Intelligent hospital guide method and system, terminal and storage medium

Abstract

The invention belongs to the technical field of medical assistance, and particularly relates to an intelligent hospital guide method and system, a terminal and a storage medium, and the method comprises the steps: obtaining disease information inputted on a patient terminal, inputting the gender and age of a patient on the patient terminal through a touch mode, and inputting disease part information based on a human body part map; or directly inputting disease information through a question and answer mode; according to the invention, the patient can input the disease information on the patient terminal through the touch control mode or the question and answer mode, so that the targeted department information and the doctor scheduling condition can be obtained, and the registration appointment can be easily completed. According to the convenient operation mode, the traditional hospital guide and registration process is simplified, and the medical experience of the patient is remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of medical assistance technology, and specifically relates to an intelligent diagnosis guidance method, system, terminal and storage medium. Background Art

[0002] In the modern medical system, patients face many inconveniences in the process of medical guidance and registration. The traditional medical guidance method mainly relies on the manual service desk of the hospital, which is not only inefficient, but also often causes patients to wait in line for a long time during peak hours, affecting the medical experience. In addition, due to the limitations of manual services, the accuracy of medical guidance cannot be fully guaranteed. Patients sometimes waste time and energy because of registering for the wrong department, and even delay their condition.

[0003] With the advancement of intelligence and informatization, and the rapid development of the "Internet + medical health" model, hospitals have begun to try to introduce new guidance methods to improve patients' medical experience. For example, some hospitals have introduced self-service registration machines and online appointment systems, which have alleviated the pressure on the manual service desk to a certain extent and improved the registration efficiency. However, these systems usually have single functions and lack intelligent guidance services. Patients still need to judge which department to register when using them, which is still a problem for patients who lack medical knowledge. Summary of the invention

[0004] In view of the defects in the prior art that the existing medical guidance system requires patients to judge by themselves which department to go to, and patients who lack medical knowledge sometimes waste time and energy or even delay their condition because of going to the wrong department, the present invention provides an intelligent medical guidance method, system, terminal and storage medium to solve the above technical problems.

[0005] In a first aspect, the present invention provides an intelligent diagnosis guidance method, comprising: Acquiring disease information inputted on a patient terminal, where the patient inputs gender and age through a touch mode and diseased part information based on a human body part diagram on the patient terminal, or directly inputs disease information through a question-and-answer mode; Selecting a pre-stored portrait disease recognition model or a pre-stored language disease recognition model according to the input mode of the patient's disease information to analyze the disease information input by the patient, outputting three disease names that are most related to the input disease information, and sending them to the patient terminal; According to one of the three disease names fed back by the patient terminal, the corresponding department information stored in the database is extracted and sent to the patient terminal; the department information includes the department name, department doctor information, doctor scheduling information and the number source of the preset appointment days corresponding to the current department; Make a registration appointment based on the appointment status feedback from the patient terminal; if the appointment is for treatment in the current department, an invitation to enter the patient's name and ID number will be sent to the patient terminal; if the appointment is not for treatment in the current department, the department information of the remaining two departments will be sent to the patient terminal until the patient successfully completes the registration appointment.

[0006] A further improvement of the technical solution is that the method for constructing the language disorder recognition model is: Collecting the symptom information inputted through the question-and-answer mode from the patient terminal, setting corresponding disease labels for the symptom information, and integrating the symptom information with disease labels into a language symptom training set and a language symptom test set; the symptom information includes symptom description, patient gender, and patient age; Use Tokenizer to segment the collected disease information and build a vocabulary containing all preset words; Use the pad_sequences() function to unify all text sequences in the vocabulary to a preset fixed length; Use the pre-trained GloVe model to convert each word in the vocabulary into a high-dimensional vector that can capture the semantic relationship between words; According to the vocabulary and high-dimensional vectors, a word embedding matrix is ​​constructed as the initial weight of the Embedding layer of the initial language disorder recognition model; Iteratively training the initial language disorder recognition model using the language disorder training set to obtain a language disorder recognition model; The Adam optimizer is used to update the parameters of the language disorder recognition model, and Binary Cross-Entropy is used as the loss function. In each iteration, the language disorder recognition model calculates the loss based on the current weights and updates the weights through back propagation. The performance indicators of the language disorder recognition model were evaluated using the language disorder test set. The performance indicators included calculating the precision, recall and F1 score. Based on the evaluation results, the language disorder recognition model is adjusted and optimized.

