Endocrine-related disease examination guidance app system and method

Abstract

According to the endocrine-related disease examination guidance app system and method provided by the invention, based on the analysis of the endocrine disease knowledge graph, the recommended examination items are more targeted, the examination required by the endocrine disease possibly suffered by the patient can be accurately covered, the accuracy of disease diagnosis is improved, the unnecessary examination items are reduced, and the examination efficiency is improved. The medical cost of the patient is reduced. The positioning information of the patient is obtained by calling the positioning function of the mobile phone, and the positioning information and the selected examination item are packaged together and sent to the medical institution adaptation module. The adaptive endocrine examination mechanism is recommended according to the positioning information of the patient, so that the patient can conveniently select a professional adaptive mechanism close to the patient, the trouble that the patient finds a proper examination mechanism is solved, and the medical experience of the patient is improved. And risk assessment is performed on the examination result and a report is generated, so that the patient can know own health condition and potential disease risk in the first time, and further measures can be taken in time by the patient.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical management technology, and in particular to an endocrine-related disease examination guidance app system, endocrine-related disease examination guidance method, electronic device and computer-readable storage medium. Background Technology

[0002] Examinations for endocrine-related diseases mainly include laboratory tests and imaging examinations. Laboratory tests cover basal hormone level measurements (such as thyroid function, sex hormones, adrenal hormones, insulin, etc.), dynamic functional tests (such as glucose tolerance test, dexamethasone suppression test, thyrotropin-releasing hormone stimulation test, etc.), autoantibody detection (such as thyroid peroxidase antibody, insulin antibody, etc.), and routine blood and urine tests, electrolyte tests, liver and kidney function tests, etc. Imaging examinations commonly use ultrasound (thyroid gland, adrenal gland, ovary, etc.), CT, MRI (pituitary gland, hypothalamus, adrenal gland, etc.), and nuclear medicine examinations (such as thyroid radionuclide scans) when necessary. In addition, some diseases require genetic testing or pathological biopsy for definitive diagnosis.

[0003] However, existing clinical endocrine-related disease tests have the following major application limitations:

[0004] The traditional medical treatment model is lagging behind: the current diagnosis of endocrine diseases relies on hospital biochemical tests, and patients need an average of 2-3 outpatient visits from the onset of symptoms to diagnosis, which delays the opportunity for early intervention (data from the Chinese Journal of Endocrinology in 2024).

[0005] Existing health apps have limitations: mainstream health management applications on the market only provide basic data recording, lack professional examination path suggestions for endocrine diseases, cannot identify the correlation between symptoms and examination items, and lack intelligent examination guidance;

[0006] Inconsistent testing standards: Different medical institutions have different testing protocols for the same disease (such as thyroid dysfunction), resulting in 22.7% of patients needing to undergo repeated testing (China National Health Commission 2023 Survey Report).

[0007] Patient cognitive blind spot: 83% of diabetic patients fail to correctly identify the typical symptoms of "three highs and one low" in the early stages (IDF Global Diabetes Report). Current technology lacks intelligent matching between symptom self-assessment and examination recommendations. Summary of the Invention

[0008] To address the technical problems existing in the prior art, the present invention provides the following technical solution:

[0009] On the one hand, an app system for guiding the examination of endocrine-related diseases is provided, the system comprising:

[0010] The symptom input module is used by patients to input endocrine symptom information and transmits the endocrine disease characteristics therein to the endocrine disease knowledge graph.

[0011] An endocrine disease knowledge graph engine is used to retrieve and analyze disease symptoms and examination indicators that match the patient's endocrine disease characteristics through a pre-deployed endocrine disease knowledge graph.

[0012] The examination item recommendation engine is used to generate corresponding examination items based on the disease symptoms and examination indicators and recommend them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and forward them to the medical institution adaptation module along with their location information.

[0013] The medical institution adaptation module is used to generate suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommend them to the app front end.

[0014] The result prediction module is used to assess the risk of endocrine test results reported by the app front-end according to pre-set endocrine disease risk assessment rules, output the corresponding risk report and recommend it to the app front-end.

[0015] Preferably, the symptom input module is further configured to:

[0016] Based on a pre-deployed NLP model, the endocrine symptom information is parsed and endocrine disease features related to endocrine-related diseases are extracted. The extracted endocrine disease features are then transferred to an endocrine disease knowledge graph.

[0017] Preferably, the symptom input module is further configured to:

[0018] If the endocrine symptom information input by the patient includes an endocrine medical image, then the image features in the endocrine medical image are identified by a pre-trained convolutional neural network model, and a symptom feature description of the image features is output.

[0019] The symptom descriptions are then transmitted to an endocrine disease knowledge graph.

[0020] Preferably, the endocrine disease knowledge graph is constructed from several sets of knowledge relationships between [endocrine disease characteristics - disease symptoms - examination indicators]; wherein:

[0021] The examination indicators include various endocrine examination items and their costs.

[0022] Preferably, the examination item recommendation engine is equipped with an endocrine examination item recommendation model, which is used to recommend corresponding examination items based on the disease symptoms and examination indicators;

[0023] The recommended model for endocrine examination items was generated by training a random forest (RF) model on data from several endocrine cases.

[0024] Preferably, the medical institution adaptation module dynamically stores the endocrine examination items configured by medical institutions in various regions, their examination fees, and medical insurance policies;

[0025] After receiving the examination items reported by the app front-end, the app matches the corresponding medical institutions based on the patient's selection information for the examination items (examination items and cost budget, whether or not they have medical insurance).

