Chat robot system based on artificial intelligence for non-face-to-face treatment

Abstract

The present invention provides an artificial intelligence-based chat robot system for non-face-to-face treatment, and the AI-based chat robot system of the present invention provides a user terminal for receiving basic information and symptom information required for non-face-to-face treatment from a user. The chat robot system analyzes a first disease candidate group based on the basic information and a second disease candidate group based on the symptom information, and receives the basic information and the symptom information from the user. And generating treatment information including the final disease candidate group by combining the first disease candidate group and the second disease candidate group. Comprising an expert terminal, and performs non-face-to-face medical treatment on the user by using the medical information generated by the server and the chat robot system server.

Description

TECHNICAL FIELD

[0001] The present application is an AI-based chatbot system for non-face-to-face treatment. More specifically, it relates to an AI-based chatbot system that generates information about a high probability disease based on information collected from a user to perform non-face-to-face treatment. BACKGROUND

[0002] In recent years, with the development of industries and technologies, AI (Artificial Intelligence) is being applied in various fields, and along with it, many services and systems that provide convenience to users are being provided. In particular, the way AI is applied is mostly to provide or recommend a result that a user wants by learning or analyzing input information in advance.

[0003] In this regard, AI is increasingly being applied to replace professional services or services that directly face users, for example, in the medical field, it is used to provide services to users who want to use non-face-to-face patients, etc. It can also be applied to the chatbot system provided.

[0004] At this time, the chatbot system for non-face-to-face treatment can apply AI to analyze information collected from patients and provide appropriate information related to diseases or medical care. However, such an AI-based non-face-to-face treatment chatbot system has limitations in accurately analyzing input information, and thus it is difficult to provide accurate information to experts such as doctors who actually provide treatment.

[0005] In addition, the AI-based non-face-to-face treatment chatbot system generally provides information by simply matching symptoms and diseases, and thus information about diseases can be determined depending on subjective symptoms input by patients, resulting in inaccuracy thereof. There is a risk of improper treatment. SUMMARY

[0006] Problems to be Solved by the Invention

[0007] The present application solves the above-mentioned problems to provide an AI-based chatbot system that generates information about a high probability disease for non-face-to-face treatment based on information collected from a user.

[0008] Means for Solving the Problems

[0009] To achieve the above object, one aspect of the present application provides an AI-based chatbot system for non-face-to-face medical treatment.

[0010] The chatbot system for non-face-to-face medical care includes a user terminal that receives basic and symptom information from the user through a user interface. Upon receiving this information, the chatbot system server and chatbot generate a second candidate disease group based on a first candidate disease group and the symptom information. The system then combines the first and second candidate disease groups to generate medical information including the following: a final candidate disease group includes an expert terminal that provides non-face-to-face medical care to the user using the medical information generated by the system server. The first candidate disease group sequentially determines factors such as gender, age, and place of residence. The chatbot system server further uses disease codes within a range defined by sequentially applying symptom types, symptom regions, periods, and intensities included in the symptom information to generate a disease history included in the basic information. This includes a database built by pre-learning disease codes, gender, age, region, symptoms, location, period, and intensity information for various diseases.

[0011] The chatbot system's server extracts primary diseases from the database's basic information, including gender, age, and region, encompassing various diseases. It then selects diseases from these primary diseases that match the patient's medical history. Next, it extracts the disease codes for secondary diseases, expands them to a preset range, and generates multiple diseases corresponding to this expanded range as the first candidate disease group.

[0012] The chatbot system's server divides the database into one of two groups—low-risk or high-risk—based on the total number of symptom types and symptom regions contained in the symptom information, and detects diseases within a preset range. Diseases with codes are extracted as primary diseases. Based on a comparison between the time period received from the user terminal and a preset reference, the disease codes of the primary diseases are adjusted within a preset range, and secondary diseases are extracted. During this process, based on a comparison between the intensity received from the user terminal and a preset reference intensity, the disease code ranges of secondary diseases are adjusted to a preset range, tertiary diseases are extracted, and the extracted tertiary diseases are classified to generate candidate diseases.

[0013] The chatbot system's server sets the range of disease code sets for high-risk groups to be wider than that for low-risk groups.