[0007] A further improvement of this technical solution is that the formula of the Binary Cross-Entropy loss function is: ; in, is the average binary cross entropy loss, which represents the average difference between the disease probability distribution predicted by the language disease recognition model and the true disease label; is the total number of samples; i is the sample index, which is an integer from 1 to N, used to traverse each sample in the language disorder training set or the language disorder test set; is the true disease label of the i-th sample; is the disease prediction probability of the i-th sample.

[0008] A further improvement of the technical solution is that the initial language disorder recognition model includes an Embedding layer, an LSTM layer, an Attention layer and an output layer, and the Embedding layer is sequentially connected to the output layer through the LSTM layer and the Attention layer; The Embedding layer is the first layer of the initial language disease recognition model. It uses the word embedding matrix to convert the disease information text input through the patient terminal into a high-dimensional vector sequence and transmits it to the LSTM layer. The LSTM layer processes the received high-dimensional vector sequence and captures the contextual information in the high-dimensional vector sequence; The Attention layer adaptively assigns different attention weights according to the importance of different parts in the input high-dimensional vector sequence to identify key disease information in the input disease information text; As the last layer of the initial language disease recognition model, the output layer uses the pre-stored softmax function to decode and classify the weighted high-dimensional vector sequence passed from the Attention layer, obtains the corresponding disease category, and outputs the three disease names in the disease category that are most correlated with the input disease information.

[0009] A further improvement of this technical solution is that the formula of the softmax function is: ; in, is the predicted probability of the h-th disease category; is the original output of the h-th disease category, that is, the confidence of the initial language disease recognition model for the disease category; is the index of the original output of the hth disease category; j is the disease category index, which is used to traverse all disease categories.

[0010] In a second aspect, the present invention provides an intelligent medical guidance system, including a data collection layer, a data processing layer, a service layer and an application layer, wherein the data collection layer is sequentially connected to the application layer through the data processing layer and the service layer; The data collection layer is used to collect basic information of the hospital and the disease information of the patients. The basic information of the hospital includes hospital information, disease information and department information; The data processing layer pre-processes the collected basic hospital information and patient disease information, including data cleaning, data classification and data mining; The service layer is used to store and analyze the basic hospital information and patient disease information processed by the data processing layer, and provide guidance information to patients; The application layer is used to provide the patient-oriented application interface.

[0011] A further improvement of this technical solution is that the service layer includes: A disease reasoning module is used to analyze the disease information input by the patient according to a pre-stored portrait disease recognition model or a pre-stored language disease recognition model, and output three disease names that are most related to the input disease information; The intelligent diagnosis guidance module is used to extract the corresponding department information stored in the database according to the disease name selected by the patient and the pre-processed basic hospital information; the department information includes the department name, department doctor information, doctor scheduling information and the number source of the preset appointment days corresponding to the current department; The appointment registration module is used to make appointments based on the appointment status reported by patients.

[0012] A further improvement of this technical solution is that the application layer includes a patient terminal APP and a mini program.

[0013] In a third aspect, a terminal is provided, including: processor, memory, wherein: The memory is used to store computer programs. The processor is used to call and run the computer program from the memory, so that the terminal executes the above-mentioned terminal method.

[0014] According to a fourth aspect, a computer storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the methods described in the above aspects.

[0015] The beneficial effects of the present invention are: Improve the efficiency and accuracy of medical guidance: By introducing the portrait disease recognition model and the language disease recognition model, the present invention can intelligently analyze the disease information input by the patient and quickly output the three disease names that are most related to the input disease information. This greatly improves the efficiency and accuracy of medical guidance and reduces the time and energy wasted by patients due to wrong department visits.

[0016] Optimize the patient medical experience: Patients can enter their symptoms through touch mode or question-and-answer mode on the patient terminal to obtain targeted department information and doctor schedule information, so as to easily complete the registration and appointment. This convenient operation method simplifies the traditional guidance and registration process, and significantly improves the patient's medical experience.

[0017] Reducing the pressure on manual service in hospitals: The intelligent medical guidance system of the present invention can automatically process the disease information and registration requirements of a large number of patients, effectively reducing the pressure on the manual service desk of the hospital. This enables the hospital to allocate resources more efficiently and improve the overall service quality and efficiency.

[0018] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention.

[0021] Figure 2 is a schematic block diagram of a system according to an embodiment of the present invention.