[0026] Output the various compatible medical institutions and recommend them to the app front end, where the patient can confirm and select one medical institution as the endocrine examination institution;

[0027] The app will recommend information about the selected endocrine testing institution to the front end.

[0028] On the other hand, a method for guiding the examination of endocrine-related diseases is provided, implemented based on the aforementioned endocrine-related disease examination guidance app system, the method comprising:

[0029] Patients log in to the app's backend to create their medical records;

[0030] Through the symptom input module, endocrine symptom information is input, and the endocrine disease characteristics are transmitted to the endocrine disease knowledge graph.

[0031] The endocrine disease knowledge graph engine retrieves and analyzes disease symptoms and examination indicators that match the patient's endocrine disease characteristics through a pre-deployed endocrine disease knowledge graph.

[0032] The examination item recommendation engine generates corresponding examination items based on the disease symptoms and examination indicators and recommends them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and the results, along with their location information, are forwarded to the medical institution adaptation module.

[0033] The medical institution matching module generates suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommends them to the app front end.

[0034] The results prediction module assesses the risk of endocrine test results reported by the app frontend based on pre-set endocrine disease risk assessment rules, outputs the corresponding risk report, and recommends it to the app frontend.

[0035] On the other hand, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions, which, when executed by the processor, implement the above-described method for guiding the examination of endocrine-related diseases.

[0036] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the at least one instruction being loaded and executed by a processor to implement the above-described method for guiding the examination of endocrine-related diseases.

[0037] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0038] Based on the analysis of an endocrine disease knowledge graph, this invention recommends more targeted examinations that accurately cover the endocrine diseases a patient may have, thereby improving the accuracy of disease diagnosis, reducing unnecessary examinations, and lowering the patient's medical expenses.

[0039] Once the disease symptoms and examination indicators are obtained, the corresponding examination items are retrieved from the database and compiled into a list for display on the app's front end. Interactive buttons for selection and confirmation are designed on the front end for easy patient operation. Simultaneously, the patient's location information is obtained by utilizing the phone's location function and packaged together with the selected examination items, then sent to the medical institution's adaptation module.

[0040] Recommending suitable endocrine testing institutions based on patient location information makes it easier for patients to choose institutions that are close to them and professionally suitable, solving the problem of finding suitable testing institutions and improving the patient's medical experience.

[0041] By conducting risk assessments and generating reports based on examination results, patients can understand their health status and potential disease risks in a timely manner. This helps patients take further measures in a timely manner, such as follow-up visits and lifestyle adjustments, which is of positive significance for the early detection and intervention of diseases. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the composition architecture of an endocrine-related disease examination guidance app system provided in an embodiment of the present invention;

[0044] Figure 2This is a schematic diagram of the input recognition components of a symptom input module provided in an embodiment of the present invention;

[0045] Figure 3 This is a training diagram of an endocrine examination item recommendation model provided in an embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0047] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0048] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0049] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0050] In this embodiment of the invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0051] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0052] This invention provides an app system and method for guiding the examination of endocrine-related diseases. The method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The diagram shown illustrates the functional architecture of an endocrine-related disease examination guidance app system. The system includes:

[0053] The symptom input module is used by patients to input endocrine symptom information and transmits the endocrine disease characteristics therein to the endocrine disease knowledge graph.

[0054] An endocrine disease knowledge graph engine is used to retrieve and analyze disease symptoms and examination indicators that match the patient's endocrine disease characteristics through a pre-deployed endocrine disease knowledge graph.

[0055] The examination item recommendation engine is used to generate corresponding examination items based on the disease symptoms and examination indicators and recommend them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and forward them to the medical institution adaptation module along with their location information.

[0056] The medical institution adaptation module is used to generate suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommend them to the app front end.

[0057] The result prediction module is used to assess the risk of endocrine test results reported by the app front-end according to pre-set endocrine disease risk assessment rules, output the corresponding risk report and recommend it to the app front-end.

[0058] The working principles of each APP functional module will be described below.

[0059] Symptom Input Module: Patients input their endocrine symptoms in this module. The system extracts features related to endocrine diseases from this information and then passes these features to the endocrine disease knowledge graph engine. This module serves as the entry point for the system to obtain clues about the patient's disease, laying the foundation for subsequent disease analysis and examination recommendations. For example, a dedicated symptom input area can be designed in the app interface, allowing patients to describe their symptoms in detail using text boxes. Simultaneously, natural language processing technology is used to parse the patient's input text and extract key endocrine disease features. For instance, for the input "I've been feeling very thirsty lately, and I've lost weight," semantic analysis extracts features such as "thirst" and "weight loss" that may be related to endocrine diseases. Through patient self-input of symptoms, the system quickly analyzes and recommends targeted examinations, avoiding patients blindly choosing tests, saving them time spent queuing and selecting examinations at the hospital, reducing the workload of medical staff, and improving overall medical efficiency.

[0060] Endocrine Disease Knowledge Graph Engine: An endocrine disease knowledge graph is pre-deployed. Upon receiving endocrine disease characteristics from the symptom input module, it searches and analyzes the knowledge graph. By leveraging existing relationships between diseases, symptoms, and examination indicators within the knowledge graph, it identifies disease symptoms and corresponding examination indicators that match the patient's endocrine disease characteristics, providing a basis for recommending appropriate examinations. First, an endocrine disease knowledge graph is constructed by collecting a large amount of medical data and clinical cases related to endocrine diseases, clarifying the relationships between diseases, symptoms, and examination indicators, and storing this data in a graph database format, such as using Neo4j. When feature information is received, a query language for the graph database (such as Cypher) is used for retrieval and analysis to find matching disease symptoms and examination indicators. Based on the analysis of the endocrine disease knowledge graph, the recommended examinations are more targeted, accurately covering the necessary examinations for the patient's potential endocrine diseases, improving the accuracy of disease diagnosis, reducing unnecessary examinations, and lowering the patient's medical costs.