[0014] When the period received from the user terminal is longer than the preset reference period, the chatbot system's server expands the disease code of the primary disease into a preset range. Conversely, if the period received from the user terminal exceeds the preset reference period but is less than it, the disease code of the primary disease is reduced into a preset range. However, for the reference period, the average symptom duration of each disease included in the primary disease is extracted from the database, and the average of the average symptom durations for each extracted disease is calculated and used as the reference period.

[0015] When the intensity received from the user terminal is greater than the preset reference intensity, or when the intensity received from the user terminal is less than the preset reference intensity, the chatbot system's server expands the disease code of the minor disease within a preset range. At the same time, the disease code of the minor disease is reduced within a preset range, but the reference intensity is the same as the symptoms of users of the same gender and age who received non-face-to-face treatment. The system checks the intensity in the database, calculates the average value of the confirmed intensities, and sets this as the reference intensity.

[0016] The chatbot system's server determines the range of disease codes by the number of digits in the text used to distinguish disease types. As the text length increases with the number of digits, the range of disease codes also increases. Conversely, reducing the number of digits in the disease code and shortening the text length results in a wider range of disease codes.

[0017] Another objective of this invention in addressing the aforementioned problems is a method for operating an AI-based chatbot system for non-face-to-face medical care, wherein the chatbot system is input by a user through a user interface. The terminal receives necessary basic information and symptom information, and receives basic information and symptom information input from the chatbot system's server through the user interface from the user terminal. Based on the basic information, a first disease candidate group and a second disease candidate group are generated. According to the symptom information, the first and second candidate disease groups are combined to generate treatment information including a final candidate disease group. The chatbot system's server in an expert terminal receives the information and executes the generated treatment. Non-face-to-face medical care is provided to the user using the received medical information, wherein the first disease candidate sequentially determines gender, age, place of residence, and medical history included in the basic information, and the second disease candidate group is generated. The chatbot system's server also includes a database built by learning about disease codes, gender, age, and symptom information. The region, symptoms, location, stage, and intensity of various diseases are known in advance.

[0018] Invention Effects

[0019] The AI-based chatbot system for non-face-to-face treatment according to the present invention, as described above, can improve the accuracy of information required for non-face-to-face treatment by using user information such as patient information, and through this system, information suitable for the user's situation or condition can be provided, which can provide guessing or prediction of the effect of the disease. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of an environment for an AI-based chatbot system for non-face-to-face medical care, according to one embodiment.

[0021] Figure 2This is a diagram of the hardware configuration of a server for an AI-based chatbot system for non-face-to-face medical care, according to an embodiment.

[0022] Figure 3 This is a diagram illustrating the operation of generating a first disease candidate group in an AI-based chatbot system for non-face-to-face treatment, according to an embodiment.

[0023] Figure 4 This is a diagram illustrating the operation of generating a second disease candidate group in an AI-based chatbot system for non-face-to-face treatment, according to an embodiment.

[0024] Figure 5 This is a representative flowchart of operations performed on a server of an AI-based chatbot system for non-face-to-face medical care, according to an embodiment.

[0025] Figure 6 This is a flowchart illustrating the process of generating a first disease candidate group in a server of an AI-based chatbot system for non-face-to-face treatment, according to an embodiment.

[0026] Figure 7 This is a flowchart illustrating the process of generating a second disease candidate group in a server of an AI-based chatbot system for non-face-to-face treatment, according to an embodiment.

[0027] Explanation of reference numerals in the attached figures

[0028] 10: User terminal; 20: Server

[0029] 30: Expert Terminal 100: Server

[0030] 110: Processor 120: Memory

[0031] 130: Transmitting / receiving device; 140: Input interface device

[0032] 150: Output interface device; 160: Storage device

[0033] 170: Bus Detailed Implementation

[0034] Because the present invention can be modified and has various embodiments, specific embodiments will be shown in the accompanying drawings and described in detail in the detailed description. However, this is not intended to limit the invention to the specific embodiments, and it should be understood to include all modifications, equivalents, and substitutions contained within the spirit and scope of the invention. In describing each drawing, similar reference numerals are used for similar parts.

[0035] Terms such as first, second, A, and B may be used to describe various components, but components should not be limited by these terms. The terms above are used only to distinguish one component from another. For example, a first component may be named a second component without departing from the scope of the invention, and similarly, a second component may be named a first component. These terms and / or include any combination of one or more of the various related statements.

[0036] When a component is referred to as "connected" or "connected" to another component, it can be understood as being able to connect directly to or be linked to another component, but other components should also exist in between. On the other hand, when it is said that a component is "directly connected" or "directly linked" to another component, it should be understood that no other components exist in between.

