Bone joint disease artificial intelligence auxiliary diagnosis and treatment system based on Internet hospital

Abstract

The invention relates to the technical field of osteoarticular diseases, in particular to an artificial intelligence auxiliary diagnosis and treatment system for osteoarticular diseases based on an internet hospital. Comprising an information management module; an intelligent diagnosis auxiliary module; a remote monitoring and rehabilitation guidance module; a disease risk prediction module; according to the system, a comprehensive system covering functions of information acquisition, data integration, image analysis, rehabilitation tracking and the like is constructed, so that whole-process information of a patient from a doctor seeing period to a rehabilitation period is systematically managed, and the continuity and personalized service level of diagnosis and treatment are effectively improved; and artificial intelligence technologies such as deep learning, natural language processing and knowledge graph are applied to automatically complete bone joint image recognition, symptom analysis and rehabilitation plan making, so that the diagnosis efficiency and accuracy of doctors are greatly improved, and meanwhile, dynamic adjustment and refined guidance of the rehabilitation plan are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bone and joint diseases, and in particular to an artificial intelligence assisted diagnosis and treatment system for bone and joint diseases based on an Internet hospital. BACKGROUND

[0002] Bone and joint diseases are common chronic disease types in the elderly population, common disease types include osteoarthritis, osteoporosis, synovitis, rheumatoid arthritis, etc., with high incidence, chronic progression and high disability rate. With the intensification of population aging trend, the demand for bone and joint disease treatment continues to grow, which brings great pressure to medical resource allocation and diagnosis and treatment efficiency.

[0003] At present, the diagnosis and rehabilitation management of bone and joint diseases in clinic mainly rely on offline hospitals, doctors make diagnosis and judgment through patient complaints, physical examination, image data, and give rehabilitation guidance combined with individual conditions. This traditional mode has the following problems: uneven distribution of medical resources, lack of professional orthopedic diagnosis and treatment ability in primary medical institutions; image interpretation relies on manual work, which is labor-intensive and easily affected by the experience level of doctors, patients cannot receive long-term and continuous rehabilitation guidance, leading to poor rehabilitation compliance, poor effect, lack of active identification and intervention for high-risk groups, insufficient disease prevention ability, limited doctor-patient interaction channels, single health education means, and difficult to improve patient self-management awareness.

[0004] In summary, the development of an artificial intelligence assisted diagnosis and treatment system for bone and joint diseases based on an Internet hospital is still a key problem that needs to be solved in the technical field of bone and joint diseases. SUMMARY

[0005] The purpose of the present application is to solve the problems in the prior art that primary medical institutions lack professional orthopedic diagnosis and treatment ability, image interpretation relies on manual work, which is labor-intensive and easily affected by the experience level of doctors, patients cannot receive long-term and continuous rehabilitation guidance, leading to poor rehabilitation compliance, poor effect, lack of active identification and intervention for high-risk groups, insufficient disease prevention ability, and limited doctor-patient interaction channels.

[0006] To achieve the above-mentioned purpose, the present application provides an artificial intelligence assisted diagnosis and treatment system for bone and joint diseases based on an Internet hospital, comprising: an information management module and an intelligent diagnosis assistance module; The information management module is used to obtain bone texture fractal dimension, mid-joint space width dynamic change rate, disease symptoms and medical knowledge graph; The intelligent diagnosis assistance module is used to calculate the disease of , for assisting diagnosis, specifically: Based on the bone texture fractal dimension and the intertarsal joint gap width dynamic change rate, the medical image is analyzed through a deep learning model to obtain the prediction probability of the disease ; Based on the symptoms and the medical knowledge graph, the matching score of the disease is calculated to obtain ; The current symptoms and the disease are calculated , Among them, Indicates the prediction probability of the disease obtained by analyzing the medical image through the deep learning model, Indicates the disease matching score calculated based on the symptom semantic similarity and the knowledge graph structure, and a and b are weight coefficients. According to , the disease is sorted, and a diagnosis basis is automatically generated for each candidate disease.

