Rheumatoid arthritis data modeling method and system based on digitization

Abstract

The invention relates to a digitalization-based rheumatoid arthritis data modeling method, and relates to the field of electric digital data processing, and the method comprises the steps: obtaining an intelligent grade identification model corresponding to a current patient; and intelligently identifying the severity level of the rheumatoid arthritis of the current patient according to the first visual data, the second visual data and the multiple related parameters of the current patient by using an intelligent level identification model. The invention further relates to a rheumatoid arthritis data modeling system based on digitization. According to the invention, aiming at the technical problem that misdiagnosis is difficult to avoid due to the fact that various disease condition associated data of the rheumatoid arthritis patient cannot be integrated to execute disease condition identification, normalization processing of the patient associated data is executed on the X-ray visual data, the MR I visual data and other disease condition associated data of the rheumatoid arthritis patient; modeling processing of intelligent grade identification models of different structures of different patients is realized, so that the technical problem is solved.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to a rheumatoid arthritis data modeling method and system based on digitization. Background Art

[0002] Rheumatoid arthritis is a chronic autoimmune disease whose symptoms and severity vary from person to person. To better describe the patient's joint function, the American College of Rheumatology (ACR) Rheumatoid Arthritis Functional Grading Criteria can be used for assessment. This grading standard divides rheumatoid arthritis joint function into four levels:

[0003] Grade 1 (mild functional impairment): Patients are essentially unrestricted in their daily activities and can complete daily activities, but may experience joint pain or fatigue after standing or walking for a long time.

[0004] Grade 2 (moderate functional impairment): The patient's joint function is somewhat affected and he or she can complete daily activities without assistance, but some activities may be limited, such as climbing stairs or lifting heavy objects.

[0005] Grade 3 (severe functional impairment): The patient requires assistance in daily life and is unable to complete activities that require considerable strength or flexibility, such as bathing, dressing, or cooking.

[0006] Level 4 (extremely severe functional impairment): The patient loses the ability to take care of himself, needs to stay in bed for a long time, and is completely dependent on others for care.

[0007] This grading standard helps doctors better understand the patient's condition, develop an appropriate treatment plan, and evaluate the patient's joint function improvement through the effectiveness of treatment. Patients of different levels may require different treatment plans, including medication, physical therapy, surgery, etc.

[0008] For example, Chinese invention patent publication CN110613430A proposes a multimodal photoacoustic / ultrasound imaging rheumatoid arthritis scoring system and its application, comprising the following steps: (1) collecting image information of the joint in vitro using photoacoustic / ultrasound dual-modal imaging; (2) analyzing the collected image information and clinical data to perform multimodal photoacoustic / ultrasound scoring; and (3) judging the disease activity of rheumatoid arthritis patients based on the multimodal photoacoustic / ultrasound scoring and combining it with local blood oxygen information. This invention is the first to use a multimodal photoacoustic / ultrasound imaging system for the evaluation of rheumatoid arthritis. The advantage is that the multimodal system can use a handheld photoacoustic / ultrasound probe, which is in line with the usage habits of clinical physicians. The implementation process and calculation process are both simple and easy, which is conducive to clinical implementation and promotion.

[0009] For example, Chinese invention patent publication CN110148465A proposes a system for analyzing rheumatoid arthritis (RA). The system comprises a patient data acquisition module, a data input module, and an analysis module, enabling timely understanding of the progression of joint damage in RA patients and monitoring the effectiveness of treatment. The proposed MRI scanner utilizes a 3.0T whole-body scanning system with an eight-channel head coil, enhancing the signal-to-noise ratio for clearer images. The average scan time is only 24 minutes, laying a solid foundation for future clinical adoption.

