KR102329416B1 - First aid guideline providing method and system based on Artificial Intelligence - Google Patents

Abstract

The present invention provides an artificial intelligence-based first aid guide system and method that, when a patient's disease name is entered in real time at the emergency site, can be derived through artificial intelligence and provided to paramedics promptly. is initiated.
The disclosed artificial intelligence-based first aid guide system includes: a storage unit for storing first aid action guidelines corresponding to disease names and symptoms for each disease as data; an input unit for receiving the disease name, symptoms for each disease, and patient-related data; Learning and storing the disease name and first aid guidelines corresponding to the symptoms of each disease in a deep learning method, and executing deep learning learning using the input data as an input variable based on the learned data, corresponding to the disease name Artificial intelligence department that provides first aid guidance; And it characterized in that it comprises an output unit for outputting the first-aid action guideline by voice or screen. According to the present invention, it is possible to increase the survival rate of critically ill patients by securing the patient's golden time by promptly providing the standard guidelines for optimal emergency activities to paramedics in real time at emergency sites for four major diseases.

Description

First aid guideline providing method and system based on Artificial Intelligence}

The present invention relates to a system and method for providing standard guidelines for first aid activities for intelligent 119 paramedics based on artificial intelligence. When the paramedics at the emergency site identify the patient's symptoms and input the disease name corresponding to it, based on the learning data, the disease name and the voice of the emergency scene are used as input variables to learn in a deep learning method, Artificial intelligence-based first aid guideline that provides paramedics with the best first aid guideline corresponding to the name of the disease to secure the patient's golden time and provide appropriate first aid treatment to the patient in a short time to increase the survival rate of seriously ill patients It relates to a delivery system and method.

In general, the standard guidelines for first aid activities are the first aid activities of paramedics who provide emergency services for emergency medical services such as the organization of an ambulance, emergency equipment operation guidelines, first aid methods for patients on-site or during transport, and guidelines for selecting a hospital to be transferred. standard guidelines for

In other words, the standard guidelines for first aid activities provide an outline explaining the basic matters of the 119 emergency service standard guidelines, and provide the operating guidelines for the first aid team that explain general matters about the operation of the first aid team. In addition, the standard guidelines for first aid activities provide on-site first aid guidelines that describe standardized first aid procedures and guidelines in general field situations, and provide guidelines for special situations that explain standard guidelines in special situations. In addition, the standard guidelines for first aid activities provide technical guidelines that explain the techniques necessary for patient evaluation and first aid treatment, and provide drug guidelines that show basic information about drugs used in the pre-hospital stage.

As fire officials across the country have to listen to lectures based on the standard guidelines for first aid six times, this standard guide for first aid is essential for paramedics to provide optimal first aid to patients at the scene.

Conventional first aid standard guidelines provide 500 pages of information for various diseases. Although it provides standard guidelines, technical guidelines, drug guidelines, and other guidelines for emergency action guidelines for each disease through a vast amount, there is a disadvantage in that it cannot provide appropriate guidelines for each patient's condition or emergency site situation in real time.

Although the standard guidelines for emergency activities provide standard guidelines, technical guidelines, drug guidelines, and other guidelines for diseases, it is almost impossible to fully understand the 500-page standard guidelines for first aid for use in actual emergency situations. In addition, due to the vast amount of standard guidelines for emergency activities, how well the paramedics are familiar with the standard guidelines for emergency activities varies greatly depending on the experience of the paramedics.

Since there are numerous variables in the actual emergency scene, it is very difficult to cope with the actual emergency scene situation only with the first aid measures listed in the standard guidelines for emergency activities. For example, in an actual emergency situation, first aid for patients cannot be generalized because symptoms are different for each patient and coping methods are different. The disadvantage is that it does not reflect

In the conventional first aid protocol, paramedics have no choice but to refer to the standard guidelines for first aid activities in order to provide appropriate first aid to the patient.

In addition, paramedics should administer appropriate first aid to the patient in real time at the emergency scene.

However, this waste of time can be fatal for critically ill patients who urgently need to secure a golden hour.

Therefore, it is essential to have an artificial intelligence-based intelligent system that provides the best first aid guide for the patient in real time and promptly to the paramedics who are dispatched to the emergency scene.

Korean Patent Publication No. 10-1875858 (Registration Date: July 02, 2018)

An object of the present invention to solve the above problems is to learn the name of the patient's disease and the first aid action guidelines corresponding to the detailed symptoms for each disease in the deep learning method and store it as data, and the paramedics at the emergency site to identify the symptoms of the patient When the disease name corresponding to the disease is entered, learning is performed in the deep learning method using the disease name and the voice of the emergency scene as input variables based on the learning data, and the optimal emergency guideline corresponding to the disease name is quickly provided to the paramedics. The goal is to provide a system and method for providing guidelines for first aid based on artificial intelligence to secure the patient's golden time and provide appropriate first aid to the patient in a short time to increase the survival rate of seriously ill patients.

In accordance with an aspect of the present invention for achieving the above object, there is provided an AI-based first aid guide providing system comprising: a storage unit for storing first aid action guidelines corresponding to disease names and symptoms for each disease as data; an input unit for receiving the disease name, symptoms for each disease, and patient-related data; Learning and storing the disease name and first aid guidelines corresponding to the symptoms of each disease in a deep learning method, and executing deep learning learning using the input data as an input variable based on the learned data, corresponding to the disease name Artificial intelligence department that provides first aid guidance; And it may include an output unit for outputting the first-aid guide in the form of voice or screen.

The input unit may include: a data input unit for receiving patient-related texts at the emergency site; a sound input unit receiving sound generated at the emergency site; and a device interlocking unit for receiving measurement values in connection with the devices for measuring the symptoms for each disease by wire or wirelessly.