[0022] Figure 3 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention.

[0023] 210 is the data collection layer, 211 is the hospital information collection module, 212 is the disease information collection module, 213 is the department information collection module, 2131 is the department setting unit, 2132 is the doctor information collection unit, 2133 is the doctor scheduling information collection unit, 2134 is the number source collection unit, 220 is the data processing layer, data cleaning, data classification, data mining, 230 is the service layer, 231 is the disease reasoning module, 232 is the intelligent guidance module, 233 is the appointment registration module, 240 is the application layer, 241 is the patient terminal APP, and 242 is the mini program. DETAILED DESCRIPTION

[0024] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in this specific embodiment. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of them. Based on the embodiments in this patent, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this patent.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0026] Figure 1 is a schematic flow chart of an intelligent diagnosis guidance method according to an embodiment of the present invention. Figure 1 The execution subject may be an intelligent medical guidance system. According to different requirements, the order of the steps in the flowchart may be changed, and some may be omitted.

[0027] like Figure 1 As shown, the method includes: Step 110, obtaining the disease information inputted on the patient terminal, the patient inputs the gender and age through the touch mode on the patient terminal and inputs the diseased part information based on the human body part diagram, or directly inputs the disease information through the question-and-answer mode; Step 120, selecting a pre-stored portrait disease recognition model or a pre-stored language disease recognition model according to the input mode of the patient's disease information to analyze the disease information input by the patient, outputting three disease names that are most relevant to the input disease information, and sending them to the patient terminal; Step 130, extracting the corresponding department information stored in the database according to one of the three disease names fed back by the patient terminal, and sending it to the patient terminal; the department information includes the department name, department doctor information, doctor scheduling information and the number source of the preset appointment days corresponding to the current department; Step 140, make a registration appointment based on the appointment status feedback from the patient terminal; if the appointment is for seeing a doctor in the current department, send an invitation to the patient terminal to enter the patient's name and patient ID number; if the appointment is not for seeing a doctor in the current department, send the department information of the remaining two departments to the patient terminal until the patient successfully completes the registration appointment.

[0028] To facilitate understanding of the present invention, the intelligent medical guidance method provided by the present invention is further described below based on the principle of the intelligent medical guidance method of the present invention and in combination with the process of performing intelligent medical guidance on patients in the embodiments.

[0029] Specifically, the method for constructing a portrait disease recognition model includes: S1211, collecting human body part images with diseased part information and the age and gender of the corresponding patients, and setting corresponding disease labels for the human body part images to form a portrait disease data set; S1212, preprocessing the human body part image in the portrait disease dataset, including cropping, scaling and denoising; S1213, inputting the preprocessed human body part image into the convolution layer and the pooling layer through the image input layer of the convolutional neural network to extract image features; S1214, inputting the feature vector output by the pooling layer into the initial portrait disease recognition model for training, thereby obtaining the portrait disease recognition model; S1215. Deploy the trained portrait symptom recognition model to the data processing layer of the intelligent diagnosis guidance system to realize intelligent analysis and diagnosis guidance of the symptom information input by the patient.

[0030] The present invention only requires patients to upload their own symptom pictures and input relevant information (such as age, gender, etc.), and then automatically analyze and give corresponding symptom identification and registration suggestions. This not only improves the accuracy of medical guidance, but also improves the efficiency of medical treatment for patients, while reducing the burden on medical staff.

[0031] In addition, the construction method of the language disorder recognition model is: S1221. Collecting symptom information inputted through the question-and-answer mode from the patient terminal, setting corresponding disease labels for the symptom information, and integrating the symptom information with the disease labels into a language symptom training set and a language symptom test set; the symptom information includes symptom description, patient gender, and patient age; S1222, using Tokenizer to perform word segmentation on the collected disease information, and construct a vocabulary containing all preset words; S1223, using the pad_sequences() function, unifying all text sequences in the vocabulary into a preset fixed length; S1224. Use the pre-trained GloVe model to convert each word in the vocabulary into a high-dimensional vector that can capture the semantic relationship between words. S1225. Construct a word embedding matrix based on the vocabulary and the high-dimensional vector as the initial weight of the Embedding layer of the initial language disease recognition model; S1226, training the initial language disorder recognition model through iteration using the language disorder training set to obtain a language disorder recognition model; S1227, using the Adam optimizer to update the parameters of the language disorder recognition model, and using BinaryCross-Entropy as the loss function; in each iteration, the language disorder recognition model calculates the loss according to the current weight, and updates the weight through back propagation; S1228. Evaluate the performance indicators of the language disorder recognition model using the language disorder test set, where the performance indicators include calculating the precision, recall and F1 (F1 Score, F Score, F Measure, balanced F score) score; S1229. Adjust and optimize the language disorder recognition model based on the evaluation results.