[0037] The terminology used in this application is for describing specific embodiments only and is not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as “comprising” or “having” are intended to indicate the presence of features, numbers, steps, operations, components, portions, or combinations thereof described in the specification, but are not intended to indicate the presence of: it should be understood that they do not preclude the possibility of the presence or addition of elements, numbers, steps, operations, components, portions, or combinations thereof.

[0038] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and should not be interpreted in an ideal or overly formal sense, unless expressly defined in this application No. 2007 / 2007.

[0039] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0040] A schematic diagram of an environment for an AI-based chatbot system for non-face-to-face medical care, according to one embodiment.

[0041] Reference Figure 1 The AI-based chatbot system 1 for non-face-to-face medical care can be executed by a user terminal 10, a chatbot system server 20, and an expert terminal 30.

[0042] User terminal 10 can refer to a user's device, such as a patient seeking non-face-to-face medical care. User terminal 10 can connect to the chatbot system's server 20 via communication and output a user interface through which the user can input the information required for non-face-to-face medical care.

[0043] The user interface can be output in a chat format, where information is provided in the form of a conversation for non-face-to-face processing, and can have a structure that allows users to input the information they want by directly writing or selecting.

[0044] The chatbot system's server 20 receives the information needed for non-face-to-face treatment from the user terminal 10.

[0045] In one embodiment, server 20 can use information received or acquired through user terminal 10 to generate medical information for non-face-to-face medical care, and more specifically, server 20 can generate medical information for non-face-to-face medical care. Medical information can be generated using basic information and symptom information, where basic information refers to relevant basic information and symptom information refers to symptoms confirmed by the user.

[0046] For example, basic information may include the user's gender and age, region (which may refer to the area where the user resides), and personal medical history. Symptom information may include the type and location of the user's symptoms, the duration and intensity of the symptoms, etc.

[0047] In addition, server 20 may include a database containing disease-related information and may be updated when providing non-face-to-face medical treatment.

[0048] For example, the database contains disease codes for various diseases, statistics on the probability of occurrence of each disease by gender, age, and region, and information on symptom types, symptom regions, symptoms, and average symptom duration for each disease. It also includes statistical data on symptom severity categorized by gender and age.

[0049] Here, server 20 can generate a first disease candidate group and a second disease candidate group based on the user's basic information and symptom information, and can use each disease candidate group to generate a final disease candidate group. Subsequently, server 20 can generate the final candidate disease group as medical information for non-face-to-face treatment and send it to expert terminal 30. For example, server 20 can extract disease codes for multiple diseases included in the final disease candidate group from a database and generate medical information in which the extracted disease codes are included.

[0050] Expert terminal 30 can refer to a device belonging to an expert (e.g., a doctor) capable of providing non-face-to-face medical treatment and can use medical methods to provide non-face-to-face medical treatment to users. Information received from server 20. Such expert terminals 30 can be pre-registered on server 20 and matched with user terminals 10 connected to server 20 to provide non-face-to-face medical care to users.

[0051] The expert terminal 30 can output multiple disease codes included in the medical information, and the expert can select at least one disease code corresponding to the user from the multiple disease codes output by the expert terminal 30. At this time, if there is no disease code corresponding to the user among the multiple disease codes output, the expert can directly input the disease code.

[0052] When a new disease code not present in the multiple disease codes output by the expert terminal 30 is input, the server 20 matches the input disease code with the type and region of the symptoms and stores it in the database. This information can be updated. In other words, the server 20 continuously updates the database as it uses the AI-based chatbot system for non-face-to-face treatment according to the embodiment to provide non-face-to-face treatment to multiple users, gradually providing users with more accurate non-face-to-face medical services.

[0053] A diagram illustrating the hardware configuration of a server for an AI-based chatbot system for non-face-to-face medical care, according to an embodiment.

[0054] refer to Figure 2 A server 100 for an AI-based chatbot system for non-face-to-face therapy may include at least one processor 110 and a memory 120 for storing instructions that instruct at least one processor 110 to perform at least one operation.

[0055] At least one of the processors 110 is a central processing unit (CPU), a graphics processing unit (GPU), or a processor that replicates the creative process in the example according to the method of execution, meaning digital.