[0007] Advantages Compared with the known prior art, the technical scheme provided by the present application has the following advantages: The present application constructs a comprehensive system covering information collection, data integration, image analysis, rehabilitation tracking and other functions, systematically manages the whole process information of patients from pre-consultation to rehabilitation period, and effectively improves the continuity of diagnosis and treatment and the individualized service level.

[0008] The present application uses deep learning, natural language processing, knowledge graph and other artificial intelligence technologies to automatically complete bone joint image recognition, symptom analysis and rehabilitation plan formulation, greatly improving the efficiency and accuracy of doctor diagnosis, and realizing dynamic adjustment and fine guidance of rehabilitation plan. DETAILED DESCRIPTION

[0009] Figure 1 The system diagram of the bone joint disease artificial intelligence auxiliary diagnosis and treatment system based on the Internet hospital of the present application. CONCRETE EMBODIMENT

[0010] In order to make the personnel in the technical field better understand the present application scheme, the technical scheme in the embodiment of the present application will be described clearly and completely in combination with the drawings in the embodiment of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the ordinary skilled in the art without creative labor should belong to the scope of protection of the present application.

[0011] ​​​It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0012] The present invention will now be described in further detail with reference to the accompanying drawings: Example: like Figure 1 As shown, the present invention provides an artificial intelligence-assisted diagnosis and treatment system for bone and joint diseases based on an internet hospital, including: an information management module; Furthermore, the operation process of the information management module includes: When patients register at the internet hospital, they enter detailed personal information, past medical history, family medical history, and lifestyle habits. The system connects with the information systems of partner medical institutions through an interface to automatically obtain the patient's past imaging examination and test reports and diagnostic records. The system then structures the obtained data.

[0013] The system uses an automated image segmentation algorithm (U-Net) to identify the edges of the femur and tibia; calculates the minimum distance between the articular surfaces and outputs the JSW value; and automatically calculates ΔJSW using time labels (such as the image acquisition date) and visualizes it.

[0014] Image data feature quantization calculates the dynamic change rate of the joint space width in the image data, using the following formula: , in, This indicates the width of the joint space measured in the current image. This indicates the joint space width measured in the previous imaging. Indicates time interval, This represents the error tolerance for image measurement or AI image segmentation. The dynamic rate of change of the joint space width is used as input to the AI ​​model.

[0015] Select the ROI (Region of Interest), such as the femoral condyle or tibial plateau; use the binarized results of the bone tissue image; automatically solve the fractal dimension based on the box-counting method; output the numerical value as input to the AI ​​model to help determine the disease stage.

[0016] The bone texture fractal dimension is obtained, and a formula of the bone texture fractal dimension is: , wherein, represents a minimum number of cubes with an edge length of required to completely cover the target bone tissue trabecula, represents the edge length of the cube, and tends to be a micro-precision when being approximately 0, represents a fractal dimension of the complexity of the bone tissue structure in the image; Specifically, when the patient registers in the Internet hospital, the patient details the basic information, the past medical history, the diagnosis and treatment experience related to bone and joint diseases, the family medical history, the living habits, the exercise mode and the labor intensity, at the same time, the system is connected with the information system of the cooperative medical institutions through the interface, and the past image examination of the patient, the X-ray, the CT, the MRI, the test report, the diagnosis record and other data are automatically obtained, the collected data are classified and stored by using the structured and unstructured database technology, and the data accuracy and consistency are ensured by means of the data cleaning and standardization processing.

[0017] An intelligent diagnosis assistance module; Further, the operation process of the intelligent diagnosis assistance module comprises: The image intelligent analysis module extracts features and identifies lesions of the bone joint image uploaded by the user based on a convolutional neural network (CNN), accurately locates the joint area through an attention mechanism, automatically identifies the lesion types such as fracture, hyperostosis and joint space narrowing by combining the classification model obtained by training, and measures quantitative indexes such as joint space width, accurately locates the joint area: adopts a UNET network with an attention gate mechanism, and the calculation formula of the attention weight of each pixel point belonging to the joint area is: , wherein, represents a splicing vector of the image feature and the position at the pixel point , represents a trainable attention weight parameter, represents all pixel points for normalization to obtain a probability distribution, represents an integration degree representing the position to the current task of fracture identification or joint contour segmentation, can automatically focus on the image area highly related to the joint lesion, and improves the accuracy of lesion detection and segmentation.