[0032] The present invention collects the symptom information input through the question-and-answer mode from the patient terminal, and sets accurate disease labels for it, thereby constructing a language symptom training set and test set containing rich symptom descriptions, patient gender, age and other information. This provides a high-quality data basis for the training of the model, and helps the model learn the accurate mapping relationship between symptom information and disease labels. Using Tokenizer to segment the symptom information and construct a vocabulary containing all preset words helps the model to be more efficient and accurate when processing text. At the same time, the pad_sequences() function is used to unify the text sequence into a preset fixed length, ensuring the consistency of the model input and improving the stability and generalization ability of the model. By using the pre-trained GloVe model to convert each word in the vocabulary into a high-dimensional vector, these high-dimensional vectors can capture the semantic relationship between words, thereby helping the model to better understand the semantic content in the symptom information. This helps the model to be more accurate and comprehensive when identifying symptoms. The word embedding matrix constructed according to the vocabulary and high-dimensional vectors is used as the initial weight of the Embedding layer of the initial language symptom recognition model, providing the model with rich semantic information. This helps the model learn the association between symptom information and disease labels more quickly during training, improving the model's training efficiency and accuracy. By using the LSTM (Long Short-Term Memory layer) layer to capture contextual information in high-dimensional vector sequences, and the Attention layer to adaptively assign attention weights to identify key symptom information in the input symptom information text, the model can understand symptom information more accurately and improve the accuracy of disease recognition.

[0033] Furthermore, the formula of Binary Cross-Entropy loss function is: ; in, is the average binary cross entropy loss, which represents the average difference between the disease probability distribution predicted by the language disease recognition model and the true disease label; is the total number of samples; i is the sample index, which is an integer from 1 to N, used to traverse each sample in the language disorder training set or the language disorder test set; is the true disease label of the i-th sample; is the disease prediction probability of the i-th sample.

[0034] The present invention uses the Adam optimizer and the Binary Cross-Entropy loss function to train the model, which helps the model converge to the optimal solution faster during the training process and improves the performance of the model.

[0035] Among them, the initial language disease recognition model includes an Embedding layer, an LSTM layer, an Attention layer and an output layer. The Embedding layer is connected to the output layer through the LSTM layer and the Attention layer in turn; the Embedding layer, as the first layer of the initial language disease recognition model, uses a word embedding matrix to convert the disease information text input through the patient terminal into a high-dimensional vector sequence and transmits it to the LSTM layer; the LSTM layer processes the received high-dimensional vector sequence and captures the contextual information in the high-dimensional vector sequence; the Attention layer adaptively allocates different attention weights according to the importance of different parts in the input high-dimensional vector sequence, and identifies the key disease information in the input disease information text; the output layer, as the last layer of the initial language disease recognition model, uses a pre-stored softmax function to decode and classify the weighted high-dimensional vector sequence passed from the Attention layer, obtains the corresponding disease category, and outputs the three disease names in the disease category that are most correlated with the input disease information.

[0036] Specifically, the formula of the softmax function is: ; in, is the predicted probability of the h-th disease category; is the original output of the h-th disease category, that is, the confidence of the initial language disease recognition model for the disease category; is the index of the original output of the hth disease category; j is the disease category index, which is used to traverse all disease categories.

[0037] The output layer in the present invention uses the pre-stored softmax function to decode and classify the weighted high-dimensional vector sequence, obtain the corresponding disease category, and output the three disease names that are most related to the input symptom information. This provides patients with more accurate and reliable disease diagnosis suggestions, helping them to seek medical treatment in time and receive appropriate treatment.

[0038] like Figure 2 As shown, the present invention provides an intelligent medical guidance system, including a data collection layer, a data processing layer, a service layer and an application layer. The data collection layer is connected to the application layer through the data processing layer and the service layer in turn.