[0010] However, the above-mentioned technical solutions in the existing technology are only limited to the collection of visual data based on a single diagnostic mechanism and the application to rheumatoid arthritis disease analysis. They are unable to integrate visual data of multiple diagnostic mechanisms and other related data to analyze the severity of the patient's rheumatoid arthritis disease. The bottleneck lies in the inability to digitize the visual data and other related data of multiple diagnostic mechanisms, and the lack of a modeling mechanism that can apply multiple related data. As a result, the various technical solutions in the existing technology have insufficient accuracy in identifying the patient's rheumatoid arthritis disease, and are prone to misdiagnosis. At the same time, it is impossible to further enrich the medical care information of rheumatoid arthritis patients on the basis of manual disease identification and machine disease identification of a single diagnostic mechanism. Summary of the Invention

[0011] In order to solve the technical problems in the prior art, the present invention provides a rheumatoid arthritis data modeling method and system based on digitalization. By performing normalization processing on the X-ray visual data, MRI visual data, and the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking mark, drinking mark, age information and gender information of the current patient suffering from rheumatoid arthritis, the modeling processing of the intelligent grade identification model with different structures for different patients is realized, thereby providing an intelligent identification mechanism for the severity level of rheumatoid arthritis of different patients, improving the identification accuracy of the patient's rheumatoid arthritis condition, avoiding misdiagnosis, and at the same time, further enriching the medical care information of rheumatoid arthritis patients on the basis of manual condition identification and machine condition identification of a single diagnostic mechanism.

[0012] According to a first aspect of the present invention, a method for modeling rheumatoid arthritis data based on digitization is provided, the method comprising:

[0013] Input an image block occupied by a joint portion in an X-ray image of a current patient suffering from rheumatoid arthritis and an image block occupied by a joint portion in an MRI image of the current patient, and output them as a first image block and a second image block respectively;

[0014] The brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block are used as the first visual data of the current patient, and the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block are used as the second visual data of the current patient;

[0015] Extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking mark, drinking mark, age information and gender information as multiple correlation parameters of the current patient;

[0016] Obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network has been trained is monotonically positively correlated with the disease course data of the current patient;

[0017] intelligently identifying the severity level of rheumatoid arthritis of the current patient using an intelligent grade identification model based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient;

[0018] The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution.

[0019] According to a second aspect of the present invention, a digitalized rheumatoid arthritis data modeling system is provided, the system comprising:

[0020] a block acquisition mechanism for recording image blocks occupied by joints in an X-ray image of a current patient suffering from rheumatoid arthritis and image blocks occupied by joints in an MRI image of the current patient, and outputting them as first image blocks and second image blocks, respectively;

[0021] a visual recording mechanism connected to the block acquisition mechanism, configured to use the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block as the first visual data of the current patient, and use the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block as the second visual data of the current patient;

[0022] A parameter extraction mechanism is used to extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information and gender information as multiple related parameters of the current patient;

[0023] A multi-layer training mechanism is used to obtain an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network is trained is monotonically positively correlated with the current patient's disease course data;

[0024] a grade identification mechanism, connected to the visual input mechanism, the parameter extraction mechanism, and the multi-layer training mechanism, respectively, for using an intelligent grade identification model to intelligently identify the severity grade of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient;

[0025] The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution.

[0026] It can be seen that the present invention has at least the following main invention points:

[0027] Invention A: Various condition-related data of a current patient suffering from rheumatoid arthritis are digitally processed to obtain various basic information used to analyze the severity level of the current patient's rheumatoid arthritis. Simultaneously, a corresponding intelligent grade identification model is established for the current patient using the current patient's disease course data. The established intelligent grade identification model is then used to identify the current patient's current condition, thereby providing a digital condition analysis mechanism for all patients suffering from rheumatoid arthritis, improving the accuracy of condition analysis while enriching the patient's medical care information.

[0028] Inventive point B: The structures of the intelligent level identification models corresponding to different patients are different. Specifically, the intelligent level identification model for each patient is a BP neural network that has been trained multiple times, and the number of BP neural network training times is monotonically positively correlated with the patient's disease course data. In addition, in each training of the BP neural network, the known severity level of a patient with rheumatoid arthritis is used as the single output content of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple related parameters of the patient are used as the item-by-item output content of the BP neural network to complete the training of the BP neural network. The customized design of different models for different patients ensures the reliability and stability of the intelligent identification results.