The artificial intelligence unit, in addition to the name of the patient's disease input from the paramedic, separates the voice of the conversation from the sound received through the sound input unit, and analyzes the separated voices of the conversation to recognize the symptoms of each disease, or through the equipment linkage unit It is possible to recognize the symptoms of each disease of the patient from the inputted biosignal of the patient, select the optimal first aid guideline according to the recognized symptom for each disease, and provide the selected optimal first aid action guideline.

The artificial intelligence unit separates the voice of the conversation between the paramedic and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian from the sound received through the sound input unit, and the separated voices of the conversation It can be analyzed to recognize detailed symptoms for each disease.

The artificial intelligence unit may provide a first aid action guide corresponding to the recognized detailed symptoms for each disease and the name of the disease, and provide a guide for first aid treatment according to the first aid action guideline.

The storage unit includes a question set to be asked by the paramedics to the subject and an answer set to answer the paramedics, and the artificial intelligence unit may include a set of questions from the sound input through the sound input unit. Based on the set and the answer set, it is possible to recognize detailed symptoms for each disease by analyzing the voices of the conversation contents of the speakers.

The artificial intelligence unit recognizes the detailed symptoms for each disease and checks the detailed situation table for each disease, or records the patient's biosignals obtained in real time through interworking with the equipment in real time in the emergency activity log, and is identified in real time According to the condition of the patient, the detailed status table for each disease or the emergency activity log may be updated in real time.

The artificial intelligence unit may record the patient's bio-signals obtained in real time through interworking with other equipment through the equipment interlocking unit in real time in the emergency activity log.

On the other hand, the artificial intelligence-based first aid guidance providing method according to another aspect of the present invention for achieving the above object is, (a) the artificial intelligence unit learns the first aid guidance corresponding to the disease name and symptoms by disease in a deep learning method. step; (b) storing the learned data by the artificial intelligence unit; (c) receiving, by an input unit, at least one of the disease name, the disease-specific symptoms, and patient-related data; (d) the artificial intelligence unit executing deep learning learning using the input data as an input variable based on the learned data to derive first aid guideline corresponding to the disease name or symptom for each disease; And (e) the artificial intelligence unit may include the step of providing the derived first aid guidance.

In the step (c), the input unit receives patient condition-related data from the paramedics at the emergency site, receives the sound generated at the emergency site through the sound input unit, or measures the symptoms of each disease through the equipment linker The measured values can be inputted by linking with each other by wire or wirelessly.

In step (d), the artificial intelligence unit separates the voice of the conversation contents from the received sound, analyzes the separated voices of the conversation to recognize the symptoms of each disease, or analyzes the patient's bio-signals input through the equipment linkage unit Thus, the symptoms of each disease can be recognized.

In the step (d), the artificial intelligence unit separates, from the sound input through the sound input unit, the voice of the conversation between the paramedics and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian. It is possible to recognize detailed symptoms for each disease by analyzing the voices of the conversation.

In step (d), the artificial intelligence unit recognizes detailed symptoms for each disease and checks the detailed situation table for each disease, or records the patient's biosignals obtained in real time through interworking with equipment in real time in the emergency activity log, According to the condition of the patient identified in real time, the detailed status table for each disease or the emergency activity log can be updated in real time.

According to the present invention, it is possible to guide the provision of first aid guidelines regarding the best first aid to the paramedics in real time so that the paramedics can provide the best first aid to the patient at the emergency scene of the 4 major diseases.

In other words, when the paramedics at the ambulance site identify the patient's condition and input the name of the patient's disease in real time, promptly provide the paramedics with guidelines for emergency activities in response to it, thereby securing the patient's golden time and providing the appropriate treatment for the patient within a short time. Emergency treatment can increase the survival rate of seriously ill patients.

In the first aid protocol according to the present invention, paramedics can be automatically provided with optimal first aid standard guidelines in real time through an artificial intelligence-based intelligent first aid standard guide system through a tablet at the scene.

Therefore, since paramedics can quickly refer to the optimal emergency guidelines without delay, the time required for emergency activities is significantly reduced compared to the prior art, and since there is no unnecessary time required, the golden time can be secured. It can increase the survival rate.

In addition, in the conventional first aid activity, there was a high possibility of causing a mistake in the process of manually referring to the standard guidelines for emergency activity, but if the AI-based knowledge DB proposed in the present invention is used, the possibility of a mistake can also be significantly reduced.

In addition, since the general first aid standard guidelines provide first aid for various diseases in 500 pages, paramedics have no choice but to feel the burden of familiarizing themselves with the 500 pages of standard first aid guidelines, but the knowledge DB of the present invention By constructing this system, paramedics' reliance on standard guidelines for first aid activities can be reduced, thereby reducing the burden and reducing the number of mistakes made by paramedics at the scene.

In addition, although first aid is provided in a vast amount in the standard guidelines for first aid activities, only generalized first aid is provided. Because it varies each time, it is difficult to take appropriate measures for the patient only with generalized first aid. You can take it, and your chances of surviving will be much higher.

1 is a diagram schematically showing an overall configuration to which an artificial intelligence-based first aid guide providing system according to an embodiment of the present invention is applied.
2A and 2B are diagrams schematically illustrating the internal configuration of an AI-based first aid guide providing system according to an embodiment of the present invention.
3 is a block diagram illustrating functional blocks of an artificial intelligence unit according to an embodiment of the present invention.
4 is a diagram illustrating, for example, a method of self-organizing a DNA mission in a self-organizing module using a neuroblock chain combination according to an embodiment of the present invention.
5 is a diagram illustrating an example of a method of self-constructing a DNA model in a self-organizing module using a neuroblock chain combination according to an embodiment of the present invention.
6 is a diagram illustrating a configuration further including a DNA tool in an artificial intelligence unit using a deep learning-based self-adaptive learning technique according to an embodiment of the present invention.
7 is a diagram illustrating an operation flowchart for explaining a method for providing an AI-based first aid guideline according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art to which the present invention pertains can easily implement them. The present invention may be embodied in various different forms and is not limited to the embodiments described herein.

