergency site, under the condition of making a call to the emergency site, a video call request is initiated at the same time, voice information of the emergency site and image information of a patient to be emergency can be directly acquired in a video call mode, the current situation information of the patient to be emergency can be effectively mastered, the acquired voice information and the image information of the patient to be emergency can be identified, emergency categories corresponding to the voice information of the emergency site and the image information of the patient can be determined, and then the corresponding emergency pre-case can be directly matched in a preset emergency pre-case library according to the emergency categories, and systematic medical guidance can be provided for the emergency site according to the matched emergency pre-case before the ambulance reaches the emergency site.

Drawings

FIG. 1 is a flow chart illustrating an exemplary conventional 120 first aid dispatch system;

FIG. 2 is a flow chart illustrating an exemplary MPDS scheduling system;

FIG. 3 is a flowchart of a first method for guiding emergency treatment according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a process for training a pre-training multi-modal identification model and determining a class of emergency by using the pre-training multi-modal identification model according to an embodiment of the present application;

FIG. 5 is a two-dimensional non-linear diagram of a patient's disease level and a desired level of life support according to an embodiment of the present application;

fig. 6 is a flowchart of a second method for guiding emergency treatment according to an embodiment of the present application;

FIG. 7 is a flowchart III of another first aid guiding method according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of an emergency dispatch system 1 according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a terminal device 2 according to an embodiment of the present application.

Detailed Description

In order to more fully understand the features and technical content of the embodiments of the present application, the following detailed description of the embodiments of the present application is provided with reference to the accompanying drawings, which are not intended to limit the embodiments of the present application.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.

In the prior art, the traditional 120 first-aid dispatching system applied to the medical field is mainly used for dispatching work between ambulances and hospitals and is mainly responsible for dispatching the ambulances, the traditional 120 first-aid dispatching system mainly calls for help and guides through voice calls, the on-site situation is limited to know, and the instruction of rescue measures cannot be provided for patients before the ambulances arrive.

The main workflow of the traditional 120 first-aid dispatching system is shown in fig. 1, when 120 phones are called in, the exchanger prompts by voice and distributes the called 120 phones to an idle receiving dispatching desk, a receiving person knows information such as contact phones, addresses and the like of an emergency scene in a phone voice mode, dispatches an ambulance, dispatches the ambulance to a destination, monitors the arrival condition of the ambulance until the ambulance reaches the scene, and completes dispatching work.

Another system for dispatching priority of emergency treatment (Medical Priority Dispatch System, MPDS) in the united states is a medical expert system MPDS for dispatching of emergency treatment, which is mainly composed of Jeff clawsen (Jeff j. Clawson, international emergency dispatch institute-IAED chairman) in the united states, starts to study the dispatching of emergency treatment in the last 70 th century, is summarized through practice for many years and developed under the common efforts of other experts in the medical field in the united states, takes software as a main use form, and is an expert system for realizing standardized, medical treatment and grading treatment in the dispatching of emergency treatment, and has strict flow, precise design and complete matching system. The MPDS is an expert knowledge base system of emergency dispatcher, and by using the MPDS, the acceptance dispatcher can evaluate the severity of the patient according to the known site situation, so that the pre-hospital mortality is reduced, and meanwhile, good guidance and support are provided for other patients.

As shown in fig. 2, when 120 phones are called in, the exchanger prompts voice and distributes the called 120 phones to an idle receiving and dispatching desk, a receiving person knows information such as contact phones, addresses and the like of an emergency scene in a phone voice mode, acquires on-site medical information from the MPDS system, rapidly and scientifically provides on-site medical guidance, sends the on-site medical information to an ambulance vehicle-mounted information terminal, and the ambulance goes to a destination according to the received information; the system can also make hierarchical dispatching decisions according to the information of the emergency site, and after receiving the dispatching instruction, the system continues to guide the site, sends the site medical information to the vehicle-mounted information terminal of the ambulance, and the ambulance goes to the destination according to the received information, so that the ambulance arrives at the emergency site to finish dispatching work.

Although the MPDS system in the united states can be used for guiding the rescue operation before the emergency ambulance arrives, the disease contained in the MPDS system is different from the national conditions of the united states, many common diseases in the country are not related to the national conditions of the country, such as gas poisoning, etc., the MPDS system is high in maintenance cost, and when the existing MPDS dispatching system is used for carrying out emergency grade classification, the MPDS system is seriously dependent on the professional knowledge and judgment of operators, and the requirements on the training and skills of the operators are high.

Therefore, the above conventional 120 emergency dispatch system and the MPDS system in the united states have the following technical problems:

(1) The existing emergency dispatch method is mainly realized by voice communication, the specific condition of the emergency site and the current situation of the caller cannot be comprehensively known only by the way of knowing the site condition by voice information, and misjudgment on the emergency condition is easy to occur;

(2) The existing emergency dispatching method mainly acquires the position information of the caller in a voice mode, can not quickly and accurately position the position information of the caller, delays the rescue time, and can not update the position of the ambulance to the caller in real time after the ambulance is dispatched;

(3) The existing emergency dispatching method is mainly used for dispatching work of ambulances and hospitals or depends on an MPDS system in the United states, is not suitable for a domestic emergency expert plan library, carries out hierarchical dispatching on different emergency conditions according to the emergency conditions, and easily causes waste of emergency resources; and when the existing MPDS dispatching system performs emergency grade classification, the special knowledge and judgment of operators are seriously relied on, and the requirements on training and skills of operators are high.