[0061] The test recommendation engine generates specific test items based on disease symptoms and test indicators provided by the endocrine disease knowledge graph engine. These items are targeted tests recommended for endocrine diseases that the patient may have. These test items are then displayed to the patient on the app's front end. The patient selects and confirms the test items on the app's front end, and the system simultaneously obtains the patient's location information, forwarding the selected test items and location information to the medical institution adaptation module. Specifically, this can be achieved by establishing a test item database to store test items corresponding to different endocrine diseases. After obtaining disease symptoms and test indicators, the system queries the database for corresponding test items and organizes them into a list for display on the app's front end. Interactive buttons for selection and confirmation are designed on the front end for easy patient operation. Simultaneously, the system uses the phone's location function to obtain the patient's location information and packages it along with the selected test items, sending it to the medical institution adaptation module.

[0062] The medical institution matching module uses algorithms or databases to filter suitable medical institutions for the patient's endocrine examination based on the received examination items and the patient's location information. These institutions are likely selected based on a combination of factors, including proximity, institution specialization, and suitability for the examination items. The matched endocrine examination institutions are then recommended to the app for easy patient selection. Specifically, this can be achieved by collecting information on medical institutions within the region, including their areas of expertise and geographical location, to build a medical institution database. Based on the received examination items and location information, Geographic Information System (GIS) technology is used to calculate the nearest medical institutions to the patient, and then, combined with each institution's areas of expertise, suitable endocrine examination institutions are selected. This information, such as name, address, and contact number, is displayed on the app. Recommending suitable endocrine examination institutions based on the patient's location information allows patients to choose a nearby and professionally suitable institution, alleviating the difficulty of finding a suitable examination institution and improving the patient's medical experience.

[0063] The results prediction module: After the patient completes the examination and submits the endocrine test results through the app, this module analyzes and evaluates the results according to pre-set endocrine disease risk assessment rules. These rules may be based on medical research and clinical experience. By interpreting various data points from the test results, it outputs a corresponding risk report, informing the patient of the potential disease risk level and recommending this risk report to the app for the patient to view. This module can formulate detailed endocrine disease risk assessment rules; for example, for blood glucose test results, different risk levels can be set according to different blood glucose value ranges. These rules are written into code logic, and when the test results are received, the module judges and analyzes various data points according to the rules to generate a risk report. A dedicated report display page is designed on the app front end to present the risk report to the patient. By assessing the risk of test results and generating reports, patients can understand their health status and potential disease risks immediately, which helps them take further measures in a timely manner, such as follow-up visits and lifestyle adjustments. This is of positive significance for the early detection and intervention of diseases.

[0064] like Figure 2 The symptom recognition system shown.

[0065] Preferably, the symptom input module is further configured to:

[0066] Based on a pre-deployed NLP model, the endocrine symptom information is parsed and endocrine disease features related to endocrine-related diseases are extracted. The extracted endocrine disease features are then transferred to an endocrine disease knowledge graph.

[0067] Preferably, the symptom input module is further configured to:

[0068] If the endocrine symptom information input by the patient includes an endocrine medical image, then the image features in the endocrine medical image are identified by a pre-trained convolutional neural network model, and a symptom feature description of the image features is output.

[0069] The symptom descriptions are then transmitted to an endocrine disease knowledge graph.

[0070] The symptom input module works as follows:

[0071] 1. Text information processing:

[0072] Based on a pre-deployed Natural Language Processing (NLP) model, this model has the ability to understand and analyze text. When it receives endocrine symptom information (in text form) from a patient, the model parses the text word by word and sentence by sentence according to the language patterns and rules it has learned during training.

[0073] During the analysis process, based on the knowledge system of endocrinology, the model focuses on extracting content describing endocrine-related diseases, which is defined as endocrine disease features. For example, if the text mentions "frequent fatigue, accompanied by sudden weight gain and excessive sweating," the model will identify features such as "fatigue," "sudden weight gain," and "excessive sweating" that may be related to endocrine diseases.

[0074] After extraction, these endocrine disease features are transferred to an endocrine disease knowledge graph. A knowledge graph is a structured knowledge base that graphically displays the relationships between various endocrine diseases, symptoms, diagnostic methods, and other information. Transferring these features allows for further matching, analysis, and reasoning within the knowledge graph.

[0075] 2. Image information processing:

[0076] When the endocrine symptom information input by the patient includes endocrine medical images, a pre-trained convolutional neural network (CNN) model begins to function. CNN models are powerful in image processing; by learning from a large number of endocrine medical images and their corresponding annotations, they are able to identify various features in the images.

[0077] Specifically, CNN models perform a series of operations such as convolution and pooling on input endocrine medical images to extract key features from the images, such as specific cell morphology and tissue texture, and then transform these image features into symptom descriptions of those features. For example, for a thyroid ultrasound image, a CNN model might output a symptom description such as "the thyroid nodule has clear boundaries and uniform internal echoes."

[0078] Similarly, the obtained symptom descriptions are transferred to an endocrine disease knowledge graph for integration and analysis with other relevant information.