[0029] Invention point C: The basic information used to analyze the severity level of rheumatoid arthritis of the current patient is digitally processed information, and the basic information includes the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient. Specifically, the first visual data of the current patient is the brightness gradient value and coordinate value of each pixel point in the image block occupied by the joint part in the X-ray image of the current patient, the second visual data of the current patient is the brightness gradient value and coordinate value of each pixel point in the image block occupied by the joint part in the MRI image of the current patient, and the plurality of associated parameters of the current patient are the disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking mark, drinking mark, age information and gender information of the current patient. The selection of the above-mentioned rich and comprehensive basic information further ensures the reliability and stability of the intelligent identification results;

[0030] Invention point D: For each component pixel point in an image block, the brightness values corresponding to each adjacent pixel point of the component pixel point in the image block and the single brightness value of the component pixel point are combined into a brightness value set corresponding to the component pixel point, and the standard deviation of the brightness value set is calculated as the brightness gradient value of the component pixel point in the image block. For example, for each component pixel point in an image block, the brightness values corresponding to each pixel point covered by a square pixel window centered on the component pixel point in the image block are combined into a brightness value set corresponding to the component pixel point, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel point is calculated as the brightness gradient value of the component pixel point in the image block, thereby completing the targeted analysis of the brightness gradient value of each component pixel point in the image block. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The embodiments of the present invention will be described below with reference to the accompanying drawings, in which:

[0032] Figure 1 Schematic diagram of the working scenario of a digital-based rheumatoid arthritis data modeling method and system according to the present invention.

[0033] Figure 2 This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 1 of the present invention.

[0034] Figure 3 This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 2 of the present invention.

[0035] Figure 4This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 3 of the present invention.

[0036] Figure 5 This is a diagram showing the internal structure of a rheumatoid arthritis data modeling system based on digitization according to Example 4 of the present invention.

[0037] Figure 6 This is a diagram showing the internal structure of a rheumatoid arthritis data modeling system based on digitization according to Example 5 of the present invention.

[0038] Figure 7 This is a diagram showing the internal structure of a digitalized rheumatoid arthritis data modeling system according to Example 6 of the present invention. DETAILED DESCRIPTION

[0039] like Figure 1 As shown, a schematic diagram of the working scenario of the rheumatoid arthritis data modeling method and system based on digitalization according to the present invention is given.

[0040] The specific technical process of the present invention is as follows:

[0041] Step 1: Digitally process various disease-related data of current patients with rheumatoid arthritis to obtain basic information for analyzing the severity level of rheumatoid arthritis of the current patients;

[0042] Specifically, the basic information used to analyze the severity of the current patient's rheumatoid arthritis is digitally processed information, such as Figure 1 As shown, the various basic information include first visual data of the current patient related to X-ray data of the current patient, second visual data of the current patient related to MRI data of the current patient, and multiple associated parameters of the current patient related to physiological information of the current patient;

[0043] More specifically, the first visual data of the current patient is the brightness gradient value and coordinate value of each pixel point in the image block occupied by the joint part in the X-ray image of the current patient; the second visual data of the current patient is the brightness gradient value and coordinate value of each pixel point in the image block occupied by the joint part in the MRI image of the current patient; and the multiple associated parameters of the current patient are the disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information of the current patient;

[0044] In fact, if Figure 1As shown, the various basic information also includes a first resolution and a second resolution. The first resolution is the resolution of the current patient's X-ray image, and the second resolution is the resolution of the current patient's MRI image. The value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution.

[0045] More specifically, for each pixel in an image block, the brightness values corresponding to each of the adjacent pixels of the pixel in the image block and the single brightness value of the pixel are combined to form a brightness value set corresponding to the pixel, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the pixel in the image block.

[0046] For example, for each component pixel in an image block, the brightness values corresponding to each pixel covered by a square pixel window centered on the component pixel in the image block are combined to form a brightness value set corresponding to the component pixel, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel is calculated as the brightness gradient value of the component pixel in the image block, thereby completing a targeted analysis of the brightness gradient value of each component pixel in the image block;

[0047] In this way, the selection of the above-mentioned rich and comprehensive basic information further ensures the reliability and stability of the intelligent identification results;

[0048] Step 2: Use the patient's disease course data to establish a corresponding intelligent grade identification model for the current patient. Different patients correspond to intelligent grade identification models with different customized structures.