Therefore, the technical scheme in the prior art cannot effectively master the specific situation of the rescue scene, cannot provide systematic medical guidance for the person calling for help before the ambulance arrives at the scene, and cannot save oneself or save each other due to the lack of systematic guidance of the person calling for help.

To solve the above problems, an embodiment of the present application provides a first aid guiding method, as shown in fig. 3, which may include:

s101, switching on an emergency call initiated by a terminal identifier, and initiating a video call to the terminal identifier.

In the embodiment of the application, the first-aid guiding method is applied to a first-aid dispatching system based on 5G information, wherein the 5G information is an upgrade of short message service, is a basic telecommunication service of an operator, is a leap for realizing service experience based on IP technology, supports more media formats and has more expression forms.

In the embodiment of the application, after a caller dials a 120 emergency call through a terminal device, an exchange receives a 120 emergency call request, the exchange distributes the received 120 emergency call request to an idle acceptance dispatching desk through voice prompt, and after an acceptance person corresponding to the acceptance dispatching desk makes a 120 emergency call dialed by the caller, the acceptance personnel initiates a video call request to the terminal device held by the caller through a 5G message.

The receiving person initiates the video call request to the terminal device held by the caller, or initiates the video call request to the terminal device held by the caller through the acquired telephone information according to the acquired telephone information of the caller.

In the embodiment of the present application, the specific type of the terminal device held by the caller is not specifically limited in the present application, and may be any user device, for example, any device with communication and storage functions, such as a smart phone, a personal computer (Personal Computer, PC), a notebook computer, a tablet computer, and a portable wearable device.

S102, under the condition that a video call is received, acquiring voice information of an emergency site and image information of a patient to be emergency; and extracting complaint symptom information from the voice information.

In the embodiment of the application, after the acceptors initiate the video call request to the caller, the caller receives the video call request through the held terminal equipment, and under the condition that the video call request is received, the acceptors can exchange the current concrete situation of the emergency scene between the connected video call and the caller.

In the embodiment of the application, after the caller receives the video call, the acceptors can receive the video information of the patient to be emergency cured, the image information of the patient to be emergency cured and the voice information of the emergency cured through the 5G message.

It should be noted that, the image information of the patient to be emergency in the emergency scene may be the image information of the patient to be emergency photographed by the 5G message, or the image information of the patient to be emergency is extracted from the obtained video information of the patient to be emergency and sent by the 5G message, specifically, the image information may be selected according to the actual situation, which is not limited in the present application.

It should be noted that, after the caller receives the video call, the receiver and the caller can describe the emergency site situation, the patient symptoms or the reasons of the accident in real time along with the video call, and the background server of the emergency dispatch system based on the 5G message can also receive the voice information of the emergency site and the video information of the patient to be emergency or the image information of the patient to be emergency in real time.

If the video information of the patient to be first-aid in the first-aid site is obtained, the image information of the patient to be first-aid needs to be extracted from the obtained video information of the patient to be first-aid, and then subsequent processing is carried out; if the image information of the patient to be emergency is directly obtained, the subsequent steps are directly executed.

Taking the example of obtaining the video information of the patient to be first-aid at the first-aid site, under the condition that the video request is received, after obtaining the voice information of the first-aid site and the video information of the patient to be first-aid, the method performs information extraction processing on the obtained voice information of the first-aid site and the obtained video information of the patient to be first-aid, and obtains the image information and the complaint symptom information of the patient to be first-aid.

In the embodiment of the application, the received voice information is converted into character information through a pre-trained voice-to-character network model, and the converted character information is subjected to complaint and symptom information extraction through a pre-trained named entity recognition network model, so that key complaint symptom information is obtained.

After obtaining the key complaint symptom information, filling the obtained key complaint symptom information into a complaint information column in the first-aid dispatching system to generate the complaint information in the dispatching list of the first-aid dispatching system 120, so that the time of manual input can be saved, and the information filling mode is more efficient.

It should be noted that the pre-trained voice-to-text network model and the pre-trained named entity recognition network model may be selected according to actual situations, which is not specifically limited in the present application.

In the embodiment of the application, the received video information of the patient to be emergency treated is subjected to key frame image extraction through a key frame extraction network model to obtain the image information of the patient to be emergency treated.

It should be noted that, the key frame extraction network model may be selected according to practical situations, which is not specifically limited in the present application.

S103, determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency.

In the embodiment of the application, after the on-site voice information of emergency treatment is processed to obtain the complaint symptom information and the video information of the patient to be emergency treated is processed to obtain the image information of the patient to be emergency treated, the corresponding emergency treatment category can be determined according to the obtained image information of the patient to be emergency treated and the complaint symptom information.

It should be noted that each emergency category corresponds to an emergency priority.

In the embodiment of the application, when the emergency category corresponding to the image information and the complaint symptom information of the patient to be emergency is determined, the image information and the complaint symptom information of the patient to be emergency are input into a pre-training multi-mode identification model, and the emergency category is output, wherein the pre-training multi-mode identification model is obtained by training by utilizing the historical complaint symptom information and the historical image information of the patient to be emergency in an emergency center dispatch list.