[0079] When implementing this module, the following system architecture can be built: Establish a system architecture that includes a data receiving module, a text processing module, an image recognition module, and a knowledge graph interaction module:

[0080] The data receiving module is responsible for receiving endocrine symptom information input by the patient, including text and possible medical images.

[0081] The text processing module integrates a pre-deployed NLP model to ensure stable model operation and effective interface with the data receiving module for real-time text information acquisition and processing. Similarly, the image recognition module integrates a pre-trained CNN model, ensuring seamless integration with the data receiving module to promptly acquire and recognize medical images.

[0082] The knowledge graph interaction module is responsible for transmitting the endocrine disease features extracted by the text processing module and the symptom feature descriptions generated by the image recognition module to the endocrine disease knowledge graph, and can obtain feedback information from the knowledge graph.

[0083] Data preprocessing:

[0084] The received text information is cleaned and formatted as necessary, such as removing irrelevant spaces and special characters, and standardizing text encoding, in order to improve the efficiency and accuracy of NLP model processing.

[0085] For medical images, preprocessing operations such as image resizing and grayscale conversion (if necessary) are performed according to the requirements of the CNN model to ensure that the image meets the format requirements of the model input.

[0086] Model Deployment and Optimization:

[0087] Deploy pre-deployed NLP models and pre-trained CNN models to a suitable server environment, and allocate computing resources such as CPU and GPU reasonably according to the actual business volume and system performance requirements to ensure the efficient operation of the models.

[0088] Regularly optimize the model by collecting more textual data on endocrine symptoms and medical image data, retraining and fine-tuning the model to improve its accuracy in extracting endocrine disease features and recognizing images. For example, fine-tune the NLP model using newly emerging endocrine disease cases to better identify new disease features; train the CNN model with new medical image samples to improve its ability to recognize different types of endocrine medical images. Establish a data backup mechanism to regularly back up patient-input endocrine symptom information, as well as intermediate data and results during model processing, to prevent data loss. Strengthen system security by employing firewalls, data encryption, and other technologies to protect patient data privacy and security, preventing data leakage and malicious attacks. For example, encrypt the descriptions of endocrine disease features and symptoms transmitted to the knowledge graph to ensure data security during transmission.

[0089] Automated text and image feature extraction and processing significantly reduces the time doctors spend manually analyzing endocrine symptom information. For example, previously, doctors might have spent 10-15 minutes reading and analyzing patient symptom descriptions and medical images in detail. Now, the system can quickly extract key features and transmit them to the knowledge graph in just 1-2 minutes. Doctors can then directly make diagnoses based on the knowledge graph's analysis results, reducing overall diagnosis time by approximately 60%-80%. NLP and CNN models, trained on large amounts of data, have high accuracy. NLP models accurately extract endocrine disease features, avoiding the possibility of missing important information during manual extraction. CNN models also provide relatively accurate symptom feature descriptions for medical images. The combination of both makes the information transmitted to the knowledge graph more comprehensive and accurate. For example, in thyroid disease diagnosis, the system-assisted diagnosis accuracy can be improved by 10%-15% compared to manual diagnosis alone, reducing the probability of misdiagnosis and missed diagnosis.

[0090] The extracted features are transferred to an endocrine disease knowledge graph, enabling the integration of information related to various endocrine diseases. The knowledge graph can perform correlation analysis based on these features, providing doctors with comprehensive diagnostic suggestions and decision support. For example, when the endocrine disease feature "goiter" is input, the knowledge graph can not only provide common thyroid disease types but also offer related examination suggestions and treatment plans, helping doctors make more scientific and reasonable diagnostic decisions. With the continuous updating of endocrine medicine knowledge and the emergence of new disease types, the system can achieve good scalability and adaptability through continuous optimization of NLP and CNN models and updates to the knowledge graph. For example, when a new subtype of endocrine disease is discovered, by collecting relevant symptom text and medical image data to train the model, and simultaneously adding relevant information about the new disease to the knowledge graph, the system can quickly adapt to new diagnostic needs, maintaining its technological advancement and practicality.

[0091] like Figure 3 As shown, preferably, the endocrine disease knowledge graph is constructed from several sets of knowledge relationships between [endocrine disease characteristics - disease symptoms - examination indicators]; wherein:

[0092] The examination indicators include various endocrine examination items and their costs.

[0093] The following will describe in detail the steps involved in constructing a knowledge graph of endocrine diseases:

[0094] 1. Data Collection

[0095] Characteristics of endocrine disorders:

[0096] Information on the characteristics of various endocrine diseases was collected from authoritative medical books, such as "Practical Internal Medicine" and "Endocrinology," including the pathogenesis, pathophysiological processes, and common inducing factors. For example, for thyroid diseases, descriptions of their pathogenesis related to iodine metabolism and autoimmunity were collected.

[0097] Search medical academic databases, such as PubMed and Wanfang Medical Network, to find the latest research papers on the characteristics of endocrine diseases and extract valuable information. For example, the latest research findings on the genetic characteristics of diabetes.

[0098] Consult with clinical endocrinologists to gain their knowledge and experience regarding the characteristics of endocrine diseases in actual clinical practice. For example, experts' insights into the unique features of rare endocrine diseases.

[0099] 2. Disease symptoms:

[0100] Referring to clinical treatment guidelines, such as the "Guidelines for the Prevention and Treatment of Type 2 Diabetes in China" and the "Guidelines for the Diagnosis and Treatment of Thyroid Nodules and Differentiated Thyroid Cancer," typical symptoms of each endocrine disease have been identified. For example, symptoms of hyperthyroidism include excessive sweating, palpitations, and hand tremors.