[0056] Memory 120 is at least one of the configurable numbers, ranging from moderate volatile to moderate non-volatile. For example, memory 120 is a read-only memory (Read-Only Memory, ROM) and a random access memory (Random Access Memory, RAM).

[0057] The operation server 100 is used to perform at least one operation, including input data, intermediate temporary data, and may also include a storage device 160 for storing output data, etc. For example, the storage device 160 is flash memory, hard disk drive (HDD), solid-state drive (SSD), or various memory cards (such as micro SD cards), etc.

[0058] Furthermore, the server 100 of the AI-based chatbot system for non-face-to-face therapy includes a digital transceiver 130 that performs communication performance sending and receiving via a wireless network. Additionally, the server 100 of the AI-based chatbot system for non-face-to-face therapy includes an input interface device 140, a power output interface device 150, a storage device 160, and more, all including digital components. Each element on the server 100 of the AI-based chatbot system for non-face-to-face therapy is connected to each other and communicates digitally via a bus 170.

[0059] For example, the server 100 of the AI-based chatbot system for non-face-to-face processing can be a desktop computer, laptop computer, mobile phone, tablet computer, smartwatch, smart glasses, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player, digital recorder, digital audio player, digital video recorder, digital video player, or PDA (personal digital assistant), etc.

[0060] A graph illustrating the operations for generating a first disease candidate group in an AI-based chatbot system for non-face-to-face treatment according to an embodiment, and Figure 4 This is a diagram illustrating AI used for non-face-to-face treatment. According to an embodiment of face-to-face treatment, this is a diagram explaining the operation of creating a second disease candidate group in a chatbot-based system.

[0061] First, refer to Figure 3 The server can set the range of multiple diseases included in the first disease candidate group by sequentially applying the user's basic information, including gender, age, region, and medical history.

[0062] Specifically, the server can confirm the user's gender, age, and region, and compare these confirmed factors with the gender, age, and region statistics for each disease that has been learned and stored, generating a first candidate disease group. For this purpose, primary diseases can be extracted from the database beforehand. Then, the server can extract diseases that match the user's gender, age, and region as the primary disease.

[0063] For example, a database could have pre-learned and stored probabilities of developing each disease at a specific age. In this case, it's possible to extract the multiple diseases with the highest probability at a user's age. That is, if a user is 20 years old, it's possible to extract the multiple diseases with the highest probability of developing at age 20.

[0064] In this way, the server can extract multiple diseases based on the user's gender, age, region, and any diseases commonly included in each of these multiple diseases. These can be extracted as primary diseases. If no common diseases exist, the server can combine multiple diseases and extract them as primary diseases.

[0065] The server can then apply the user's medical history to the primary disease to extract secondary diseases. More specifically, among the various diseases included in the user's primary disease, the server can extract the diseases the user has a history of. By selection, secondary diseases can be extracted from the primary disease.

[0066] At this point, if the user's medical history does not exist, the server can set the primary disease as a secondary disease. If none of the multiple diseases included in the primary disease list have a corresponding medical history, then the primary disease is set as a secondary disease. This can be achieved by adding additional diseases that correspond to the history of the primary disease.

[0067] You can view the disease codes for each disease included in a secondary disease, expand each confirmed disease code into a preset range, and select multiple diseases corresponding to the disease codes within the expanded range as the first candidate disease group. Here, the preset range can refer to the size of a single digit in the disease code.

[0068] Specifically, a disease code can be represented as 6-digit text, and the disease code can be divided into different diseases or disease groups based on the text set for each digit. In other words, a disease code can be determined by the number of digits in the text used to distinguish the disease type, and as the text length increases due to the increase in the number of digits, it may mean that the disease is specified by a disease code.

[0069] Specifically, when the server narrows down the range of disease codes, it can select a preset number of diseases from among the multiple diseases identified by increasing the number of digits, in order of their highest probability of occurrence. For example, as shown in Table 1 below, the database can pre-learn and store information about the disease code for each digit and the probability of having each disease.

[0070] Table 1

[0071]

[0072]

[0073] By narrowing down the range of disease codes, and given that the disease code range is M100 with 4 digits, the server can adjust the number of digits in the disease code range from 4 to 5. At this point, from the multiple disease codes (M1000 to M1009, a total of 10 disease codes) contained within the 4-digit M100 range, a preset number of diseases can be selected based on the probability of each disease code's prevalence.