In the embodiment of the application, as shown in fig. 4, the training is performed by using the historical complaint symptom training data, the test data and the historical image information training data and the test data of the patient to be first-aid in the first-aid center schedule, respectively using the CNN neural network and the Bert neural network to obtain a pre-training multi-mode recognition model, extracting the text information of the acquired first-aid site, inputting the acquired text information and the image information of the patient to be first-aid into the pre-training multi-mode recognition model to judge, and outputting the corresponding first-aid category.

In the embodiment of the application, when the pre-training multi-mode recognition model is utilized to output the emergency category, before the image information of the patient to be emergency and the complaint symptom information are input into the pre-training multi-mode recognition model, training is firstly carried out by the pre-training multi-mode recognition model, and training is carried out by utilizing the historical complaint symptom information and the historical patient image information to be emergency in the emergency center schedule, wherein the training can be carried out by utilizing the historical complaint symptom information and the historical patient image information to be emergency in the emergency scene in the 120 emergency center schedule according to 8:2, the proportion is randomly divided into training data and test data respectively.

In the embodiment of the application, the obtained training data and test data are marked and cleaned by emergency specialists aiming at the data in the emergency plan library, the corresponding text information and the belonging emergency category are marked by the historical complaint voice information, and the belonging emergency category is marked by the historical patient image information to be emergency.

It should be noted that the cleaning function of the training data and the test data is to remove some interference information, such as some false information, and the finally obtained data is real data with reference value.

It should be noted that, the proportion of dividing training data and test data for the first-aid site historic complaint symptom information and the historic patient image information to be first-aid in the first-aid dispatching center dispatching list can be selected according to actual situations, and the application is not limited specifically.

In the embodiment of the application, firstly, the divided training data is utilized to carry out multi-mode recognition model training, after mask processing is carried out on the injured part in the image of the patient to be first-aid, the image is input into a convolutional neural network CNN for pre-training, and a pre-trained image feature extraction network of the patient to be first-aid is obtained; and meanwhile, after mask masking is carried out on key complaint symptom information in the extracted text information, pretraining is carried out by using a Bert network, and a pretrained complaint symptom information feature extraction network is obtained.

It should be noted that, by performing mask training on the image information of the patient to be first-aid and the key complaint information in the text information extracted from the voice information, the network obtained after training can be more sensitive to the content of the mask part, and in the subsequent multi-mode recognition model, the content and the characteristics of the key information provided by the patient to be first-aid are more focused.

It should be noted that, the pre-trained image feature extraction network of the patient to be first-aid and the pre-trained complaint symptom information feature extraction network may be selected according to actual situations, and are not limited to the convolutional neural network CNN and the Bert neural network in the present application.

In the embodiment of the application, after a pre-trained complaint symptom information feature extraction network is obtained, text information in test data is extracted to obtain the complaint symptom information feature, and training and fine-tuning are performed on the pre-trained complaint symptom information feature extraction network; meanwhile, after the pre-trained image feature extraction network of the patient to be emergency trained is obtained, the image information of the patient to be emergency trained in the test data is input into the pre-trained image feature extraction network of the patient to be emergency trained, the image feature of the patient to be emergency trained is extracted, and training and fine adjustment are carried out on the pre-trained image feature extraction network of the patient to be emergency trained.

In the embodiment of the application, after the characteristic of the complaint symptom information and the characteristic of the image of the patient to be first-aid are extracted, the extracted characteristic of the complaint symptom information and the characteristic of the image of the patient to be first-aid are subjected to weighted fusion, then the first-aid category recognition result is output through the full-connection layer, and the model parameters are continuously optimized until the model parameters are stable, so that the pre-training multi-mode recognition model is obtained.

In the embodiment of the application, a pre-training multi-mode recognition model is used for prediction, the acquired image information of the patient to be first-aid in the first-aid scene and the complaint symptom information extracted from the voice information are input into the pre-training multi-mode recognition model, and the corresponding first-aid class recognition result can be obtained through feature extraction of the image of the patient to be first-aid and feature extraction and fusion of the obtained complaint symptom information.

In the embodiment of the application, the on-site voice information of emergency and the image information of the patient to be emergency are used for input identification, so that the on-site situation can be judged and identified more effectively and conveniently, and the on-site situation can be processed more accurately and efficiently.

S104, determining an emergency plan according to the emergency category, and outputting through the video call.

In the embodiment of the application, the emergency category, the emergency priority corresponding to the emergency category and the corresponding emergency plan are stored in the emergency plan library system. After the image information and the complaint symptom information of the patient to be emergency-treated at the emergency site are input into the pre-training multi-mode identification model to obtain the corresponding emergency category, the corresponding emergency plan can be matched in the emergency plan library according to the obtained emergency category.

It should be noted that, the first-aid category stored in the first-aid plan library system may correspond to A, B, C, D, E, F, where the priority of the category F is highest, the priority of the category a is lowest, and may be selected according to the actual situation, which is not specifically limited in the present application.