[0101] We collect medical records of patients with endocrine disorders from the hospital's electronic medical record system, statistically analyze the frequency and specific manifestations of symptoms for different diseases, and supplement and improve symptom information. For example, we analyze the occurrence of various complications in the medical records of diabetic patients.

[0102] By leveraging patient health management platforms and medical forums, we can understand patients' self-described disease symptoms and obtain symptom information that more closely reflects their actual experiences. For example, we can gather information on atypical endocrine disease symptoms reported by patients.

[0103] 3. Inspection indicators:

[0104] Based on the Medical Laboratory Manual, the specific content of various endocrine examination items under different indicators is clarified, such as the examination items corresponding to indicators such as T3, T4, and TSH in thyroid function tests.

[0105] Collect the fee schedules of laboratory departments in major hospitals to understand the costs of various endocrine tests. Also, pay attention to the differences in fees between hospitals in different regions and at different levels.

[0106] Consult the medical insurance reimbursement catalog and related policy documents to understand which endocrine examination items are covered by medical insurance and the reimbursement ratio, so as to reflect the comprehensive information related to costs in the chart.

[0107] 4. Data processing and cleaning

[0108] Data compilation:

[0109] The collected data on endocrine disease characteristics, symptoms, and examination indicators are categorized and organized according to disease type. For example, thyroid-related characteristics, symptoms, and examination indicators are grouped into one category, while diabetes-related data are grouped into another.

[0110] For each disease category, the data is organized according to the structure of [endocrine disease characteristics - disease symptoms - examination indicators] to ensure the correlation and logical consistency of the data. For example, the genetic characteristics of diabetes, symptoms of polydipsia, polyphagia, and polyuria, and examination indicators such as blood glucose and glycated hemoglobin are organized accordingly.

[0111] Data cleaning:

[0112] Remove duplicate data, such as descriptions of the same endocrine disease characteristics collected from different sources, and retain only one accurate and authoritative piece of information.

[0113] Check the accuracy of the data, verifying and correcting any unclear or contradictory data. For example, if there are inaccurate descriptions of certain symptoms or abnormal test results, correct them by reviewing relevant materials or consulting experts.

[0114] Standardize data formats, such as unifying the language style for describing disease symptoms and standardizing the units of test indicators.

[0115] 5. Knowledge Graph Construction

[0116] Determine nodes and relationships:

[0117] Each endocrine disease is treated as a master node, with its endocrine disease characteristics, symptoms, and examination indicators serving as sub-nodes.

[0118] Define the relationships between nodes: the relationship between endocrine disease characteristics and diseases is "characteristic"; the relationship between diseases and disease symptoms is "symptom manifestation"; and the relationship between diseases and examination indicators is "examination required". For example, the main node of diabetes has a "characteristic" relationship with the genetic characteristics sub-node, a "symptom manifestation" relationship with the polydipsia, polyphagia, and polyuria symptom sub-node, and a "examination required" relationship with the blood glucose examination indicator sub-node.

[0119] 6. Atlas drawing:

[0120] Choose a suitable knowledge graph drawing tool, such as Neo4j or Graphviz.

[0121] Based on the defined nodes and relationships, a knowledge graph of endocrine diseases is created using this tool. The relationships between nodes are displayed graphically, making the knowledge graph more intuitive and easier to understand. For example, in Neo4j, the knowledge system of endocrine diseases can be visualized by creating nodes and relationships.

[0122] 7. Graph Verification and Update

[0123] Experts in the field of endocrinology were invited to review the constructed knowledge graph, checking its accuracy, completeness, and logical consistency. For example, the experts assessed the reasonableness of the correlations between disease characteristics and symptoms within the graph.

[0124] The knowledge graph is applied to practical scenarios such as clinical auxiliary diagnosis and medical education. Its practicality and effectiveness are tested through practice, feedback is collected, and potential problems are identified. For example, in clinical auxiliary diagnosis, it is observed whether the diagnostic accuracy of doctors improves after using the graph, and whether there is any unreasonable guidance.

[0125] Regularly monitor the latest research findings, clinical practice experiences, and changes in medical policies in the medical field, and update the relevant content in the knowledge graph in a timely manner. For example, when new drugs for the treatment of endocrine diseases emerge, update the characteristic information related to the treatment of the disease.

[0126] Based on the issues identified during the knowledge graph validation process, the graph is revised and improved to ensure it maintains high quality and effectiveness. This includes correcting inaccurate symptom descriptions or incorrect relationships between test indicators.

[0127] The verification can be done internally by the hospital.

[0128] like Figure 3 As shown, preferably, the examination item recommendation engine is equipped with an endocrine examination item recommendation model, which is used to recommend corresponding examination items based on the disease symptoms and examination indicators;

[0129] The recommended model for endocrine examination items was generated by training a random forest (RF) model on data from several endocrine cases.

[0130] The model training steps are as follows:

[0131] 1. Data Collection: Collect a large amount of endocrine case data, including detailed disease symptoms, examination indicators, and corresponding completed examinations. Ensure the data covers multiple types of endocrine diseases to guarantee its comprehensiveness and diversity, providing rich samples for model learning.

[0132] 2. Data preprocessing:

[0133] Data cleaning: Check the data for missing values ​​and outliers. For missing values, depending on the data characteristics, imputation methods such as mean imputation, median imputation, or machine learning-based algorithms can be used. For outliers, analyze their causes; if they are data entry errors, correct them; if they are true extreme values, consider their impact on the model and decide whether to retain them.