In order to clearly explain the present invention, parts irrelevant to the description are omitted, and the same reference numerals are given to the same or similar elements throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, it includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, this means that other components may be further included rather than excluding other components unless otherwise stated.

When a part is referred to as being “above” another part, it may be directly on top of the other part, or the other part may be involved in between. In contrast, when a part refers to being "directly above" another part, no other part is involved in between.

Terms such as first, second and third are used to describe, but are not limited to, various parts, components, regions, layers and/or sections. These terms are used only to distinguish one part, component, region, layer or section from another part, component, region, layer or section. Accordingly, a first part, component, region, layer or section described below may be referred to as a second part, component, region, layer or section without departing from the scope of the present invention.

The terminology used herein is for the purpose of referring to specific embodiments only, and is not intended to limit the invention. As used herein, the singular forms also include the plural forms unless the phrases clearly indicate the opposite. The meaning of "comprising," as used herein, specifies a particular characteristic, region, integer, step, operation, element and/or component, and includes the presence or absence of another characteristic, region, integer, step, operation, element and/or component. It does not exclude additions.

Terms indicating a relative space such as “below” and “above” may be used to more easily describe the relationship of one part shown in the drawings to another part. These terms, along with their intended meanings in the drawings, are intended to include other meanings or operations of the device in use. For example, if the device in the figures is turned over, some parts described as being "below" other parts are described as being "above" other parts. Thus, the exemplary term “down” includes both the up and down directions. The device may be rotated 90 degrees or at other angles, and terms denoting relative space are interpreted accordingly.

Although not defined otherwise, all terms including technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs. Commonly used terms defined in the dictionary are additionally interpreted as having a meaning consistent with the related technical literature and the presently disclosed content, and unless defined, they are not interpreted in an ideal or very formal meaning.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily carry out the embodiments of the present invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein.

1 is a diagram schematically showing an overall configuration to which an artificial intelligence-based first aid guide providing system according to an embodiment of the present invention is applied.

Referring to FIG. 1 , an artificial intelligence-based first aid guide providing system 100 according to an embodiment of the present invention may be applied and implemented to a tablet 100 carried by paramedics. Therefore, in the embodiment of the present invention, the tablet 100 refers to the artificial intelligence-based first aid guide system 100, and the same reference numerals are used.

In addition, the AI-based first aid guidance providing system 100 according to an embodiment of the present invention is applied to the tablet 100 as an example, but is not limited thereto, and a smart phone carried by paramedics in addition to the tablet 100 . It can also be applied to wireless terminals including (Smart Phone).

As shown in FIG. 1 , when an ambulance crew arrives at an ambulance scene together with an ambulance, the ambulance crew drives a portable tablet 100 , and an artificial intelligence-based first aid activity guideline providing system 100 according to an embodiment of the present invention ) works.

In this case, the measurement equipment 200 at the emergency site may transmit the measurement values measured for the emergency patient to the cloud server and store it in the cloud server.

Accordingly, the tablet 100 at the emergency site may receive the measurement values of the corresponding emergency patient from the cloud server and provide video or audio so that the ambulance crew can check it.

In addition, the tablet 100 may connect with the measurement equipment 200 for measuring the symptoms of an emergency patient at the emergency site by wire or wirelessly, and receive measurement values related to the condition of the emergency patient from the measurement equipment 200 . .

In addition, the tablet 100 is an AI-based first aid guideline providing system, and may use a first aid action guideline protocol that provides first aid action guideline for each of the four serious diseases.

In order to input the disease name according to the patient's symptoms into the tablet 100, the paramedics check the patient's consciousness and breathing, and check the trauma and the pulse. After confirmation, the paramedics input the disease name determined as one of the four major diseases according to the patient's consciousness, respiration, trauma and pulse into the tablet 100 .

Here, the artificial intelligence of the tablet 100 utilizes voice, bio-signals, images, and other information (severity classification, event screening for each disease) in addition to the disease name input from the paramedic.

Artificial intelligence-based voice technology is used to assist paramedics by recording the patient treatment situation and patient condition of paramedics in real time, and checking events for each situation according to the detailed situation table. In the case of voice, first, by removing numerous noises generated at the actual emergency scene, the speaker (a conversation between a paramedic and a patient, a conversation between an ambulance crew and a witness, or a conversation between a paramedic and a caregiver) is separated. In addition, it effectively removes noise based on artificial intelligence in noisy ambulance situations, and performs accurate voice analysis through speaker-to-speaker voice separation. Second, in the conversation content in which the noise is removed and the speaker is separated, the DB format is based on the Question Set that the paramedics should ask and the Answer Set that can be answered by the paramedics at the actual emergency scene. Based on the QA Set constructed with

In the case of biosignals, there are biosignals obtained in real time through interworking with equipment for measuring biosignals, and biosignals obtained through images displayed on equipment because interlocking is impossible. The biosignals obtained in real time through interworking with the equipment are recorded in the emergency activity log in real time and updated according to the patient's condition in real time. In the case of bio-signals that cannot be obtained through interworking with equipment, artificial intelligence-based image recognition technology is used to recognize the equipment in which bio-signals appear. The patient's bio-signals that appear on the monitor of the equipment through image recognition technology are automatically entered into the detailed situation table for each disease.

The AI-based video technology recognizes the actions of paramedics during emergency activities and is used as a first aid procedure or as an aid to treatment. Therefore, when paramedics recognize first aid and provide appropriate first aid, it is necessary to provide the next first aid sequence, so it assists in providing the next first aid sequence by recognizing the first aid activities of the first aid crew. For example, when performing CPR, the degree or frequency of pressure applied to the patient may be guided through an image.