[0049] Specifically, the intelligent grade identification model corresponding to the current patient is a BP neural network that has been trained multiple times, and the number of BP neural network trainings is monotonically positively correlated with the patient's disease course data. In each training of the BP neural network, the known severity level of a patient suffering from rheumatoid arthritis is used as a single output content of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple associated parameters of the patient are used as item-by-item output contents of the BP neural network to complete this training of the BP neural network.

[0050] In this way, the reliability and stability of the intelligent identification results are guaranteed through the customized design of different models for different patients mentioned above;

[0051] Step 3: Using the corresponding intelligent grade identification model established for the current patient in Step 2, and based on the basic information of the current patient's rheumatoid arthritis severity grade that has been digitally processed in Step 1, complete the intelligent identification of the current patient's rheumatoid arthritis severity grade;

[0052] Specifically, there are four levels of severity for rheumatoid arthritis: level 1, level 2, level 3, and level 4, which correspond to four levels of severity: mild functional impairment, moderate functional impairment, severe functional impairment, and extremely severe functional impairment, respectively. The severity level of rheumatoid arthritis of the current patient obtained by intelligent identification is one of the above four levels;

[0053] Step 4: wirelessly transmitting the severity level of the rheumatoid arthritis of the current patient intelligently identified in Step 3 to a remote medical care information server via a mobile communication link;

[0054] In this way, distributed and synchronous condition management of patients in various locations is achieved at the remote medical care information server.

[0055] It can be seen that the present invention performs normalization processing on the X-ray visual data, MRI visual data, and disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking mark, drinking mark, age information and gender information of the current patient suffering from rheumatoid arthritis, thereby realizing the modeling processing of intelligent grade identification models with different structures for different patients, thereby providing an intelligent identification mechanism for the severity level of rheumatoid arthritis in different patients, improving the identification accuracy of the patient's rheumatoid arthritis condition, avoiding misdiagnosis, and further enriching the medical care information of rheumatoid arthritis patients on the basis of manual condition identification and machine condition identification of a single diagnostic mechanism.

[0056] The key points of the present invention are: normalization processing of various patient-related data including the current patient's X-ray visual data, MRI visual data and the current patient's medical course data, customized design of different intelligent level identification models for different patients, targeted design of single training of BP neural network and further enrichment of medical care information for rheumatoid arthritis patients.

[0057] Hereinafter, a digitalized rheumatoid arthritis data modeling method and system of the present invention will be specifically described in the form of an embodiment.

[0058] Example 1

[0059] Figure 2 This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 1 of the present invention.

[0060] like Figure 2As shown, the digitalized rheumatoid arthritis data modeling method comprises the following steps:

[0061] Step S201: inputting image blocks occupied by joints in an X-ray image of a current patient suffering from rheumatoid arthritis and image blocks occupied by joints in an MRI image of the current patient, and outputting them as first image blocks and second image blocks respectively;

[0062] For example, for different patients with rheumatoid arthritis, the same X-ray scanner is used to perform X-ray scans of the patients, and for different patients with rheumatoid arthritis, the same magnetic resonance imaging device is used to perform MRI examinations, i.e., magnetic resonance imaging examinations, of the patients;

[0063] Step S202: using the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block as the first visual data of the current patient, and using the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block as the second visual data of the current patient;

[0064] For example, the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block are used as the first visual data of the current patient, and the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block are used as the second visual data of the current patient, including: the coordinate value of each pixel point includes a horizontal coordinate value and a vertical coordinate value in a two-dimensional coordinate system;

[0065] Step S203: extracting the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information as multiple correlation parameters of the current patient;

[0066] For example, multiple different information extraction components may be selected to respectively extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information;

[0067] For further example, respectively extracting the disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information of the current patient includes: using different values of smoking indicators to indicate whether the current patient smokes, and using different values of drinking indicators to indicate whether the current patient drinks;

[0068] Step S204: obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network has been trained is monotonically positively correlated with the disease course data of the current patient;