[0039] Among them, the data collection layer is used to collect the basic information of the hospital and the patient's disease information. The basic information of the hospital includes hospital information, disease information and department information; the data processing layer pre-processes the collected basic information of the hospital and the patient's disease information. The preprocessing includes data cleaning, data classification and data mining; the service layer is used to store and analyze the basic information of the hospital and the patient's disease information processed by the data processing layer, and provide guidance information to patients; the application layer is used to provide an application interface for patients.

[0040] Specifically, the data collection layer includes a hospital information collection module, a disease information collection module and a department information collection module. The department information collection module includes a department setting unit, a doctor information collection unit, a doctor scheduling information collection unit and an appointment number situation collection unit. The hospital information collection module is used to collect information such as the geographical location of the hospital and the hospital grade. The disease information collection module is used to obtain symptoms of several categories of diseases and to obtain patient symptom information through the patient terminal in touch mode and question-and-answer mode. The department setting unit is used to set the department name and department doctor configuration information. The doctor information collection unit is used to obtain the basic information of doctors in each department, including name, education, major, etc. The doctor scheduling information collection unit is used to obtain the scheduling information of doctors in each department. The appointment number source collection unit is used to obtain the registration number source information in each department.

[0041] In addition, the service layer includes a disease reasoning module, an intelligent diagnosis guidance module and an appointment registration module; the disease reasoning module is used to analyze the symptom information input by the patient according to a pre-stored portrait symptom recognition model or a pre-stored language symptom recognition model, and output the three disease names that are most correlated with the input symptom information; the intelligent diagnosis guidance module is used to extract the pre-stored corresponding department information from the database according to the disease name selected by the patient and combined with the pre-processed basic hospital information; the department information includes the department name, department doctor information, doctor scheduling information and the number source for the preset appointment days corresponding to the current department; the appointment registration module is used to make registration appointments based on the appointment situation reported by the patient.

[0042] In addition, the application layer includes the patient terminal APP (Application) and mini-programs. The patient terminal APP or mini-program provides an application interface corresponding to the touch mode (portrait disease recognition model) and an application interface corresponding to the question and answer mode (language disease recognition model), allowing patients to upload a human body part map with diseased part information (the human body part map is stored in the system, and the patient can directly click on the diseased part in the touch mode) and describe the disease through the question and answer interface (the question and answer mode includes text question and answer and voice question and answer, in which the voice question and answer needs to be converted into text for disease recognition after the question and answer) to obtain targeted department information and registration suggestions.

[0043] Figure 3 The present invention provides a schematic diagram of the structure of a terminal 300 provided in an embodiment of the present invention. The terminal 300 can be used to execute the intelligent diagnosis guidance method provided in an embodiment of the present invention.

[0044] The terminal 300 may include: a processor 310, a memory 320 and a communication module 330. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention, and it may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0045] The memory 320 can be used to store the execution instructions of the processor 310, and the memory 320 can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the terminal 300 can perform some or all of the steps in the following method embodiments.

[0046] The processor 310 is the control center of the storage terminal, and uses various interfaces and lines to connect various parts of the entire electronic terminal. It runs or executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic terminal and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 310 can only include a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0047] The communication module 330 is used to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.

[0048] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0049] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes, including several instructions for enabling a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention.

[0050] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0051] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or modules, which can be electrical, mechanical or other forms.

[0052] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0053] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0054] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. An intelligent diagnosis guidance method, characterized in that: include: Acquiring disease information inputted on a patient terminal, where the patient inputs gender and age through a touch mode and diseased part information based on a human body part diagram on the patient terminal, or directly inputs disease information through a question-and-answer mode; Selecting a pre-stored portrait disease recognition model or a pre-stored language disease recognition model according to the input mode of the patient's disease information to analyze the disease information input by the patient, outputting three disease names that are most related to the input disease information, and sending them to the patient terminal; According to one of the three disease names fed back by the patient terminal, the corresponding department information stored in the database is extracted and sent to the patient terminal; the department information includes the department name, department doctor information, doctor scheduling information and the number source of the preset appointment days corresponding to the current department; Make a registration appointment based on the appointment status feedback from the patient terminal; if the appointment is for treatment in the current department, an invitation to enter the patient's name and ID number will be sent to the patient terminal; if the appointment is not for treatment in the current department, the department information of the remaining two departments will be sent to the patient terminal until the patient successfully completes the registration appointment.