Since other AI-based technologies, including severity classification and disease-specific event screening, have different major events for each disease depending on each patient's condition, each major event is configured separately for each patient's condition. Measures which event will occur accordingly.

2A and 2B are diagrams schematically showing the main configuration of an AI-based first aid guide providing system according to an embodiment of the present invention.

Referring to FIGS. 2A and 2B , the AI-based first aid guide providing system 100 according to an embodiment of the present invention includes a data storage unit 110 , an input unit 120 , an artificial intelligence unit 130 , and an output unit. (140).

The data storage unit 110 stores the name of the disease and the first aid guide corresponding to the symptom for each disease as data.

In addition, the AI-based first aid guidance providing system 100 according to an embodiment of the present invention, in addition to using the data storage unit 110 internally, as shown in FIG. 2B, a database (DB) installed in a remote place (112) can be used. That is, the database 112 stores the name of the disease and the first aid guide corresponding to the symptom for each disease as data. In this case, the database 110 is a separate device that operates in conjunction with the tablet 100 at a remote location.

In this case, the tablet 100 may wirelessly retrieve data from the database 112 , and may also wirelessly transmit data processed by the tablet 100 to the database 112 and store it in the database 112 . . For example, when a specific first aid guide corresponding to the disease name of the patient is provided to a patient at the emergency site, the tablet 100 transmits and stores the currently provided first aid treatment to the database 112 . can This can be used to inform that first aid in one stage of the sequence of first aid protocols for a specific disease stored in the structured database 112 has been provided, so that the next stage of first aid should be provided. .

The input unit 120 receives disease name or symptom information as well as patient information from paramedics at the emergency site.

To this end, the input unit 120 includes a data input unit for receiving patient-related texts at the emergency site; a sound input unit receiving sound generated at the emergency site; an image input unit for acquiring image data by photographing the emergency scene; And it may include a device linkage unit for receiving measurement values from the measurement devices in connection with the devices for measuring the symptoms of each disease in a wired or wireless connection.

The artificial intelligence unit 130 learns the name of the disease and the emergency action instructions corresponding to the detailed symptoms for each disease in a deep learning method and stores it in the data storage unit 110 or the database 112, and the input unit 120 ), executes deep learning learning as an input variable to derive the optimal first aid guideline corresponding to the disease name and detailed symptoms for each disease, and provides the derived best first aid action guideline.

The output unit 140 may output the emergency action instructions by voice or screen.

To this end, the output unit 140 may include a speaker for outputting first aid instructions by voice, and a display unit for outputting the first aid instructions in the form of text on the screen.

In addition, the artificial intelligence unit 130 separates the voice of the conversation content from the received sound, and analyzes the separated voice of the conversation to recognize the symptoms of each disease and use it to derive the optimal first aid guide, or input it through the equipment linkage unit. Recognizes the symptoms of each disease according to the patient's disease name from the patient's biosignals, uses it to derive the optimal first aid guideline corresponding to the recognized patient's disease-specific symptom, and provides the derived optimal first aid action guideline. can

In addition, the artificial intelligence unit 130 separates the voice of the conversation between the paramedic and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian from the sound received through the sound input unit, and separated It can be used to derive the optimal first aid guideline by recognizing the detailed symptoms of each disease by analyzing the voices of the conversation.

Accordingly, the artificial intelligence unit 130 receives one of severe trauma disease, cerebrovascular disease, cardiovascular disease, and cardiac arrest disease as a severe disease name through the input unit 120 , and the received severe disease name and detailed symptoms for each disease as described above. Using as an input variable, it is possible to derive first aid action guidelines corresponding to severe disease names in a short time based on the learning data.

In addition, the data storage unit 110 and the database 112 may store a set of questions asked by the paramedics to the subject and a set of answers to be answered by the paramedics. Accordingly, the artificial intelligence unit 130 may recognize the detailed symptoms of each disease by analyzing the voices of the conversation contents of the speakers based on the question set and the answer set from the sound received through the sound input unit.

In addition, the artificial intelligence unit 130 receives, based on the learning data, one of severe trauma disease, cerebrovascular disease, cardiovascular disease, and cardiac arrest disease as a serious disease name, or patient-related text, conversational voice, image data, patient By executing deep learning learning using at least one or more of the biosignals of

In addition, the artificial intelligence unit 130 recognizes detailed symptoms for each disease and checks the detailed situation table for each disease, or records the patient's biosignals obtained in real time through interworking with equipment in real time in the emergency activity log, and in real time According to the condition of the patient identified by

In addition, the artificial intelligence unit 130 records the patient's biosignals obtained in real time through interworking with other equipment through the equipment linkage unit in real time in the emergency activity log, and biosignals that cannot be obtained through interlocking with the equipment. In the case of , it is possible to automatically record the patient's bio-signals input through the image input unit in the detailed situation table for each disease.

3 is a block diagram illustrating functional blocks of an artificial intelligence unit according to an embodiment of the present invention.

Referring to FIG. 3 , the artificial intelligence unit 130 according to an embodiment of the present invention may include an element derivator 10 , a learning engine 20 , and an action unit 30 .

That is, in the present invention, the element is derived by processing structured data and unstructured data in the element derivator 10 , and self-adaptive learning can be performed using the element in the learning engine 20, and the learning result is used. Including the action unit 30, it is possible to provide a system capable of understanding and scheduling, decision making and prediction, recommendation and situational action in a modular way, and a system suitable for various situations can be provided in a custom manner. .