[0069] Specifically, obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of BP neural network training times is monotonically positively correlated with the current patient's disease course data, including: if the current patient's disease course data is 1 year, the selected BP neural network training times is 800 times; if the current patient's disease course data is 3 years, the selected BP neural network training times is 900 times; if the current patient's disease course data is 5 years, the selected BP neural network training times is 1000 times, and so on;

[0070] Step S205: using an intelligent grade identification model to intelligently identify the severity grade of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient;

[0071] Specifically, a programmable logic controller can be used to implement simulation and testing of a data processing process for intelligently identifying the severity level of rheumatoid arthritis of a current patient based on a first resolution, a second resolution, first visual data of the current patient, second visual data of the current patient, and multiple associated parameters of the current patient using an intelligent grade identification model;

[0072] The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution;

[0073] Among them, the severity level of rheumatoid arthritis has four levels: level 1, level 2, level 3 and level 4, corresponding to the four severity levels of mild functional impairment, moderate functional impairment, severe functional impairment and extremely severe functional impairment respectively;

[0074] In each training of the BP neural network, the known severity level of a patient suffering from rheumatoid arthritis is used as a single output of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple associated parameters of the patient are used as outputs of the BP neural network item by item to complete the training of the BP neural network.

[0075] For each pixel in an image block, the brightness values corresponding to each of the adjacent pixels of the pixel in the image block and the single brightness value of the pixel are combined to form a brightness value set corresponding to the pixel, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the pixel in the image block.

[0076] And wherein, for each component pixel point in the image block, combining the respective brightness values corresponding to each adjacent pixel point of the component pixel point in the image block and the single brightness value of the component pixel point into a brightness value set corresponding to the component pixel point, and calculating the brightness value standard deviation of the brightness value set as the brightness gradient value of the component pixel point in the image block includes: for each component pixel point in the image block, combining the respective brightness values corresponding to each pixel point covered by a square pixel window centered on the component pixel point in the image block into the brightness value set corresponding to the component pixel point, and calculating the standard deviation of all brightness values in the brightness value set corresponding to the component pixel point as the brightness gradient value of the component pixel point in the image block;

[0077] For example, for each component pixel in an image block, the brightness values corresponding to each pixel covered by a square pixel window centered on the component pixel in the image block are combined to form a brightness value set corresponding to the component pixel, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel is calculated as the brightness gradient value of the component pixel in the image block, including: the square pixel window centered on the component pixel in the image block is a square pixel window of 3 pixels by 3 pixels.

[0078] Example 2

[0079] Figure 3 This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 2 of the present invention.

[0080] like Figure 3 As shown, after the intelligent grade identification model is used to intelligently identify the severity grade of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient, that is, after step S205, the digitalized rheumatoid arthritis data modeling method further includes:

[0081] Step S206: receiving the severity level of the rheumatoid arthritis of the current patient, and wirelessly transmitting the severity level of the rheumatoid arthritis of the current patient to a remote medical care information server via a mobile communication link;

[0082] Specifically, receiving the severity level of rheumatoid arthritis of the current patient and wirelessly transmitting the severity level of rheumatoid arthritis of the current patient to a remote healthcare information server via a mobile communication link includes: the mobile communication link can be implemented using a frequency division duplex communication link or a time division duplex communication link.

[0083] Example 3

[0084] Figure 4 This is a flowchart of the steps of a digitalized rheumatoid arthritis data modeling method according to Example 3 of the present invention.

[0085] like Figure 4 As shown, after obtaining the intelligent level identification model corresponding to the current patient, the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of BP neural network training times is monotonically positively correlated with the disease course data of the current patient, that is, after step S204, the digitalized rheumatoid arthritis data modeling method further includes:

[0086] Step S207: receiving the intelligence level identification model corresponding to the current patient, and completing the model storage of the intelligence level identification model corresponding to the current patient by storing various model data of the intelligence level identification model corresponding to the current patient;

[0087] For example, the object storage mechanism is connected to the level identification mechanism, and is used to receive the intelligent level identification model corresponding to the current patient, and complete the model storage of the intelligent level identification model corresponding to the current patient by storing various model data of the intelligent level identification model corresponding to the current patient, including: TF storage chip, SD storage chip or static storage chip can be selected to implement the object storage mechanism.