[0134] Data encoding: For some categorical data, such as disease names and symptom classifications, one-hot encoding or label encoding is used to convert them into numerical data so that the RF model can process them.

[0135] Feature engineering: Extracting meaningful features from raw data, such as quantifying disease symptoms and assigning different values ​​according to the severity of symptoms; normalizing or standardizing examination indicators to bring different indicators to the same order of magnitude and avoid affecting model performance due to large differences in feature scale.

[0136] 3. Model Selection and Initialization: The Random Forest (RF) model was chosen. RF models consist of multiple decision trees, exhibiting good resistance to overfitting and the ability to handle complex data. During model initialization, parameters such as the number of decision trees (n_estimators) and the maximum depth (max_depth) were set. Typically, default parameters were used for initial training, followed by fine-tuning to determine the optimal parameters.

[0137] 4. Model Training: The preprocessed data is divided into training and testing sets, typically in a 70% - 30% or 80% - 20% ratio. The RF model is trained using the training set data. During training, the RF model automatically learns the complex relationships between disease symptoms, examination indicators, and examination items. In the construction of the decision tree, features are split to find the optimal split point, improving the purity of the child nodes and thus constructing a decision tree. Multiple decision trees are built in parallel and ultimately combined to form the Random Forest model.

[0138] 5. Model Evaluation: Evaluate the trained model using test set data. Evaluation metrics can include accuracy, recall, and F1 score. These metrics determine the accuracy and reliability of the model's recommendations for endocrine examinations. If the evaluation results are unsatisfactory, model adjustments are needed, such as readjusting parameters, increasing the amount of data, or further optimizing data preprocessing steps.

[0139] 6. Model Optimization and Tuning: Based on the evaluation results, model parameters were fine-tuned using methods such as grid search and random search. For example, different combinations of parameters, such as the number of decision trees, maximum depth, and minimum number of sample splits, were tried to find the optimal parameter settings for the evaluation metrics. Simultaneously, further feature selection was considered to remove features that did not contribute to improving model performance or even had a negative impact, thereby improving the model's efficiency and accuracy. After multiple optimizations and adjustments, a superior endocrine examination recommendation model was obtained.

[0140] Integrate the trained endocrine test recommendation model into the test recommendation engine. Ensure the model can interact well with other components of the recommendation engine; for example, it should be able to receive disease symptoms and test indicator data from the front end and accurately return recommended tests to the front end for display. Design a concise and efficient interface for receiving and processing input data. The interface should be able to validate the input data to ensure its integrity and correctness. For example, it should require the input disease symptoms and test indicator data to conform to a predefined format, and return an error message if the format is incorrect.

[0141] When a user inputs disease symptoms and examination indicators into the system, the recommendation engine quickly invokes the endocrine examination recommendation model to perform calculations and returns a list of recommended examinations within a short time. To improve the real-time performance of recommendations, an asynchronous processing mechanism can be adopted to avoid excessive user waiting time due to long model calculation times. Simultaneously, model caching optimization can be implemented; for some common input combinations, recommendation results can be directly retrieved from the cache, reducing the time spent on repetitive model calculations.

[0142] Establish a user feedback channel to collect user evaluations of the recommendation results. If users find the recommended checks unreasonable or inaccurate, they can report the issue to the system administrator through the feedback function. Based on user feedback, the system administrator will further optimize and adjust the model to continuously improve the quality of recommendations.

[0143] Random forest (RF) models, based on the ensemble of multiple decision trees, can comprehensively consider the complex relationships between various factors. By learning from a large amount of endocrine case data, it can capture the potential patterns between disease symptoms, examination indicators, and examination items, thus providing more accurate recommendations for examination items. This helps doctors to understand patients' conditions more comprehensively and accurately, improving diagnostic accuracy. Because RF models consist of multiple decision trees, each constructed using randomly selected samples and feature subsets, there is a certain degree of variability between each tree. This randomness reduces the risk of overfitting in individual decision trees, giving the entire model good generalization ability to new data, maintaining stable performance even when the data contains noise or features are correlated. RF models do not have strict requirements on the distribution and feature type of the data, and do not require complex transformations to satisfy specific distribution assumptions. They can handle various types of features, including numerical and categorical features, and are relatively simple in the data preprocessing stage, reducing the difficulty and workload of data processing, while also effectively utilizing various information in the data for model training.

[0144] Endocrine case data may contain a large number of disease symptoms and examination indicators, making it high-dimensional data. When processing high-dimensional data, the random forest model can automatically select features, identifying those that have a significant impact on the recommendation results. This avoids the curse of dimensionality caused by too many features, while fully utilizing the information in the high-dimensional data to improve the accuracy and comprehensiveness of the recommendations.

[0145] Preferably, the medical institution adaptation module dynamically stores the endocrine examination items configured by medical institutions in various regions, their examination fees, and medical insurance policies;

[0146] After receiving the examination items reported by the app front-end, the app matches the corresponding medical institutions based on the patient's selection information for the examination items (examination items and cost budget, whether or not they have medical insurance).

[0147] Output the various compatible medical institutions and recommend them to the app front end, where the patient can confirm and select one medical institution as the endocrine examination institution;

[0148] The app will recommend information about the selected endocrine testing institution to the front end.