In the embodiment of the application, before the ambulance arrives at the rescue scene, the acceptors can pacify the patient to be first-aid by using the connected video call, and instruct the scene personnel to carry out corresponding first-aid operation on the patient to be first-aid according to the matched treatment advice of the first-aid plan, such as cardiopulmonary resuscitation, use instruction of an automatic external defibrillator (Automated External Defibrillator, AED) and the like, until the ambulance arrives at the scene.

It can be appreciated that in the emergency guiding method provided by the embodiment of the application, in the process of obtaining the emergency guiding method corresponding to the emergency site, a video call request is initiated at the same time under the condition of making a call to the emergency site, the voice information of the emergency site and the image information of the patient to be emergency can be directly obtained in a video call mode, the current status information of the patient to be emergency can be effectively mastered, the obtained voice information and the image information of the patient to be emergency can be identified, the emergency category corresponding to the voice information of the emergency site and the image information of the patient can be determined, and then the corresponding emergency pre-case can be directly matched in a preset emergency pre-case library according to the emergency category, and before the ambulance reaches the emergency site, the systemized medical guidance can be provided for the emergency site according to the matched emergency pre-case.

Optionally, in case the video request is received, acquiring location information of the patient to be first-aid; after the emergency category is determined, matching corresponding rescue vehicle identifications from a preset emergency plan library according to the emergency category; based on the rescue vehicle identification, the position information is sent to the corresponding rescue vehicle so that the rescue vehicle can conduct route planning according to the position information; and acquiring the position information of the rescue vehicle and sending the position information of the rescue vehicle to the terminal identifier.

In the embodiment of the application, after the caller receives the video call request on the held terminal, the acceptors can also acquire the current position information of the patient to be first-aid through the 5G message.

In the embodiment of the application, under the condition of initiating video call to a terminal identifier, if the permission of the terminal to allow the position access is detected to be preset, acquiring the position information by using a Global Positioning System (GPS); if the terminal is detected not to set the permission for allowing the position access, a request for acquiring the position permission is initiated to the terminal identifier, and the position information is acquired by using a Global Positioning System (GPS) under the condition that a response for agreeing to the request is received.

In the embodiment of the application, when a receiver sends out a video call, if detecting that the terminal identifier held by the caller has set the authority of allowing the 5G message to use the position information, the position information of the patient to be first-aid is directly acquired through the GPS, if detecting that the terminal identifier held by the caller has not set the authority of allowing the 5G message to use the position information, the receiver initiates the video call, and initiates a request popup window for acquiring the position authority, and after the caller agrees with the request information on the terminal, the position information of the patient to be first-aid is acquired through the GPS, so that the specific position information of the patient to be first-aid is acquired according to the 5G message.

In the embodiment of the application, after the emergency category is determined according to the image information and the complaint symptom information of the patient to be emergency, the corresponding emergency vehicle identification information can be matched in the emergency plan library according to the emergency category.

It should be noted that different emergency categories in the emergency plan library correspond to different emergency vehicle identifications, i.e. the emergency vehicle resources corresponding to the different emergency categories are different.

It should be noted that, the manner of determining the emergency category may refer to step S103, which is not described herein.

In the embodiment of the application, after the corresponding emergency vehicle identification information is determined, the corresponding emergency vehicle dispatching is carried out, the acquired position information of the patient to be emergency is synchronously transmitted to the corresponding emergency vehicle, and after the emergency vehicle acquires the position information of the patient to be emergency, the emergency vehicle is matched with a qualified hospital closest to the patient to be emergency according to the position information, and an optimal route is planned in combination with the current traffic condition.

In the embodiment of the application, after the ambulance starts, the first-aid dispatching system synchronously sends the information of the ambulance to the terminal identifier.

It should be noted that the information of the rescue vehicle may include a license plate number of the rescue vehicle, a doctor phone following the rescue vehicle, a real-time position of the rescue vehicle, and the like.

Optionally, in the embodiment of the present application, the voice information of the emergency site and the image information of the patient to be emergency may be synchronized to the terminal device of the hospital, so that the hospital may prepare corresponding medical resources according to the received image information and voice information of the patient to be emergency.

In the embodiment of the application, under the condition that the initiated video call is received, all the voice information, the video information of the patient to be first-aid and the image information of the patient to be first-aid acquired through the 5G message can be automatically sent to the terminal equipment of the treatment hospital, and after the voice information, the video information of the patient to be first-aid and the image information of the patient to be first-aid are received, the doctor can perform disease condition assessment in advance through the information displayed on the terminal equipment of the hospital to prepare relevant medical resources such as an operating room, medicines and the like.

It should be noted that, if a special situation occurs in the emergency site, the emergency dispatch system may also notify the relevant personnel to arrive at the emergency site for processing, and synchronize the relevant information to the relevant personnel.

In the embodiment of the application, after the ambulance arrives at the first-aid site to receive the first-aid patient, the acceptors or the caller can hang up the video call, at the moment, the medical staff on the ambulance communicates with the doctor of the first-aid hospital, and according to the guidance of the doctor of the first-aid hospital, the medical staff can carry out the best rescue on the ambulance until the first-aid patient is sent to the first-aid hospital by the ambulance, so that the whole dispatching process is completed.

It should be noted that, through this first aid dispatch system, can really accomplish first aid, from answering the telephone, whole dispatch system is the relay of a first aid, accepts the people to the ambulance from the dispatch center, goes to the whole flow of hospital again and carries out the rescue, best effort protection wait for first aid patient's life safety.