For example, the element derivator 10 receives disease names, detailed symptoms for each disease, and patient status related data, and derives main elements to be used for learning, and the learning engine 20 uses the main elements to provide general field Self-adaptation learning is carried out by classifying the guidelines for use among on-site first aid guidelines, special situation-specific guidelines, technical guidelines, and drug guidelines, and in the action stage (30), the patient's (severe) disease name is It is to recommend and provide guidelines for the best first aid action.

The element derivator 10 may process input data to derive Elements. That is, the element derivator 10 may derive an element that is input information of the learning engine 20 from input data including structured data and unstructured data.

The element derivator 10 may include a text conversion module 11 , an information extraction module 12 , and an element derivation module 13 .

The text conversion module 11 may convert unstructured data excluding text among input data into text data (Text Conversion). In particular, the text conversion module 11 may convert unstructured data including images, images, and voices other than text into text data.

The information extraction module 12 may extract information from the text data converted by the text conversion module 11 (Information Extraction). In addition, the information extraction module 12 may extract necessary information from input data in text form, which is not a target of conversion in the text conversion module 11 .

The element derivation module 13 may derive an element to be input to the learning engine 20 from the extracted information (Element Identification & Elicitation).

The learning engine 20 self-organizes a DNA mission using the elements derived from the element derivator 10, and a deep learning-based artificial neural network DNA model using the self-organized DNA mission. ) can be self-constructed, and self-constructed DNA models can be trained. The present invention combines self-adaptation technology and deep learning-based learning technology, and includes a learning engine 20 that self-organizes DNA missions and self-configures DNA models, so that the It can effectively implement the human brain mechanism to solve situations by creating 1 , the learning engine 20 may be configured to include an organization module 21 , a configuration module 22 , and a learning module 23 .

The organization module 21 may self-organize a DNA mission for each emergency activity using the elements derived from the element derivator 10 (Self-Organization of DNA Mission). More specifically, the organization module 21, based on the input main factors, i) the outline of the standard guidelines for first aid activities, ii) the operation guidelines of the ambulance team, iii) the first aid guidelines in general field situations, iv) the guidelines for special situations , v) technical guidelines, vi) drug guidelines, etc., can be compared and evaluated to self-organize and create a first-aid DNA mission. Here, the mission is a mission of a predefined organization, and the DNA mission is a mission self-organized by the organization module 21 of the present invention and is different from each other.

On the other hand, the first aid DNA mission organized by the organization module 21 may be a combination of blocks of the first aid guide organization and chains. That is, the organization module 21 can organize the DNA mission by combining the block and chain of the organization using the Neuro Block Chain Combination technology. In addition, according to the embodiment, the DNA mission may include a special DNA mission consisting of a combination of chains.

In addition, the DNA mission may be composed of the sum of mission modules, and the mission module may be a function of the element received from the element derivator 10 and the position of the first aid action guide organization (Positions of Organization). In this case, the position of the organization member may be predetermined.

In the present invention, in order for the artificial intelligence unit 130 to support human tasks (missions), the patient's disease name (severe traumatic disease, severe traumatic disease, cerebrovascular disease, cardiovascular disease, cardiac arrest disease, etc.), by disease The first aid action that performs the mission for the symptom, and the position (instruction) of the first aid action guide to which the first aid action guide belongs or is located may be considered. Such an organization has, for example, a hierarchical tree structure or a parallel structure, and in the hierarchical tree structure, nodes and nodes are connected (that is, positions and positions within a disease name (chain)) If this is extended to the entire organization, it can be viewed as a chain, that is, a connection between groups (instructions within the organization) composed of a certain tree structure (blocks).

When using the neuro blockchain combination technology in the organization module 21, a chain is a link between a position (guideline 1) and a position (guideline 2) within first aid, and groups with certain first aid action guidelines (in general field situations). On-site first aid guide group, special situation guide group, technique guide group, drug guide group), and the block of first aid guide organization is a group in which positions and positions within a certain first aid action are connected and grouped together. A first aid action guide may consist of one block, or one first aid action may consist of several guide blocks.

4 is a diagram illustrating, for example, a method of self-organizing a DNA mission in the organization module 21 using a neuroblock chain combination according to an embodiment of the present invention.

The organization module 21 of the deep learning-based learning engine 20 according to an embodiment of the present invention includes a block (Block i, Block j, Block k, etc.) of an organization as shown in FIG. 4 and a chain (Chain l) , Chain m, Chain n, etc.) can be combined to form a DNA mission. The constructed DNA mission can be expressed as the sum of the elements and mission modules that are a function of the first aid position of the disease name.

In addition, the DNA mission may include a special DNA mission consisting of a combination of chains. That is, depending on the embodiment, a DNA mission may be configured by a combination of only chains without an organizational block.

That is, in FIG. 4 , each block may be one of an on-site first aid instruction block in a general field situation, an instruction block for each special situation, a technique instruction block, and a drug instruction block.

The configuration module 22 may self-compose a deep learning-based artificial neural network DNA model using a self-organized DNA mission (Self-Composition of DNA Model). That is, the configuration module 22 may receive the DNA mission from the organization module 21 , and configure and create an artificial neural network DNA model that can be learned based on deep learning by itself. The DNA model self-constructed by the configuration module 22 may be a model that flexibly changes according to input data because it is constructed using a DNA mission self-organized by elements input over time.

The DNA model may be a combination of Blocks of Function and Chains. That is, the configuration module 22 can organize a DNA model by combining functional blocks and chains using Neuro Block Chain Combination technology.

In addition, the configuration module 22 may self-configure a DNA model composed of a sum of functional submodels.

Here, the functional block is a function-specific set of situations that enable learning in the artificial neural network model by mimicking the situation judgment method of the human brain. can be configured. It can be assumed that a simple situation can be judged only by a single thought, but a complex situation can be judged by multiple thoughts instead of one thought. In the process of constructing a model to solve a mission, using this concept, a DNA model can be self-organized by combining functional blocks and chains in such a way that complex situations are classified by function and grouped for judgment.