[0018] Through the natural language symptom description, a text is converted into a feature vector model by using a pre-training model such as BERT, and the formula is: , wherein, Indicates the knee joint night dull pain, morning stiffness lasts 60 minutes, Reasoning with medical knowledge graph, calculating the correlation strength between symptoms and diseases, The calculation formula is:

[0019] Among them, the vector of the user's current input symptom, Indicates the vector of the Standard symptom node in the graph, Indicates the semantic similarity between the current symptom and the symptom node In the knowledge graph, Indicates the graph edge weight of symptom To disease , indicating the correlation strength between the two, Indicates the number of all symptom nodes connected to the candidate disease In the graph.

[0020] The calculation formula of the semantic matching strength between the current symptom and the disease node is: , Among them, Indicates the disease probability output by the image, Indicates the final score of the candidate disease , Indicates the prediction probability of the disease obtained by analyzing the medical image through the deep learning model (such as CNN), Indicates the disease matching score calculated based on the symptom semantic similarity and the knowledge graph structure.

[0021] The system sorts the disease list According to the comprehensive score, and automatically generates the diagnosis basis for each candidate disease, including typical image features such as osteoarthritis, joint space narrowing, soft tissue degeneration area location, symptom similarity analysis pain site and disease onset area overlap degree and corresponding clinical differential suggestions, distinguish the morning stiffness duration of osteoarthritis and rheumatoid arthritis, serological indicators, assist doctors in accurate diagnosis and reasonable disposal; Specifically, image intelligent analysis: for the uploaded bone and joint image data, deep learning algorithm is used, convolutional neural network automatically identifies the shape and structural changes of bone and joint, detects fracture, hyperostosis, joint space stenosis and other pathological changes, and generates detailed image analysis report, marks the location, range and degree of the lesion, provides diagnosis reference for doctors, and inputs the current symptom pain site, degree, onset time and accompanying symptoms of the patient, the system uses machine learning algorithm, combines with massive clinical case data and medical knowledge graph, analyzes the correlation between symptoms and diseases, provides possible disease diagnosis list for doctors, and sorts them according to the possibility, and gives corresponding diagnosis basis and differential diagnosis suggestion.

[0022] Remote monitoring and rehabilitation guidance module Rehabilitation equipment connection: 1. The system provides open Internet of Things access capability, supports patient binding and authorization of multiple wearable devices and home intelligent rehabilitation equipment.

[0023] 2. Real-time or scheduled collection of joint range of motion, muscle strength, gait parameters, pain subjective score, physiotherapy equipment usage data, etc., and automatic uploading to the system platform through an encrypted channel.

[0024] Personalized rehabilitation program development: 1. Based on the patient's final diagnosis, current rehabilitation stage, physical function baseline evaluation and personal goals, the system uses rule-based reasoning combined with reinforcement learning RL to automatically generate personalized and stepwise rehabilitation training plans, including specific action illustrations / videos, group number, frequency, intensity / resistance setting and rehabilitation therapy scheme.

[0025] 2. The scheme clearly sets the stage rehabilitation goals and expected progress schedule.

[0026] Rehabilitation process tracking and adjustment: 1. The system continuously receives and analyzes the rehabilitation monitoring data uploaded by the patient, and dynamically compares the actual progress of ROM improvement and pain relief with the preset goals.

[0027] 2. If the algorithm detects that the rehabilitation progress significantly lags behind the expected progress, the key indicators worsen or abnormal patterns appear, the system will automatically trigger an early warning and notify the responsible doctor for remote or offline clinical evaluation.

[0028] 3. After the doctor's evaluation, the system can combine the evaluation conclusion and use feedback-based model predictive control MPC to intelligently adjust the rehabilitation program, including modifying the action difficulty, increasing or decreasing the training amount, and adjusting the physiotherapy parameters.

[0029] To ensure that the patient executes correctly, the system provides multimedia guidance, and can combine wearable device data to provide real-time action correction feedback during patient practice.