Optionally, before the image information of the patient to be first-aid and the complaint symptom information are input into the pre-training multi-mode recognition model, a plurality of preset first-aid categories can be acquired, acquiring historical image information and historical complaint symptom information of a patient to be first-aid corresponding to a plurality of preset first-aid categories respectively; respectively inputting historical image information and historical complaint symptom information of a patient to be first-aid corresponding to a plurality of preset first-aid categories into a multi-mode identification model, and outputting a plurality of second first-aid categories corresponding to the plurality of preset first-aid categories; training the multi-modal identification model by utilizing a plurality of preset first emergency categories and a plurality of second emergency categories to obtain a pre-trained multi-modal identification model.

In the embodiment of the application, aiming at the common emergency call reasons in China and China, the emergency call reasons are classified into 9 major categories and 84 minor categories, the current reasons and conditions of the emergency call ambulance before national hospital can be covered comprehensively, and all emergency categories are classified into A, B, C, D, E, F five categories according to the patient disease grade and the required life support grade, wherein F represents the most urgent, the scheduling priority is the highest, and a two-dimensional nonlinear schematic diagram of the patient disease grade and the required life support grade is shown as 5.

In the embodiment of the present application, the vertical axis in fig. 5 represents the patient disease level, and is divided into four classes, class iv: refers to the situation that the patient is endangered and possibly endangers life at any time, and the patient immediately enters a resuscitation room or a rescue room for rescue; III grade: refers to the critical or rapid deterioration of the patient's condition and life-threatening; stage II: refers to the urgent illness state of patients with potential life threat; stage I: it means that the patient is in a non-emergency condition and the disease degree is general. The horizontal axis represents the level of life support required, where life support can be divided into three stages: basic life support BLS, advanced life support ALS and extended life support PLS; A. b, C, D, E, F are five levels of emergency category.

In addition, the main treatment measure for basic life support is free-hand cardiopulmonary resuscitation, the Haimrick method can be implemented on the person with airway obstruction, and the contents of wound first aid in a wider sense are sometimes classified into basic life support, and basic life support can be carried out through first aid training generally.

Advanced life support is the establishment and maintenance of more efficient ventilation and blood circulation, typically performed by a healthcare professional, using auxiliary equipment and special techniques such as electrocardiographic monitoring, defibrillators, artificial respirators, and drugs, while continuing BLS on a basic life support basis.

By life support is meant a series of measures taken against primary morbidity or resuscitation complications after recovery of spontaneous circulation in a patient, including correction of hypotension/shock, correction of water, electrolyte or acid-base imbalance, prevention of renal failure, improvement of cardiopulmonary function and systemic perfusion, in particular cerebral perfusion, prevention of cerebral hypoxia and cerebral oedema.

In the embodiment of the application, the emergency call scenes of 9 major categories and 84 minor categories in the emergency plan library are classified according to the emergency priority classification rule to form corresponding emergency dispatch codes.

It should be noted that, each first-aid dispatch code corresponds to one first-aid category and one dispatch priority.

In the embodiment of the application, when the training of the pre-training multi-mode identification model is performed, first, a plurality of first-aid dispatching codes corresponding to different categories, and the historical image information and the historical complaint symptom information of the patient to be first-aid corresponding to the first-aid dispatching codes are obtained, the historical image information and the historical complaint symptom information of the patient to be first-aid corresponding to the first-aid dispatching codes of different categories are respectively trained, the identification category corresponding to each preset first-aid category is output, and the multi-mode identification model is continuously optimized and trained according to the relation between the preset first-aid category and the identification category, so that the pre-training multi-mode identification model is obtained.

It should be noted that, the training process for the historical image information and the historical complaint symptom information of the patient to be first-aid corresponding to different categories may refer to the training process in S103, which is not described herein.

The scheduling codes are 1-2-F-1, and represent a first major class, a second minor class, the scheduling priority is F, and the last 1 represents a first scene of the codes of 1-2-F, and a data set is formed by collecting image information of a patient to be emergency and emergency site voice information corresponding to each scheduling code scene, so that the scheduling codes can be used for training a pre-training multi-mode identification model, and the pre-training multi-mode identification model is used for carrying out artificial intelligent identification and judgment on the emergency site to obtain the emergency category corresponding to the scheduling codes.

In the embodiment of the application, in order to avoid errors of artificial intelligence in the emergency category identification process, a detailed inquiry flow is formulated for each dispatching code in the emergency plan library, the emergency dispatching system can automatically prompt different inquiry questions according to the answers of the caller, and about 3-6 dialogs are needed for each emergency event corresponding to the patient to be emergency, wherein when the patient to be emergency is serious, the shorter the judgment time is, the faster the vehicle is dispatched, and the life of the patient to be emergency is more beneficial to rescuing. In the whole emergency call acceptance process, the acceptors strictly follow the information in the expert plan library, do not occupy excessive time for inquiry, do not leak important information, and the problems prepared in the expert plan library are shallow and easy to understand, so that the medical problems can not be asked by the ordinary people by using special medical terms, and meanwhile, the language habits of China are met.