As described above, if the name of severe disease, that is, severe traumatic disease, cerebrovascular disease, cardiovascular disease, cardiac arrest, etc., has different symptoms and different first aid action guidelines, the organization blocks are 4, For the performance of the four organizational blocks, it can be converted into four functional blocks by taking sub-tasks for each function.

5 is a diagram illustrating, for example, a method of self-configuring a DNA model in the configuration module 22 using a neuroblock chain combination according to an embodiment of the present invention.

The configuration module 22 of the deep learning-based learning engine 20 according to an embodiment of the present invention, as shown in FIG. 5, functions blocks and chains through neuroblock chain combination technology. It is possible to construct a functional submodel (Functional Submodel i, Functional Submodel j, Functional Submodel k, Functional Submodel m, Functional Submodel n, etc.)

For example, when cardiovascular disease is input as the patient's disease name, in addition to the field-measured measurement data and patient condition-related data, the tissue module 21 performs a general field situation in the tissue module 21 based on the input data. Using self-organized on-site first aid instruction block, special situational instruction block, skill instruction block, drug instruction block, etc. select at least one or more of which guidelines to use; select at least one or more of which guidelines to use in special contextual guidelines; select at least one or more of which guidelines to use in technical guidelines; It is to select at least one or more to use and connect them with a chain to form a functional block.

Meanwhile, in FIG. 3 , the learning module 23 may self-lear a self-constructed DNA model (Self-Learning of DNA Model). That is, the learning module 23 is a configuration for learning the DNA model configured in the configuration module 22 , and can learn through artificial neural network technology, and can deliver the learning result to the action unit 30 .

The action unit 30 may perform a function according to the learning result of the learning engine 20 through the understanding and scheduling module 31 , the determination and prediction module 32 , and the recommendation and action module 33 . As shown in FIG. 3 , the action unit 30 may be configured to include an understanding and scheduling module 31 , a judgment and prediction module 32 , and a recommendation and action module 33 , and each module includes common software. can be In the present invention, the structural aspect of the action device 30 will be described.

The understanding and scheduling module 31 may understand a given situation or grasp an intention, and may provide scheduling to a decision maker using a result of understanding the situation or understanding the intention (Understanding & Scheduling).

The decision and prediction module 32 may provide judgment and analysis results for a given situation, and predict and provide a possible situation (Decision & Prediction).

The recommendation and action module 33 may use the analysis result and the prediction result to recommend a decision for a given situation and provide an action accordingly (Recommendation & Action). To this end, the recommendation and action module 33 may receive analysis results and prediction results from the determination and prediction module 32 .

6 is a diagram illustrating a configuration further including a DNA tool 40 in the artificial intelligence unit using a deep learning-based self-adaptive learning technique according to an embodiment of the present invention.

As shown in FIG. 6 , the artificial intelligence unit 130 using deep learning-based self-adaptive learning technology according to an embodiment of the present invention may further include a DNA tool 40 .

The DNA tool 40 may provide a plurality of tools to the element derivator 10 , the learning engine 20 , and the action unit 30 . That is, the DNA tool 40 may provide a tool that helps the element derivator 10 , the learning engine 20 , and the action group 30 perform their respective functions.

More specifically, the DNA tool 40 includes a Conversion Tool 41 for converting unstructured data into text, an Extraction Tool 42 for extracting information, and a self-contained DNA mission. A combination tool 43 that connects blocks and chains necessary for self-organization of tissue and DNA models may be included.

In addition, the organization module 21 , the configuration module 22 , and the learning module 23 are self-organizing DNA missions, self-organizing and self-learning DNA models. Self-Adapted Tool (44) may be configured to further include.

7 is a diagram illustrating an operation flowchart for explaining a method for providing an AI-based first aid guideline according to an embodiment of the present invention.

Referring to FIG. 7 , in the AI-based first aid guide system 100 according to an embodiment of the present invention, first, the artificial intelligence unit 130 responds to the disease name and disease-specific symptoms stored in the data storage unit 110 . It learns the first aid action instructions in a deep learning method (S710).

That is, the artificial intelligence unit 130 dips the data on the name of the disease stored in the data storage unit 110 or the database 112 and the first aid guideline corresponding to the symptoms for each disease as shown in FIGS. 4 and 5 . learning by learning. For example, in the artificial intelligence unit 130, the patient is unconscious, before the spinal cord injury stage, there is paralysis and sensory loss of motor nerves after the neurological examination, there is other damage that does not recognize the spinal cord injury, and the spinal cord examination result is abnormal. In the case of , learning to output a message indicating that the patient needs spinal fixation as an emergency action guideline corresponding thereto.

Next, the artificial intelligence unit 130 stores the learned data in the data storage unit 110 or the database 112 (S720).

This is to use the learned data when the artificial intelligence unit 130 receives the disease name or detailed symptoms for each disease from the paramedics at the emergency site, and derives the optimal first aid guide for the disease. .

Next, the input unit 120 receives at least one of a disease name, a symptom for each disease, and data related to a patient's condition at the emergency site ( S730 ).

At this time, the input unit 120 receives data related to the patient's condition from the paramedics at the emergency site, receives the sound generated at the emergency site through the sound input unit, or wired with equipment for measuring symptoms for each disease through the equipment linkage. Alternatively, the measured values may be input by wirelessly interworking.

Next, the artificial intelligence unit 130 executes deep learning learning using the input data as an input variable based on the learned data to derive first aid guidelines corresponding to disease names or symptoms for each disease (S740).