2. The intelligent diagnosis guidance method according to claim 1, characterized in that: The construction method of the language disorder recognition model is as follows: Collecting the symptom information inputted through the question-and-answer mode from the patient terminal, setting corresponding disease labels for the symptom information, and integrating the symptom information with disease labels into a language symptom training set and a language symptom test set; the symptom information includes symptom description, patient gender, and patient age; Use Tokenizer to segment the collected disease information and build a vocabulary containing all preset words; Use the pad_sequences() function to unify all text sequences in the vocabulary to a preset fixed length; Use the pre-trained GloVe model to convert each word in the vocabulary into a high-dimensional vector that can capture the semantic relationship between words; According to the vocabulary and high-dimensional vectors, a word embedding matrix is ​​constructed as the initial weight of the Embedding layer of the initial language disorder recognition model; Iteratively training the initial language disorder recognition model using the language disorder training set to obtain a language disorder recognition model; The Adam optimizer is used to update the parameters of the language disorder recognition model, and Binary Cross-Entropy is used as the loss function. In each iteration, the language disorder recognition model calculates the loss based on the current weights and updates the weights through back propagation. The performance indicators of the language disorder recognition model were evaluated using the language disorder test set. The performance indicators included calculating the precision, recall and F1 score. Based on the evaluation results, the language disorder recognition model is adjusted and optimized.

3. The intelligent diagnosis guidance method according to claim 2, characterized in that: The formula for the Binary Cross-Entropy loss function is: ； in, is the average binary cross entropy loss, which represents the average difference between the disease probability distribution predicted by the language disease recognition model and the true disease label; is the total number of samples; i is the sample index, which is an integer from 1 to N, used to traverse each sample in the language disorder training set or the language disorder test set; is the true disease label of the i-th sample; is the disease prediction probability of the i-th sample.

4. The intelligent diagnosis guidance method according to claim 2, characterized in that: The initial language disorder recognition model includes an Embedding layer, an LSTM layer, an Attention layer, and an output layer. The Embedding layer is connected to the output layer through the LSTM layer and the Attention layer in turn. The Embedding layer is the first layer of the initial language disease recognition model. It uses the word embedding matrix to convert the disease information text input through the patient terminal into a high-dimensional vector sequence and transmits it to the LSTM layer. The LSTM layer processes the received high-dimensional vector sequence and captures the contextual information in the high-dimensional vector sequence; The Attention layer adaptively assigns different attention weights according to the importance of different parts in the input high-dimensional vector sequence to identify key disease information in the input disease information text; As the last layer of the initial language disease recognition model, the output layer uses the pre-stored softmax function to decode and classify the weighted high-dimensional vector sequence passed from the Attention layer, obtains the corresponding disease category, and outputs the three disease names in the disease category that are most correlated with the input disease information.

5. The intelligent diagnosis guidance method according to claim 4, characterized in that: The formula for the softmax function is: ； in, is the predicted probability of the h-th disease category; is the original output of the h-th disease category, that is, the confidence of the initial language disease recognition model for the disease category; is the index of the original output of the hth disease category; j is the disease category index, which is used to traverse all disease categories.

6. An intelligent medical guidance system, characterized in that: It includes data collection layer, data processing layer, service layer and application layer. The data collection layer is connected to the application layer through the data processing layer and service layer in turn. The data collection layer is used to collect basic information of the hospital and the disease information of the patients. The basic information of the hospital includes hospital information, disease information and department information; The data processing layer pre-processes the collected basic hospital information and patient disease information, including data cleaning, data classification and data mining; The service layer is used to store and analyze the basic hospital information and patient disease information processed by the data processing layer, and provide guidance information to patients; The application layer is used to provide the patient-oriented application interface.

7. The intelligent medical guidance system according to claim 6, characterized in that: The service layer includes: A disease reasoning module is used to analyze the disease information input by the patient according to a pre-stored portrait disease recognition model or a pre-stored language disease recognition model, and output three disease names that are most related to the input disease information; The intelligent diagnosis guidance module is used to extract the corresponding department information stored in the database according to the disease name selected by the patient and the pre-processed basic hospital information; the department information includes the department name, department doctor information, doctor scheduling information and the number source of the preset appointment days corresponding to the current department; The appointment registration module is used to make appointments based on the appointment status reported by patients.

8. The intelligent medical guidance system according to claim 6, characterized in that: The application layer includes patient terminal APP and mini-programs.

9. A terminal, characterized in that: include: processor; A memory for storing execution instructions of the processor; The processor is configured to execute the method according to any one of claims 1 to 5.

10. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images