In the embodiment of the application, the emergency dispatching system merges the illness states which cannot be clearly judged into the higher grade of the emergency category, requires to dispatch the ambulance to the position of the patient to be emergency for treatment as soon as possible, can avoid the occurrence of some medical disputes, and simultaneously, provides 22 emergency or mutual emergency treatment plans for the emergency category corresponding to different emergency reasons, corresponds to each emergency reason, and is used for acceptors to guide the emergency scene to truly realize emergency.

By means of the emergency dispatch system, waste of emergency resources is avoided, emergency patients can be guaranteed to send vehicles in time preferentially, and emergency site personnel are guided in a video call mode to rescue lives of patients. Meanwhile, the waste of emergency resources and aggravation or death of patients to be emergency caused by no vehicle to be dispatched or slow dispatch are avoided, so that complaints, medical disputes and the like are caused.

By absorbing international advanced first-aid management experience and method, a unified national expert plan library is established, the national first-aid standardization process is quickened, the classification disorder of the national first-aid symptoms is improved, the unified standard specification is not available, a scientific ambulance classification guiding principle and method are provided, the first-aid resources are conveniently and reasonably utilized, a comprehensive acceptance scheduling quality monitoring management method is provided, and scientific statistical data is utilized to provide decision basis for the first-aid center post setting and the first-aid workflow adjustment.

It should be noted that, based on the emergency dispatch system of 5G message, call for help, dispatch and guide through the 5G message, establish the expert's plan library of first aid that suits the national situation of China at the same time, add artificial intelligence and assist and discern, through the multimode recognition model, carry on the automatic recognition and match of the category of plan, compare with traditional emergency dispatch system, this system can be through the voice information, wait first aid patient's image, wait first aid patient's video information, position information etc. judge the situation of first aid scene at the same time, more comprehensive, more accurate, expert's plan library is more suitable for the national situation of China at the same time, carry on the automatic recognition through the multimode recognition model, lighten the dependence to the specialized dispatcher.

Based on the above embodiment, the first-aid guiding method provided in the present application, as shown in fig. 6, specifically includes the following steps:

step 1, through an emergency dispatch system based on a 5G message, the acquired voice information of an emergency site, video information of a patient to be emergency, image information of the patient to be emergency and the position information of the patient to be emergency are obtained;

step 2, inputting video information of a patient to be emergency, image information of the patient to be emergency and voice information of an emergency site into a pre-training multi-mode recognition model; judging the first-aid dispatching priority and the first-aid on-site first-aid plan;

Step 3, utilizing video call, and guiding emergency site personnel to rescue the patient to be emergency by the acceptors according to the emergency site emergency plan;

and step 4, synchronizing the position information of the patient to be first-aid to the ambulance for the ambulance to carry out route planning, acquiring the position information of the ambulance, and sending the position information to a terminal mark held by the patient to be first-aid.

It should be noted that the execution sequence of the step 3 and the step 4 is not sequential, and may be specifically selected according to the actual situation, and the present application is not specifically limited.

Based on the above embodiment, another first aid guiding method provided in the present application, as shown in fig. 7, specifically includes the following steps:

step 1, initiating 120 a call for help through a terminal identifier; initiating a video call to the terminal identifier;

step 2, receiving voice information of an emergency scene, image information of a patient to be emergency and position information of the patient to be emergency based on the 5G message under the condition that the video call is received;

step 3, extracting the main complaint symptom information of the received voice information, inputting the extracted main complaint symptom information and the acquired image information of the patient to be first-aid into a pre-training multi-mode recognition model, and outputting the first-aid category;

Step 4, determining an emergency plan according to the emergency category; outputting the determined first-aid plan through video call;

step 5, matching corresponding ambulance identifications according to the determined first-aid categories, and sending the position information to the corresponding ambulance based on the ambulance identifications so as to enable the ambulance to carry out route planning according to the position information;

step 6, acquiring the position information of the rescue vehicle, and sending the position information of the rescue vehicle to a terminal mark for initiating 120 telephone call for help;

and 7, monitoring the arrival condition of the rescue vehicle until the rescue vehicle arrives at the emergency site, and ending the flow.

It should be noted that, the execution steps of step 4 and steps 5-7 are parallel execution relationships, and the execution sequence is not divided.

Based on the above-described embodiments, in another embodiment of the present application, there is provided an emergency dispatch system 1, as shown in fig. 8, the emergency dispatch system 1 including:

and the switching-on unit 10 is used for switching on the emergency call initiated by the terminal identifier and initiating the video call to the terminal identifier.

An acquisition unit 11 for acquiring voice information of an emergency scene and image information of a patient to be emergency in the case where a video call is received; and extracting complaint symptom information from the voice information.

A determining unit 12 for determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through video call.

Optionally, the acquiring unit 11 is further configured to acquire location information of the patient to be first-aid, in a case where the video call is received.

Optionally, the first-aid dispatching system 1 further comprises: a matching unit for matching the number of the first and second units,

and the matching unit is used for matching corresponding ambulance identifications from a preset ambulance protocol library according to the first-aid category after the first-aid category is determined.

Optionally, the first-aid dispatching system 1 further comprises: the transmitting unit is configured to transmit the data,

the sending unit is used for sending the position information to the corresponding rescue vehicle based on the rescue vehicle identification so as to enable the rescue vehicle to conduct route planning according to the position information; and acquiring the position information of the rescue vehicle and sending the position information of the rescue vehicle to the terminal identifier.