[0074] For example, if the preset quantity is 3, the server can select disease codes M1009, M1008, and M1007 in the order of the highest probability of occurrence among the multiple disease codes contained in M100. Accordingly, the range of existing 4-digit disease codes can be reduced from one range to a 5-digit range.

[0075] Expanding the disease code range: Given a disease code range of M1009 with 5 digits, the server can adjust the number of digits in the disease code range from 5 to 4. For example, if the current disease code is M1009, the server can truncate the last digit (9) to change it to M100, thus expanding the disease code range from the current 5 digits to 4 digits.

[0076] In other words, if the disease code range is 5 digits and only disease code M1009 exists, then when the disease code range expands to 4 digits, the 10 disease codes contained in M100 can be selected, thus expanding the disease code range. The range of disease codes may mean increasing the number of diseases that can be selected.

[0077] In other words, disease codes can mean that as text length decreases due to a reduction in the number of bits, the disease group including the diseases indicated by the disease code becomes wider. Conversely, disease codes can mean that as text length increases due to an increase in the number of bits, the disease group including the diseases indicated by the disease code becomes narrower.

[0078] In this way, the server can generate the first candidate disease group.

[0079] At the same time, refer to Figure 4 The server can set the range of multiple diseases included in the second disease candidate group by sequentially applying the symptom types, locations, periods, and intensities included in the user's symptom information.

[0080] Specifically, the server can check the number of symptom types and symptom locations exhibited by the user, and calculate the total number of confirmed cases. The server can then compare the calculated total with preset values ​​to classify risk groups, and can categorize users into either a high-risk group or a low-risk group based on the comparison results.

[0081] For example, within a risk group, a comparable value can be pre-set so that the higher the sum of the number of symptom types and the number of symptom regions, the higher the risk group. For instance, if the preset value for dividing the risk group is 3, and the user's symptom types are fever, cough, and headache, and the symptom region is identified as the head and neck, then the sum of the number of symptom types and the number of symptom regions is 5, and the user can be classified as a high-risk individual.

[0082] Subsequently, the server can extract multiple diseases from the database that correspond to a preset disease range of the classified risk group, and extract the multiple diseases as the main diseases in the first range.

[0083] Here, the server can set different preset disease ranges for each risk group. More specifically, the high-risk group can be set to have a wider disease range than the low-risk group. For example, the server can set a 3-digit disease code for high-risk individuals and a 4-digit disease code for low-risk individuals.

[0084] In other words, the first range can represent a range corresponding to the risk group categorized for the user. Ultimately, the server extracts multiple diseases from the database based on the user's symptom type and symptom region, and extracts the primary disease within the first range by selectively extracting the disease corresponding to the number of digits in the determined first range. This is how you do it for at-risk groups.

[0085] Subsequently, the server can compare the time period (which can refer to the duration of symptoms) included in the symptom information of the primary disease within the first range with a preset reference time period. Here, the preset reference period can be set as the average duration of symptoms of multiple diseases included in the primary disease. For example, if the primary disease includes diseases a, b, and c, then the average duration of symptoms for disease a is 3 days, the average duration of symptoms for disease b is 2 days, and the average duration of symptoms for disease c is 5 days. The averages of 3 days, 2 days, and 5 days can be calculated, and approximately 3 days can be set as the reference period.

[0086] Subsequently, the server can compare the period included in the symptom information with a standard period. If the period included in the symptom information is longer than the standard period, the server can determine a second range by expanding the first range by a preset range. Conversely, if the period included in the symptom information is shorter than the reference period, the server can determine a second range by reducing the first range by a preset range. The server can then consider the newly determined second range based on the primary disease in the first range and extract secondary diseases from the database within the second range.

[0087] Subsequently, the server can adjust the second range to a third range by comparing the intensity of the user's perceived secondary illness within the extracted second range with a preset reference intensity. Here, intensity can be expressed as a number or unit that can distinguish stages, such as 1 to 10 (a larger number may indicate a more severe intensity) to indicate the severity of symptoms.

[0088] Specifically, the server can determine the symptom intensity of users of the same gender by applying statistical data on the intensity of secondary disease symptoms for gender and age in the database, along with the user's gender, age, and symptoms. It can also extract the gender of users of the same age.

[0089] For example, if a user is a 20-year-old male with a cold and cough symptom intensity of 3, the server can calculate the average intensity of cold and cough symptoms over 20 years in the database – for older men. If the server then determines that the calculated average intensity is 5, it can set 5 as the standard intensity. The server can then compare 3 (the intensity value entered by the user as symptom information) with 5 (the standard intensity set by the database).