[0030] Disease risk prediction module Risk factor analysis: 1. System aggregates multi-dimensional data in patient profile: demographic data, genetic information, detailed lifestyle habits, past medical history, bone density, specific biomarkers.

[0031] 2. Utilize association rule mining Apriori and feature importance analysis algorithm SHAP value to identify significant risk factors related to specific bone and joint disease occurrence and progression, including primary osteoarthritis OA, osteoporosis OP, rheumatoid arthritis RA, including intervenable and non-intervenable risk factors.

[0032] Risk prediction model construction: 1. Based on the historical accumulation of massive de-identified bone and joint disease patient cohort data, train and verify multiple machine learning risk prediction models: ① Logistic regression: for binary classification risk prediction.

[0033] ② Cox proportional hazards model: predict the incidence risk within 5 years, 10 years in the future.

[0034] ③ Gradient Boosting Machine or Deep Learning Model: handle more complex non-linear relationships and high-dimensional features to improve prediction accuracy.

[0035] 2. Model output results are the quantitative risk probability of patients suffering from specific target bone and joint diseases.

[0036] Early warning and prevention recommendations: 1. When the system predicts that the patient is at medium to high risk, automatically generate early warning information, and push it to the patient himself and his contracted / supervising doctor through platform messages, SMS or email.

[0037] 2. Based on the identified dominant risk factors, the system generates personalized and actionable prevention intervention recommendations: ① Lifestyle intervention: such as weight loss goal setting, recommended low-impact exercise, avoidance of specific professional posture / load.

[0038] ② Nutrition supplement recommendations: such as calcium, vitamin D intake guidance.

[0039] ③ Preventive exercise prescription: targeted strengthening of joint surrounding muscle strength, training program to improve joint stability.

[0040] ④ Regular monitoring recommendations: recommend high-risk groups to undergo regular bone density tests or specific joint imaging screening.

[0041] Doctor-patient interaction and health popularization module Online consultation: 1. Provide a secure and confidential online consultation channel for real-time text, voice, high-definition video, and other forms of communication.

[0042] 2. When a doctor receives a patient, the system automatically pushes the patient's integrated health record and the analysis results of the intelligent diagnosis assistance module, including image reports and diagnosis suggestion lists, to help the doctor quickly and comprehensively understand the patient's condition, improve the efficiency and accuracy of online consultation, and provide more accurate treatment recommendations.

[0043] Health popularization: 1. The system maintains a structured knowledge base of bone and joint diseases, including popular articles, short videos, animations, and information graphics from authoritative sources.

[0044] 2. The content covers disease prevention, early identification, diagnosis methods, treatment options, postoperative care, home rehabilitation, nutrition and health care, and psychological adjustment throughout the cycle.

[0045] 3. The application uses content recommendation algorithms based on collaborative filtering or content similarity to recommend systems that can accurately push popular content based on the patient's specific disease diagnosis, current health status, rehabilitation stage, and historical browsing / interaction behavior, improving the patient's disease awareness and self-health management capabilities.

[0046] Patient feedback and evaluation: 1. After completing online consultation, using rehabilitation guidance services, or receiving popular information, patients can evaluate and feedback on the doctor's service attitude, professional level, answer clarity, and system functionality, ease of use, stability, and help through standardized questionnaires or open comments.

[0047] 2. The system collects structured feedback data and uses sentiment analysis technology to mine and analyze the results, which can be used to continuously improve medical service quality, optimize system user experience and functional design.