Optionally, the first-aid dispatching system 1 further comprises: an output unit, which is used for outputting the output signals,

the output unit is also used for inputting the image information and the complaint symptom information of the patient to be first-aid into the pre-training multi-mode identification model and outputting the first-aid category; the pre-training multi-mode identification model is obtained by training historical complaint symptom information and historical patient image information to be first-aid in a first-aid center schedule.

Optionally, the first-aid dispatching system 1 further comprises: the synchronization unit is used for synchronizing the data,

the synchronous unit is used for synchronizing the voice information of the emergency scene and the image information of the patient to be emergency to the terminal equipment of the hospital, so that the hospital can be used for preparing corresponding medical resources according to the received image information and voice information of the patient to be emergency.

Optionally, the acquiring unit 11 is further configured to acquire a plurality of preset first emergency categories, and acquiring historical image information and historical complaint symptom information of the patient to be first-aid corresponding to the plurality of preset first-aid categories respectively.

Optionally, the output unit is further configured to input historical image information and historical complaint symptom information of the patient to be first-aid respectively corresponding to the plurality of preset first-aid categories into the multi-mode identification model respectively, and output a plurality of second first-aid categories corresponding to the plurality of preset first-aid categories.

Optionally, the terminal device 1 further includes: the training unit is used for training the training device,

the training unit is used for training the multi-modal identification model by utilizing a plurality of preset first emergency categories and a plurality of second emergency categories to obtain a pre-trained multi-modal identification model.

Optionally, the acquiring unit 11 is further configured to, when initiating a video call to the terminal identifier, acquire the location information by using a global positioning system GPS if it is detected that the terminal preset permission to allow the location access; if the terminal is detected not to set the permission for allowing the position access, initiating a request for acquiring the position permission to the terminal identifier; and acquires the location information using the global positioning system GPS in the case of receiving a response to the consent request.

The first-aid dispatching system provided by the embodiment of the application is used for switching on the first-aid call initiated by the terminal identifier and initiating the video call to the terminal identifier; under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information; determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through video call. Therefore, in the emergency dispatch system provided by the embodiment of the application, in the process of acquiring the emergency instruction method corresponding to the emergency site, under the condition of switching on an emergency call, a video call request is initiated at the same time, the voice information of the emergency site and the image information of the patient to be emergency can be directly acquired in a video call mode, the current situation information of the patient to be emergency can be effectively mastered, the voice information and the image information of the patient to be emergency can be identified, the emergency category corresponding to the voice information of the emergency site and the image information of the patient can be determined, and then the corresponding emergency instruction can be directly matched in the preset emergency instruction library according to the emergency category, and before the ambulance reaches the emergency site, the systemized medical instruction can be provided for the emergency site according to the matched emergency instruction.

Fig. 9 is a schematic diagram of a composition structure of a terminal device 2 according to an embodiment of the present application, and in practical application, the terminal device 2 according to the present embodiment includes a processor 20, a memory 21 and a communication bus 22 as shown in fig. 9 under the same disclosure concept of the above embodiment.

In a specific embodiment, the switching-on unit 10, the acquiring unit 11, the determining unit 12, the matching unit, the transmitting unit, the output unit, the synchronization unit, and the training unit may be implemented by a processor 20 located on the terminal device 2, where the processor 20 may be at least one of an ASIC (application specific integrated circuit ), a DSP (digital signal processor, digital Signal Processor), a DSP (digital signal processor), a DSPD, digital Signal Processing Device), a PLD (programmable logic device ), a FPGA (field programmable gate array, field Programmable Gate Array), a CPU, a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronics for implementing the above-described processor functions may be other for different devices, and the present embodiment is not particularly limited.

In the embodiment of the present application, the communication bus 22 is used to implement connection communication between the processor 20 and the memory 21; the processor 20 implements the following emergency guidance method when executing the operation program stored in the memory 21:

the first-aid call initiated by the terminal identification is connected, and the video call is initiated to the terminal identification; under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information; determining the emergency category according to the image information and the complaint symptom information of the patient to be emergency; and determining an emergency plan according to the emergency category, and outputting the emergency plan through video call.

Further, the processor 20 is further configured to obtain location information of the patient to be first-aid when the video call is received; after the emergency category is determined, matching corresponding rescue vehicle identifications from a preset emergency plan library according to the emergency category; based on the rescue vehicle identification, the position information is sent to the corresponding rescue vehicle so that the rescue vehicle can conduct route planning according to the position information; and acquiring the position information of the rescue vehicle and sending the position information of the rescue vehicle to the terminal identifier.

Further, the processor 20 is further configured to input the image information and the complaint symptom information of the patient to be first-aid into the pre-training multi-mode recognition model, and output the first-aid category; the pre-training multi-mode identification model is obtained by training historical complaint symptom information and historical patient image information to be first-aid in a first-aid center schedule.

Further, the processor 20 is further configured to synchronize the voice information of the emergency site and the image information of the patient to be emergency to the terminal device of the hospital, so that the hospital prepares corresponding medical resources according to the received image information and voice information of the patient to be emergency.