[0149] The medical institution adaptation module dynamically stores (a dynamic database) information on endocrine examination items, examination fees, and medical insurance policies configured by medical institutions in various regions. This data is saved in a dynamically updatable manner to reflect the latest information of medical institutions at any time.

[0150] The medical institution matching module receives endocrine examination information reported from the app's front end, and also obtains the patient's selection information, including specific examination items, set cost budget, and whether they have medical insurance. Based on the received patient selection information, it filters and matches medical institutions from stored data across various regions. For patients with medical insurance, priority is given to medical institutions that comply with medical insurance policies and are within their budget; for patients without medical insurance, the matching is mainly based on their cost budget and the examination items provided. Information on successfully matched medical institutions is output and recommended to the patient on the app's front end. The patient confirms and selects one of these recommended medical institutions as their endocrine examination institution. After the patient completes their selection, the module again recommends detailed information about the selected endocrine examination institution to the app's front end for the patient's further understanding and operation.

[0151] In practical implementation, data connection channels can be established with medical institutions across various regions to regularly collect information such as endocrine examination items, costs, and medical insurance policies. Simultaneously, data interfaces should be set up for manual data entry or modification to handle temporary adjustments. A relational database (such as MySQL) should be used to store the data, with a well-designed table structure, including tables for medical institutions (containing basic institution information), examination items (linked to the medical institution table, recording examination items for each institution), costs (costs for corresponding examination items), and medical insurance policies (explanations of medical insurance policies for different medical institutions and items), ensuring structured data storage and efficient querying. A data update mechanism should be established to regularly (e.g., weekly or monthly) confirm data accuracy with medical institutions, updating the database promptly if changes occur. A real-time update interface should also be provided so that medical institutions can promptly modify their data in emergency situations. An API interface for interaction with the app frontend should be designed to ensure accurate reception of examination items and patient selection information reported by the frontend, while accurately pushing adaptation results and institution information to the frontend. For example, a RESTful API design style should be adopted to ensure the interface's universality and ease of use.

[0152] During data transmission, encryption technologies (such as SSL / TLS protocols) can be used to encrypt sensitive information (such as fees and medical insurance information) to prevent data leakage. Simultaneously, the transmitted data can be compressed to reduce the amount of data transmitted and improve transmission efficiency.

[0153] When making recommendations, different matching rules can be established based on whether the patient has medical insurance. For patients with medical insurance, medical institutions that comply with the insurance policy are first screened out, and then further matched among these institutions based on the budget and examination items. For patients without medical insurance, matching is directly based on the budget and examination items. A matching algorithm is written using a programming language (such as Python) to match and filter according to the established rules by querying relevant information in the database. Reasonable priorities are set in the algorithm; for example, when multiple medical institutions meet the criteria, institutions that are closer to the patient (assuming the app front-end can obtain the patient's location information) or have a better reputation (obtained through third-party review platform data) are prioritized.

[0154] Therefore, patients no longer need to manually select suitable endocrine testing institutions from numerous medical facilities. A simple operation via the app provides a list of compatible institutions, saving significant time and effort. This one-stop service model improves the convenience of medical care, allowing patients to focus more on their health. Based on the patient's specific needs, such as testing requirements, budget, and medical insurance coverage, the app provides precise and personalized medical institution recommendations. Compared to traditional uniform recommendations, this approach better meets the diverse needs of different patients and increases patient satisfaction with the recommendations.

[0155] The system can dynamically store relevant information about medical institutions, reflecting timely changes in their services, costs, and medical insurance policies. This ensures patients always receive up-to-date information, avoiding inconvenience caused by outdated information, such as discrepancies in pricing or inapplicability to medical insurance policies. Through a reasonable matching algorithm, patients can be guided to suitable medical institutions, preventing over-concentration at a few popular institutions and promoting a more balanced distribution of medical resources, thus improving the overall efficiency of medical resource utilization. Simultaneously, it also helps increase patient volume at less popular but reliable medical institutions, promoting the overall development of the healthcare industry.

[0156] On the other hand, a method for guiding the examination of endocrine-related diseases is provided, implemented based on the aforementioned endocrine-related disease examination guidance app system, the method comprising:

[0157] Patients log in to the app's backend to create their medical records;

[0158] Through the symptom input module, endocrine symptom information is input, and the endocrine disease characteristics are transmitted to the endocrine disease knowledge graph.

[0159] The endocrine disease knowledge graph engine retrieves and analyzes disease symptoms and examination indicators that match the patient's endocrine disease characteristics through a pre-deployed endocrine disease knowledge graph.

[0160] The examination item recommendation engine generates corresponding examination items based on the disease symptoms and examination indicators and recommends them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and the results, along with their location information, are forwarded to the medical institution adaptation module.

[0161] The medical institution matching module generates suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommends them to the app front end.

[0162] The results prediction module assesses the risk of endocrine test results reported by the app frontend based on pre-set endocrine disease risk assessment rules, outputs the corresponding risk report, and recommends it to the app frontend.

[0163] Please understand and implement the above methods and steps in conjunction with the working principles of the system described above.

[0164] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 4 As shown, optionally, electronic device 410 may include a first processor 2001.

[0165] Optionally, the electronic device 410 may also include a memory 2002 and a transceiver 2003.

[0166] The first processor 2001, memory 2002, and transceiver 2003 can be connected via a communication bus.

[0167] The following is combined Figure 4 A detailed description of each component of electronic device 410 is provided below:

[0168] The first processor 2001 is the control center of the electronic device 410. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0169] Optionally, the first processor 2001 can perform various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0170] In a specific implementation, as one example, the first processor 2001 may include one or more CPUs, for example... Figure 4 CPU0 and CPU1 are shown in the diagram.