For example, the artificial intelligence unit 130 executes the process as shown in FIGS. 4 and 5, and uses data related to the patient's condition, including consciousness change including loss of consciousness, high-risk damage mechanism, high-speed collision accident, and the patient's stomach. In the event of a passing accident, a fall from 1 m or more or 5 stairs or more, damage to the spine such as diving, bicycle or motorcycle accident, pain or tenderness in the middle of the spine, movement and sensory abnormalities, etc. In case of emergency, guidelines for first aid can be derived.

At this time, the artificial intelligence unit 130 separates the voice of the conversation content from the received sound, analyzes the separated voices of the conversation to recognize the symptoms of each disease, or analyzes the patient's bio-signals input through the equipment linker to analyze the patient's condition can recognize

In addition, the artificial intelligence unit 130 separates the voice of the conversation between the paramedic and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian from the sound input through the sound input unit, and the separated conversation It is possible to recognize detailed symptoms of each disease by analyzing the contents and voices.

Next, the artificial intelligence unit 130 provides the derived first aid guide in the form of voice or screen (S750).

For example, the artificial intelligence unit 130 as the name of the patient's disease, when a cardiovascular disease is input, as an emergency guideline, auscultation of lung sounds, calming the patient, taking a comfortable posture, listening to a medical history, , it can provide guidelines to confirm the history of taking medications for erectile dysfunction.

For example, the artificial intelligence unit 130 manipulates the degree of opening the airway as a first aid guide when receiving a head injury without a vomiting reflex as data related to the patient's condition, and blood to the nose or ear according to the signs of skull fracture It is said to wipe without blocking the flow, and for all head injury patients, it is possible to provide instructions to fix the cervical vertebrae, and to stop hemostasis by compression with sterile sterile gauze in case of external bleeding.

For example, the artificial intelligence unit 130 applies a top sealing dressing for an open pneumothorax when a penetrating image of an abdominal and pelvic injury is input as data related to the patient's condition, and if there is life-threatening external bleeding, it is compressed with sterile gauze They can provide instructions to stop bleeding.

For example, the artificial intelligence unit 130 provides instructions regarding hemostatic treatment, such as pack / dressing, compression dressing, hemostatic dressing, and tourniquet application, when a severe bleeding state is input due to a limb injury as data related to the patient's condition, In cases where bleeding is not severe, instructions are given to apply gauze and bandages, fixation to minimize movement, and ice packs for edema regarding treatment and protection of the injured area. Do not try to forcefully subdue the extremity, apply it including the upper and lower parts of the piercing when fixing the splint, and provide instructions to fix the protruding part of the body and the hard part of the splint with a thick pad.

In addition, the detailed symptoms of the severe disease may include at least one or more of multiple trauma, head injury, chest injury, abdominal pelvic injury, limb injury, ophthalmic injury, and dental injury.

Here, the multiple trauma includes a bleeding site, loss of consciousness or head and vertebra damage, an injured site, a fracture site, and a head injury includes an impairment of consciousness, asymmetric pupil, vomiting reflex, skull fracture, and signs of brain prolapse. can

In addition, chest injury includes data on penetrating injury, life-threatening external bleeding, blunt trauma, shaking chest, cervical spine injury, and ECG monitoring. It includes the presence or absence of an impingement and the presence or absence of a fracture of the pelvic bone, and the limb injury may include the presence or absence of bleeding beyond a certain standard, the presence of a fracture, and the presence or absence of amputation.

In addition, the ophthalmic damage may include data regarding the presence or absence of intraocular bleeding, pupil size, trauma, and electrocardiogram monitoring, and the dental damage may include data regarding the presence or absence of an intraoral wound and missing teeth.

As described above, according to the present invention, the name of the patient's disease and first aid guidelines corresponding to the detailed symptoms for each disease are learned in the deep learning method and stored as data, and the paramedics at the emergency site identify the patient's symptoms and respond accordingly. When a disease name is entered, based on the learning data, learning is performed in the deep learning method using the disease name and the voice of the first aid site as input variables, and the optimal emergency guidance corresponding to the disease name is quickly provided to the paramedics. It is possible to realize an artificial intelligence-based first aid guide system and method to secure the golden time and provide appropriate first aid to the patient in a short time to increase the survival rate of seriously ill patients.

Those skilled in the art to which the present invention pertains should understand that the present invention can be embodied in other specific forms without changing the technical spirit or essential characteristics thereof, so the embodiments described above are illustrative in all respects and not restrictive. only do The scope of the present invention is indicated by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

100: artificial intelligence-based first aid guidance system 200: measuring equipment
110: data storage unit 120: input unit
130: artificial intelligence unit 140: output unit
10: Element Derivator 11: Text Transformation Module
12: information extraction module 13: element derivation module
20: learning engine 21: organization module
22: configuration module 23: learning module
30: action group 31: understanding and scheduling module
32: judgment and prediction module 33: recommendation and action module
40: DNA tool 41: transformation tool
42: extraction tool 43: combination tool
44: self-adaptation tool 112: database

Claims (13)