[0090] Subsequently, if the user's intensity is higher than the reference intensity, the server can determine the third range by expanding the second range (i.e., the range of secondary diseases) by a preset range. Conversely, if the user's intensity is lower than the reference intensity, the server can determine the third range by reducing the second range (which is the range of secondary diseases) by a preset range. The server can then extract the third disease from the database by considering the newly determined third range from the second diseases within the second range.

[0091] The third disease range extracted in this way can be used as a second disease candidate group.

[0092] The server can be configured as shown in the reference. Figure 3 and Figure 4 The generated first and second disease candidate groups are used to generate the final disease candidate group, and disease codes for various diseases included in the final disease candidate group can be generated. This can be created using medical information.

[0093] The server can transmit medical information to the expert's terminal, allowing the expert to output the medical information and provide non-face-to-face medical treatment based on the information provided by the server. Meanwhile, the content entered by the expert about the user can be output to the user interface of the user terminal through the system server.

[0094] When providing non-face-to-face medical care, if the disease code determined by the expert terminal exists on the server, the server will increase the correlation between the user's disease and symptoms and the disease code. The treatment information provided to the expert terminal can be configured. Conversely, when providing non-face-to-face medical care, if the expert terminal determines a disease code for the user, the server will determine the correlation between the user's disease and symptoms and the disease code. This correlation is not present in the medical information provided to the expert terminal. This can be set to low.

[0095] Subsequently, if the correspondence between a disease or symptom and its disease code is lower than a preset value, the server can exclude newly added cases from the disease codes for that disease or symptom. On the other hand, if the correspondence between a disease and its symptom and its disease code is greater than a preset value, and the corresponding disease, symptom, and disease code are output from the expert terminal, this information can be separately marked and output as highly reliable information due to its high correspondence.

[0096] For example, the server can shade or color diseases, symptoms, and disease codes with a match score higher than a preset value to distinguish them from the other disease codes provided, thus ensuring a high match score and providing highly matched disease codes. You can request information through the expert terminal.

[0097] Figure 5 This is a representative flowchart illustrating the operations performed on a server in an AI-based chatbot system for non-face-to-face medical care according to an embodiment, and Figure 6 This is a representative flowchart illustrating an AI-based chatbot system for non-face-to-face medical care. Face-to-face medical care according to an embodiment is illustrated by a flowchart showing the operation of generating a first disease candidate group in a server, and... Figure 7 This is a flowchart illustrating the operation of generating a second disease candidate group in a server. According to one embodiment, an AI-based chatbot system for non-face-to-face treatment is also described.

[0098] First, see Figure 5 The operation method of the AI-based non-face-to-face medical chatbot system can be executed by a user terminal, a chatbot system server, and an expert terminal. It mainly involves steps S100, receiving basic information and symptom information from the user terminal, generating a first candidate disease group in the chatbot system server S200, creating a second candidate disease group S300, generating medical information S400, and providing medical care through the expert terminal non-face-to-face, which may include steps S500.

[0099] Specifically, see Figure 6The chatbot system uses gender, history, and region to extract primary diseases to create a first disease candidate group on the server S210, and applies history to the primary diseases to obtain the extraction step S220, and generates the first disease candidate group by expanding secondary diseases to a preset range S230.

[0100] In this process, the server uses the disease codes for each disease previously learned and stored in the database to extract and create primary, secondary, and minor disease candidates, adjusting the range of disease codes by varying the number of candidates. The numbers in the disease codes can be processed, and related details can be referenced from previous documents. Figure 3 and Figure 4 The details described are similar or identical.

[0101] Additionally, refer to Figure 7 The chatbot system uses symptom type and symptom region to divide risk groups to generate a second disease candidate group 310 in the server, and can perform the steps of extracting the primary disease based on the range of the risk group 310S320, extracting the secondary disease by adjusting the range of the primary disease usage cycle S330, extracting the tertiary disease by adjusting the range of the secondary disease usage intensity 340, and extracting the tertiary disease S350 to generate the second disease candidate group.

[0102] In this process, the server can use the disease codes for each disease, pre-learned and stored in the database, to extract first to third diseases to create a first candidate disease group and a second candidate disease group. The range of disease codes is determined by adjusting the number of bits in the codes. Further details can be found in the previous reference. Figure 3 and Figure 4 The details described are similar or identical.