Further, the processor 20 is further configured to obtain a plurality of preset first emergency categories, and obtain historical image information and historical complaint symptom information of the patient to be emergency corresponding to the plurality of preset first emergency categories respectively; respectively inputting historical image information and historical complaint symptom information of a patient to be first-aid corresponding to a plurality of preset first-aid categories into a multi-mode identification model, and outputting a plurality of second first-aid categories corresponding to the plurality of preset first-aid categories; training the multi-modal identification model by utilizing a plurality of preset first emergency categories and a plurality of second emergency categories to obtain a pre-trained multi-modal identification model.

Further, the processor 20 is further configured to, when it is detected that the terminal sets permission to allow the location access in advance, obtain location information using a global positioning system GPS under the condition that a video call is initiated to the terminal identifier; if the terminal is detected not to set the permission for allowing the position access, initiating a request for acquiring the position permission to the terminal identifier; and acquiring location information using a global positioning system GPS in case a response is received agreeing to the request.

Based on the above embodiments, the embodiments of the present application provide a storage medium having stored thereon a computer program, where the computer readable storage medium stores one or more programs, where the one or more programs are executable by one or more processors and applied to a terminal device, where the computer program implements a data processing method as described above.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present disclosure may be embodied essentially or in a part contributing to the related art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), including several instructions for causing an image display device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present disclosure.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. A method of emergency instruction for application to an emergency dispatch system, the method comprising:

The first-aid call initiated by the terminal identification is connected, and the video call is initiated to the terminal identification;

under the condition that the video call is received, acquiring voice information of an emergency scene and image information of a patient to be emergency; extracting complaint symptom information from the voice information;

determining an emergency category according to the image information of the patient to be emergency and the complaint symptom information;

and determining an emergency plan according to the emergency category, and outputting the emergency plan through the video call.

2. The method according to claim 1, wherein the method further comprises:

acquiring position information of a patient to be first-aid under the condition that the video call is received;

after the emergency category is determined, matching corresponding ambulance identifications from a preset emergency plan library according to the emergency category;

based on the rescue vehicle identification, the position information is sent to a corresponding rescue vehicle so that the rescue vehicle can conduct route planning according to the position information;

and acquiring the position information of the rescue vehicle, and sending the position information of the rescue vehicle to a terminal identifier.

3. The method of claim 1, wherein the determining an emergency category from the patient image information to be emergency and the complaint symptom information comprises:

Inputting the image information of the patient to be first-aid and the complaint symptom information into a pre-training multi-mode identification model, and outputting a first-aid category; the pre-training multi-mode identification model is obtained by training historical complaint symptom information and historical patient image information to be first-aid in a first-aid center schedule.

4. The method according to claim 1, wherein the method further comprises:

synchronizing the voice information of the emergency site and the image information of the patient to be emergency to the terminal equipment of the hospital, so that the hospital prepares corresponding medical resources according to the received image information of the patient to be emergency and the voice information.

5. The method of claim 3, wherein prior to the inputting the patient image information to be rescued and the complaint symptom information into a pre-trained multimodal recognition model, the method further comprises:

acquiring a plurality of preset first emergency categories, and acquiring historical image information and historical complaint symptom information of a patient to be emergency corresponding to the plurality of preset first emergency categories respectively;

respectively inputting the historical image information and the historical complaint symptom information of the patient to be first-aid corresponding to the preset first-aid categories into a multi-mode identification model, and outputting a plurality of second first-aid categories corresponding to the preset first-aid categories;

And training the multi-modal identification model by utilizing the plurality of preset first emergency categories and the plurality of second emergency categories to obtain a pre-trained multi-modal identification model.

6. The method of claim 2, wherein the obtaining location information of the patient to be emergency, comprises:

under the condition of initiating a video call to the terminal identifier, if the permission of allowing the position access is detected to be preset by the terminal, acquiring the position information by using a Global Positioning System (GPS);

if the terminal is detected not to set the permission for allowing the position access, initiating a request for acquiring the position permission to the terminal identifier; and acquiring the location information using the global positioning system GPS in case a response is received that agrees with the request.

7. An emergency dispatch system, the emergency dispatch system comprising:

the switching-on unit is used for switching on the first-aid call initiated by the terminal identifier and initiating a video call to the terminal identifier;

the acquisition unit is used for acquiring voice information of an emergency scene and image information of a patient to be emergency under the condition that the video call is received; extracting complaint symptom information from the voice information;

The determining unit is used for determining the emergency category according to the image information of the patient to be emergency and the complaint symptom information; and determining an emergency plan according to the emergency category, and outputting the emergency plan through the video call.

8. The emergency dispatch system of claim 7, wherein the first processor is configured to,

the acquisition unit is further used for acquiring the position information of the patient to be first-aid under the condition that the video call is received;

the matching unit is used for matching corresponding ambulance identifications from a preset ambulance protocol library according to the first-aid category after the first-aid category is determined;

the sending unit is used for sending the position information to the corresponding rescue vehicle based on the rescue vehicle identification so as to enable the rescue vehicle to conduct route planning according to the position information; and acquiring the position information of the rescue vehicle, and sending the position information of the rescue vehicle to a terminal identifier.

9. A terminal device, characterized in that the terminal device comprises: a processor, a memory, and a communication bus; the processor, when executing a memory-stored operating program, implements the method of any one of claims 1-6.

10. A storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1-6.