a storage unit for storing disease names and first aid guidelines corresponding to symptoms of each disease as data;
an input unit for receiving the disease name, symptoms for each disease, and patient-related data;
Learning and storing the disease name and first aid guidelines corresponding to the symptoms of each disease in a deep learning method, and executing deep learning learning using the input data as an input variable based on the learned data, corresponding to the disease name Artificial intelligence department that provides first aid guidance; and
It includes an output unit for outputting the first-aid action instructions by voice or screen,
The artificial intelligence unit,
a factor derivator for deriving main factors to be used for deep learning learning from the received disease name, disease-specific symptoms, and patient-related data;
a learning engine that executes self-adaptation learning by classifying the guidelines for use of first aid guidelines in general field situations, guidelines for special situations, technique guidelines, and drug guidelines using the above main elements; and
Including an action device that recommends the best first aid guideline corresponding to the disease name,
The learning engine is
Organization module that self-organizes DNA missions by combining organizational blocks and chains; and
It uses the self-organized DNA mission and includes a configuration module that self-constructs a DNA model by combining functional blocks and chains,
The block of the above organization corresponds to one of a first aid instruction block in a general field situation, an instruction block specific to a special situation, a skill instruction block, and a drug instruction block,
The block of the function selects at least one guideline to be used in the on-site first aid guideline in the general field situation, selects at least one guideline to use in the special situation-specific guideline, and in the technical guideline Artificial intelligence-based first aid guidance providing system, characterized in that it is configured by selecting at least one or more guidelines to use, selecting at least one or more guidelines to use from the drug guidelines, and linking each selected guideline with a chain .
The method of claim 1,
The input unit,
a data input unit for receiving patient-related texts at the emergency scene;
a sound input unit receiving sound generated at the emergency site; and
an equipment linkage unit for receiving measurement values in connection with the equipment for measuring the symptoms of each disease by wire or wirelessly;
Including, artificial intelligence-based first aid guidance system.
3. The method of claim 2,
The artificial intelligence unit separates the voice of the conversation from the received sound, analyzes the separated voices of the conversation to recognize the symptoms of each disease, or recognizes the symptoms of each disease of the patient from the patient's bio-signals input through the equipment linkage unit, , AI-based first aid guideline providing system that selects the optimal first aid guideline according to the recognized symptoms of each patient's disease and provides the selected optimal first aid action guideline.
4. The method of claim 3,
The artificial intelligence unit separates the voice of the conversation between the paramedic and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian from the sound received through the sound input unit, and the separated voices of the conversation An AI-based first aid guide system that analyzes and recognizes detailed symptoms for each disease.
5. The method of claim 4,
The artificial intelligence unit provides first aid guidance corresponding to the recognized detailed symptoms for each disease and the disease name, and guides the first aid procedure according to the first aid guideline.
3. The method of claim 2,
The storage unit includes a set of questions asked by the paramedics to the subject, and an answer set to answer the paramedics,
The artificial intelligence unit, based on the question set and the answer set, from the sound input through the sound input unit, analyzes the voices of the conversation contents of the speakers to recognize detailed symptoms for each disease, an artificial intelligence-based first aid guide providing system.
7. The method of claim 6,
The artificial intelligence unit recognizes the detailed symptoms for each disease and checks the detailed situation table for each disease, or records the patient's biosignals obtained in real time through interworking with the equipment in real time in the emergency activity log, and is identified in real time An artificial intelligence-based first aid guide providing system that updates the detailed situation table for each disease or the emergency activity log in real time according to the patient's condition.
8. The method of claim 7,
The artificial intelligence unit, an artificial intelligence-based first aid guide providing system that records the patient's biosignals obtained in real time through interworking with other equipment through the equipment linkage unit in real time in a first aid activity log.
(a) the artificial intelligence unit learning the name of the disease and the first aid guide corresponding to the symptoms of each disease in a deep learning method;
(b) storing the learned data by the artificial intelligence unit;
(c) receiving, by an input unit, at least one of the disease name, the disease-specific symptoms, and patient-related data;
(d) the artificial intelligence unit executing deep learning learning using the input data as an input variable based on the learned data to derive first aid guideline corresponding to the disease name or symptom for each disease; and
(e) the artificial intelligence unit comprising the step of providing the derived first aid guidance,
The artificial intelligence unit,
a factor derivator for deriving main factors to be used for deep learning learning from the received disease name, disease-specific symptoms, and patient-related data;
a learning engine that executes self-adaptation learning by classifying the guidelines for use of first aid guidelines in general field situations, guidelines for special situations, technique guidelines, and drug guidelines using the above main elements; and
Including an action device that recommends the best first aid guideline corresponding to the disease name,
The learning engine is
Organization module that self-organizes DNA missions by combining organizational blocks and chains; and
It uses the self-organized DNA mission and includes a configuration module that self-constructs a DNA model by combining functional blocks and chains,
The block of the above organization corresponds to one of a first aid instruction block in a general field situation, an instruction block specific to a special situation, a skill instruction block, and a drug instruction block,
The block of the function selects at least one guideline to be used in the on-site first aid guideline in the general field situation, selects at least one guideline to use in the special situation-specific guideline, and in the technical guideline Selecting at least one or more guidelines to use, selecting at least one or more guidelines to use from the drug guidelines, and linking each selected guideline with a chain .
10. The method of claim 9,
In the step (c), the input unit receives patient condition-related data from the paramedics at the emergency site, receives the sound generated at the emergency site through the sound input unit, or measures the symptoms of each disease through the equipment linkage unit A method of providing artificial intelligence-based first aid guidance that receives measurement values by linking them with wired or wireless devices.
11. The method of claim 10,
In step (d), the artificial intelligence unit separates the voice of the conversation contents from the received sound, analyzes the separated voices of the conversation to recognize the symptoms of each disease, or analyzes the patient's bio-signals input through the equipment linkage unit A method of providing artificial intelligence-based first aid guidance that recognizes symptoms by disease.
10. The method of claim 9,
In the step (d), the artificial intelligence unit separates, from the sound input through the sound input unit, the voice of the conversation between the paramedic and the patient, the voice of the conversation between the paramedic and the witness, and the voice of the conversation between the paramedic and the guardian. A method of providing artificial intelligence-based first aid guidance that recognizes detailed symptoms for each disease by analyzing voices from conversations.
10. The method of claim 9,
In step (d), the artificial intelligence unit recognizes detailed symptoms for each disease and checks the detailed situation table for each disease, or records the patient's biosignals obtained in real time through interworking with equipment in real time in the emergency activity log, An AI-based first aid guide method that updates the detailed status table or emergency activity log for each disease in real time according to the patient's condition identified in real time.

Images