[0088] And in any one of the above embodiments 1-3, optionally, in the digitalized rheumatoid arthritis data modeling method:

[0089] For each pixel in the image block, the vertical coordinate value and the horizontal coordinate value of the pixel in the image block are used as the coordinate value corresponding to the pixel;

[0090] The intelligent grade identification model is used to intelligently identify the severity grade of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient, including: the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution;

[0091] For example, the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution. This includes: the total number of pixel rows in the first resolution is actually the number of pixels occupied by each pixel column, and the total number of pixel columns in the first resolution is actually the number of pixels occupied by each pixel row;

[0092] Wherein, using the intelligent grade identification model to intelligently identify the severity level of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient further comprises: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient, and then inputting them into the intelligent grade identification model in parallel;

[0093] Wherein, using the intelligent grade identification model to intelligently identify the severity level of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient further comprises: the severity level of the rheumatoid arthritis of the current patient output by the intelligent grade identification model is a numerical representation of a normalized value;

[0094] The step of performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent grade identification model in parallel includes: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient using a numerical processing device;

[0095] And wherein, the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient are respectively subjected to numerical normalization processing and then input in parallel into the intelligent level identification model, which also includes: using a parallel control device connected to the numerical processing device to input the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient after the numerical normalization processing are respectively performed into the intelligent level identification model in parallel.

[0096] Example 4

[0097] Figure 5 This is a diagram showing the internal structure of a rheumatoid arthritis data modeling system based on digitization according to Example 4 of the present invention.

[0098] like Figure 5 As shown, the digitalized rheumatoid arthritis data modeling system includes the following components:

[0099] a block acquisition mechanism for recording image blocks occupied by joints in an X-ray image of a current patient suffering from rheumatoid arthritis and image blocks occupied by joints in an MRI image of the current patient, and outputting them as first image blocks and second image blocks, respectively;

[0100] For example, for different patients with rheumatoid arthritis, the same X-ray scanner is used to perform X-ray scans of the patients, and for different patients with rheumatoid arthritis, the same magnetic resonance imaging device is used to perform MRI examinations, i.e., magnetic resonance imaging examinations, of the patients;

[0101] a visual recording mechanism connected to the block acquisition mechanism, configured to use the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block as the first visual data of the current patient, and use the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block as the second visual data of the current patient;

[0102] For example, the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block are used as the first visual data of the current patient, and the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block are used as the second visual data of the current patient, including: the coordinate value of each pixel point includes a horizontal coordinate value and a vertical coordinate value in a two-dimensional coordinate system;

[0103] A parameter extraction mechanism is used to extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information and gender information as multiple related parameters of the current patient;

[0104] For example, multiple different information extraction components may be selected to respectively extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information;

[0105] For further example, respectively extracting the disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information, and gender information of the current patient includes: using different values of smoking indicators to indicate whether the current patient smokes, and using different values of drinking indicators to indicate whether the current patient drinks;

[0106] A multi-layer training mechanism is used to obtain an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network is trained is monotonically positively correlated with the current patient's disease course data;

[0107] Specifically, obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of BP neural network training times is monotonically positively correlated with the current patient's disease course data, including: if the current patient's disease course data is 1 year, the selected BP neural network training times is 800 times; if the current patient's disease course data is 3 years, the selected BP neural network training times is 900 times; if the current patient's disease course data is 5 years, the selected BP neural network training times is 1000 times, and so on;

[0108] a grade identification mechanism, connected to the visual input mechanism, the parameter extraction mechanism, and the multi-layer training mechanism, respectively, for using an intelligent grade identification model to intelligently identify the severity grade of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient;

[0109] Specifically, a programmable logic controller can be used to implement simulation and testing of a data processing process for intelligently identifying the severity level of rheumatoid arthritis of a current patient based on a first resolution, a second resolution, first visual data of the current patient, second visual data of the current patient, and multiple associated parameters of the current patient using an intelligent grade identification model;

[0110] The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution;

[0111] Among them, the severity level of rheumatoid arthritis has four levels: level 1, level 2, level 3 and level 4, corresponding to the four severity levels of mild functional impairment, moderate functional impairment, severe functional impairment and extremely severe functional impairment respectively;

[0112] In each training of the BP neural network, the known severity level of a patient suffering from rheumatoid arthritis is used as a single output of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple associated parameters of the patient are used as outputs of the BP neural network item by item to complete the training of the BP neural network.