[0048] The above examples are only used to illustrate the technical solutions of the present application, and not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that they can modify the technical solutions described in the foregoing examples, or make equivalent substitutions for some technical features; and these modifications or substitutions do not change the essence of the corresponding technical solutions beyond the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An Internet hospital-based artificial intelligence-assisted diagnosis and treatment system for bone and joint diseases, characterized in that, Comprise: An information management module and an intelligent diagnosis assistance module; The information management module is used to obtain bone texture fractal dimension, mid-joint space width dynamic change rate, symptoms of diseases and medical knowledge graph; The intelligent diagnosis assistance module is used for calculating diseases of , for assisting in diagnosis, in particular: Based on the fractal dimension of bone texture and the dynamic change rate of the midtarsal joint gap width, the prediction probability of the disease is obtained by analyzing the medical image through a deep learning model . ​ Based on symptom and medical knowledge graphs, calculate diseases The matching score is obtained. ; computing the current symptoms and diseases : ， wherein, represents the prediction probability of the disease obtained by analyzing the medical image through the deep learning model, represents the disease matching score calculated based on the symptom semantic similarity and the knowledge graph structure, and a and b are weight coefficients. According to , the diseases are ranked, and a diagnosis rationale is automatically generated for each candidate disease . 2.The Internet hospital-based osteoarticular disease artificial intelligence auxiliary diagnosis and treatment system according to claim 1, characterized in that, The information management module is used to: Quantify image features and calculate the mid-joint space width dynamic change rate: , wherein, represents the joint space width measured in the current image, represents the joint space width measured in the previous image, represents the time interval, represents the error tolerance of the image measurement or AI image segmentation. 3.The Internet hospital-based osteoarticular disease artificial intelligence auxiliary diagnosis and treatment system according to claim 2, characterized in that, In the information management module: The bone texture fractal dimension judgment formula is: wherein, represents the minimum number of cubes with edge length required to completely cover the target bone tissue trabeculae, represents the edge length of the cube, which tends to 0 when approximating the micro-precision, represents the fractal dimension of the bone tissue structure in the image. 4.The Internet hospital-based osteoarticular disease artificial intelligence auxiliary diagnosis and treatment system according to claim 1, characterized in that, The system further comprises a remote monitoring and rehabilitation guidance module, which is used to: Receive joint range of motion, muscle strength, gait parameters, pain subjective score and physiotherapy equipment use data uploaded from wearable devices or home smart rehabilitation devices bound to patients; Based on the diagnosis results, rehabilitation stages, physical function baseline evaluation and personal goals of patients, a personalized step-by-step rehabilitation training plan is automatically generated by using rule-based reasoning combined with reinforcement learning model, which includes specific action guidance, group number, frequency, intensity setting and stage rehabilitation goals; Continuously analyze the rehabilitation monitoring data uploaded by patients, dynamically compare the actual rehabilitation progress with the preset goals, and automatically trigger an alarm and notify the responsible doctor when the rehabilitation progress lags behind, the key indicators worsen or abnormal patterns appear; Receive the evaluation feedback of the doctor, and dynamically adjust the rehabilitation scheme based on the feedback by using model predictive control algorithm. 5.The Internet hospital-based osteoarticular disease artificial intelligence auxiliary diagnosis and treatment system according to claim 1, characterized in that, The disease risk prediction module of the system is used to: Aggregate multi-dimensional data in the patient file, including demographic data, genetic information, lifestyle, past medical history, bone density and specific biomarkers; Use association rule mining and feature importance analysis algorithm to identify risk factors significantly related to the occurrence and progression of specific bone and joint diseases; Based on the historical de-identified patient cohort data, train and verify the machine learning risk prediction model, which includes at least one of the following: logistic regression, Cox proportional hazards model, gradient boosting machine or deep learning model, to output the quantitative risk probability of patients suffering from specific bone and joint diseases; When the predicted risk reaches the medium-high risk threshold, automatically generate an early warning message and push it to the patient and the doctor, and generate personalized prevention intervention suggestions based on the dominant risk factors. 6.The Internet hospital-based osteoarticular disease artificial intelligence auxiliary diagnosis and treatment system according to claim 1, characterized in that, The doctor-patient interaction and health popularization module of the system is used to: Provide online consultation channels in the form of text, voice or video, and actively push the patient's health record and the analysis results of the intelligent diagnosis assistance module to the doctor during the consultation process; Maintain a structured bone and joint disease knowledge base containing disease prevention, diagnosis, treatment, rehabilitation and health-related popular science content; Based on the specific disease diagnosis, health status, rehabilitation stage and historical behavior of patients, use recommendation algorithms to achieve personalized and accurate push of popular science content; Receive the evaluation and feedback of patients on the consultation service, rehabilitation guidance and popular science content, and use sentiment analysis technology to mine and analyze the feedback data to continuously improve the service quality and system functions.

Images