[0103] The method according to the invention can be implemented in the form of program instructions, which can be executed by various computer devices and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., which may be included individually or in combination. The program instructions recorded on the computer-readable medium may be specifically designed and constructed for this invention, or may be known and available to those skilled in the art of computer software.

[0104] Examples of computer-readable media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions may include machine language code, such as code generated by a compiler, and high-level language code that can be executed by a computer using an interpreter. The aforementioned hardware devices may be configured to operate in conjunction with at least one software module to perform the operations of the present invention, and vice versa.

[0105] Furthermore, the above-described methods or apparatus are implemented by combining all or part of their configurations or functions, or they can be implemented individually.

[0106] The invention has been described above with reference to preferred embodiments; however, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention as set forth in the appended claims. I understand you can do it.

Claims

1. An AI-based chatbot system for non-face-to-face therapy, wherein, The AI-based chatbot system for non-face-to-face therapy includes: The user terminal receives basic and symptom information required for non-face-to-face treatment from the user through a user interface. The server receives basic information and symptom information from the user terminal, generates a first candidate disease group based on the basic information and a second candidate disease group based on the symptom information, and combines the first and second candidate disease groups. The chatbot system then generates medical information, including the final candidate disease groups. Expert terminals utilize medical information generated by the chatbot system's server to provide users with non-face-to-face medical care. A first disease candidate group is created using disease codes within a range defined by sequentially applying basic information including gender, age, place of residence, and medical history. Symptom information is then used to include disease codes within a range defined by sequentially applying symptom type, symptom region, duration, and intensity. The chatbot system's server also includes a database built by pre-learning information such as disease codes, gender, age, region, symptoms, location, duration, and intensity of various diseases.

2. The AI-based chatbot system for non-face-to-face therapy according to claim 1, wherein, In the chatbot system's server, From the database, the primary diseases of the first disease candidate group are extracted from multiple diseases included in the basic information such as gender, age, and region. Diseases matching the medical history are then extracted from the extracted secondary diseases to form the first candidate disease group. The disease codes of the secondary diseases in the first candidate disease group are expanded to a preset range. Multiple diseases corresponding to the disease codes within the expanded range are selected as the first candidate diseases to create a disease candidate group. Based on the number of symptom types and symptom regions included in the symptom information, as well as the total number of diseases in the disease codes, the risk group is divided into one of a low-risk group and a high-risk group. The pre-defined range of the divided risk group is designated as the first risk group. Primary diseases of candidate disease groups are extracted, and the range of the first risk group is determined by comparing the period received from the user terminal with a pre-defined reference period. The disease codes of the primary diseases in the second candidate disease group are adjusted to a pre-defined range. Secondary diseases of the second candidate disease group are extracted and their disease code ranges are pre-defined. Based on the comparison of the intensity received from the user terminal with a pre-defined standard intensity, tertiary diseases are extracted by adjusting the set range, and these extracted tertiary diseases are generated as the second disease candidate group. The chatbot system's server sets the range of disease code sets for high-risk groups to be wider than that for low-risk groups.

3. The AI-based chatbot system for non-face-to-face therapy according to claim 2, wherein, In the chatbot system's server, If the period received from the user terminal is greater than the preset reference period, the disease code of the primary disease is expanded to a preset range. If the period received from the user terminal is less than the preset reference period, the disease code of the primary disease is expanded to a predetermined range and the disease code is reduced to a preset range. However, for the reference period, the average duration of symptoms for each disease included in the primary disease is extracted from the database, and the average of the extracted average duration of symptoms for each disease is calculated and set as the reference period. If the intensity received from the user terminal is greater than the preset reference intensity, the disease code for the minor disease is expanded within the preset range. If the intensity received from the user terminal is less than the preset reference intensity, the disease code is expanded to the preset range. However, the standard intensity is determined by examining the intensity of symptoms identical to those of users receiving non-face-to-face treatment. This involves comparing the symptoms of users with the same gender and age in the database, determining the intensity of their symptoms, calculating the average intensity, and setting this average as the reference intensity. The range of disease codes is determined by the number of digits in the text used to distinguish disease types. As the text length increases with the number of digits, the range of disease codes narrows, and as the text length increases, the number of digits decreases, and the range of disease codes expands.

Images