[0171] In a specific implementation, as one example, the electronic device 410 may also include multiple processors, for example... Figure 4 The first processor 2001 and the second processor 2004 are shown in the diagram. Each of these processors can be a single-core processor or a multi-core processor. Here, a processor can refer to one or more devices, circuits, and / or processing cores used to process data (such as computer program instructions).

[0172] The memory 2002 is used to store the software program that executes the present invention, and is controlled by the first processor 2001 to execute it. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0173] Optionally, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 2002 may be integrated with the first processor 2001 or may exist independently and be connected via the interface circuit of the electronic device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0174] The transceiver 2003 is used to communicate with network devices or with terminal devices.

[0175] Alternatively, transceiver 2003 may include a receiver and a transmitter. Figure 4 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function.

[0176] Optionally, the transceiver 2003 can be integrated with the first processor 2001, or it can exist independently and be connected via the interface circuit of the electronic device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0177] It should be noted that, Figure 4 The structure of the electronic device 410 shown does not constitute a limitation on the router. Actual knowledge structure identification devices may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0178] Furthermore, the technical effects of the electronic device 410 can be referred to the technical effects of the endocrine-related disease examination guidance app system and method described in the above method embodiments, and will not be repeated here.

[0179] It should be understood that the first processor 2001 in the embodiments of the present invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0180] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0181] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0182] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0183] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0184] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0185] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0186] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0187] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0188] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0189] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0190] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0191] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An app system for guiding the examination of endocrine-related diseases, characterized in that, The system includes: The symptom input module is used by patients to input endocrine symptom information and transmits the endocrine disease characteristics therein to the endocrine disease knowledge graph. An endocrine disease knowledge graph engine is used to retrieve and analyze disease symptoms and examination indicators that match the patient's endocrine disease characteristics through a pre-deployed endocrine disease knowledge graph. The examination item recommendation engine is used to generate corresponding examination items based on the disease symptoms and examination indicators and recommend them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and forward them along with their location information to the medical institution adaptation module. The medical institution adaptation module is used to generate suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommend them to the app front end. The result prediction module is used to assess the risk of endocrine test results reported by the app front-end according to pre-set endocrine disease risk assessment rules, output the corresponding risk report and recommend it to the app front-end.

2. The endocrine-related disease examination guidance app system and method according to claim 1, characterized in that, The symptom input module is also used for: Based on a pre-deployed NLP model, the endocrine symptom information is parsed and endocrine disease features related to endocrine-related diseases are extracted. The extracted endocrine disease features are then transferred to an endocrine disease knowledge graph.

3. The endocrine-related disease examination guidance app system and method according to claim 1, characterized in that, The symptom input module is also used for: If the endocrine symptom information input by the patient includes an endocrine medical image, then the image features in the endocrine medical image are identified by a pre-trained convolutional neural network model, and a symptom feature description of the image features is output. The symptom descriptions are then transmitted to an endocrine disease knowledge graph.

4. The endocrine-related disease examination guidance app system and method according to claim 1, characterized in that, The knowledge graph of endocrine diseases is constructed from several sets of knowledge relationships between [endocrine disease characteristics - disease symptoms - examination indicators]; wherein: The examination indicators include various endocrine examination items and their costs.

5. The endocrine-related disease examination guidance app system and method according to claim 1, characterized in that, The examination item recommendation engine includes an endocrine examination item recommendation model, which recommends corresponding examination items based on the disease symptoms and examination indicators. The recommended model for endocrine examination items was generated by training a random forest (RF) model on data from several endocrine cases.

6. The endocrine-related disease examination guidance app system and method according to claim 1, characterized in that, The medical institution adaptation module dynamically stores the endocrine examination items, examination fees, and medical insurance policies configured by medical institutions in various regions. After receiving the examination items reported by the app front-end, the app matches the corresponding medical institutions based on the patient's selection information for the examination items (examination items and cost budget, whether or not they have medical insurance). Output the various compatible medical institutions and recommend them to the app front end, where the patient can confirm and select one medical institution as the endocrine examination institution; The app will recommend information about the selected endocrine testing institution to the front end.

7. A method for guiding the examination of endocrine-related diseases, implemented based on the endocrine-related disease examination guidance app system according to any one of claims 1-6, characterized in that, The method includes: Patients log in to the app's backend to create their medical records; Through the symptom input module, endocrine symptom information is input, and the endocrine disease characteristics are transmitted to the endocrine disease knowledge graph. The endocrine disease knowledge graph engine uses a pre-deployed endocrine disease knowledge graph to retrieve and analyze disease symptoms and examination indicators that match the patient's endocrine disease characteristics. The examination item recommendation engine generates corresponding examination items based on the disease symptoms and examination indicators and recommends them to the front end of the app. Patients select and confirm the corresponding examination items through the front end of the app, and the results, along with their location information, are forwarded to the medical institution adaptation module. The medical institution matching module generates suitable endocrine examination institutions for the patient based on the examination items and the patient location information reported by the app front end, and recommends them to the app front end. The results prediction module assesses the risk of endocrine test results reported by the app frontend based on pre-set endocrine disease risk assessment rules, outputs the corresponding risk report, and recommends it to the app frontend.

8. An electronic device, characterized in that, The electronic device includes: processor; A memory storing computer-readable instructions that, when executed by the processor, implement the method as described in claim 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in claim 7.

Images