[0113] For each pixel in an image block, the brightness values corresponding to each of the adjacent pixels of the pixel in the image block and the single brightness value of the pixel are combined to form a brightness value set corresponding to the pixel, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the pixel in the image block.

[0114] And wherein, for each component pixel point in the image block, combining the respective brightness values corresponding to each adjacent pixel point of the component pixel point in the image block and the single brightness value of the component pixel point into a brightness value set corresponding to the component pixel point, and calculating the brightness value standard deviation of the brightness value set as the brightness gradient value of the component pixel point in the image block includes: for each component pixel point in the image block, combining the respective brightness values corresponding to each pixel point covered by a square pixel window centered on the component pixel point in the image block into the brightness value set corresponding to the component pixel point, and calculating the standard deviation of all brightness values in the brightness value set corresponding to the component pixel point as the brightness gradient value of the component pixel point in the image block;

[0115] For example, for each component pixel in an image block, the brightness values corresponding to each pixel covered by a square pixel window centered on the component pixel in the image block are combined to form a brightness value set corresponding to the component pixel, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel is calculated as the brightness gradient value of the component pixel in the image block, including: the square pixel window centered on the component pixel in the image block is a square pixel window of 3 pixels by 3 pixels.

[0116] Example 5

[0117] Figure 6 This is a diagram showing the internal structure of a rheumatoid arthritis data modeling system based on digitization according to Example 5 of the present invention.

[0118] like Figure 6 As shown, the digitalized rheumatoid arthritis data modeling system further includes:

[0119] a mobile transmission mechanism connected to the level identification mechanism, configured to receive the severity level of the rheumatoid arthritis of the current patient and wirelessly transmit the severity level of the rheumatoid arthritis of the current patient to a remote medical care information server via a mobile communication link;

[0120] Specifically, receiving the severity level of rheumatoid arthritis of the current patient and wirelessly transmitting the severity level of rheumatoid arthritis of the current patient to a remote healthcare information server via a mobile communication link includes: the mobile communication link can be implemented using a frequency division duplex communication link or a time division duplex communication link.

[0121] Example 6

[0122] Figure 7 This is a diagram showing the internal structure of a digitalized rheumatoid arthritis data modeling system according to Example 6 of the present invention.

[0123] like Figure 7 As shown, the digitalized rheumatoid arthritis data modeling system further includes:

[0124] The object storage mechanism is connected to the level identification mechanism, and is used to receive the intelligent level identification model corresponding to the current patient, and complete the model storage of the intelligent level identification model corresponding to the current patient by storing various model data of the intelligent level identification model corresponding to the current patient;

[0125] For example, the object storage mechanism is connected to the level identification mechanism, and is used to receive the intelligent level identification model corresponding to the current patient, and complete the model storage of the intelligent level identification model corresponding to the current patient by storing various model data of the intelligent level identification model corresponding to the current patient, including: TF storage chip, SD storage chip or static storage chip can be selected to implement the object storage mechanism.

[0126] And in any one of the above embodiments 4-6, optionally, in the digitalized rheumatoid arthritis data modeling system:

[0127] For each pixel in the image block, the vertical coordinate value and the horizontal coordinate value of the pixel in the image block are used as the coordinate value corresponding to the pixel;

[0128] The intelligent grade identification model is used to intelligently identify the severity grade of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient, including: the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution;

[0129] For example, the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution. This includes: the total number of pixel rows in the first resolution is actually the number of pixels occupied by each pixel column, and the total number of pixel columns in the first resolution is actually the number of pixels occupied by each pixel row;

[0130] Wherein, using the intelligent grade identification model to intelligently identify the severity level of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient further comprises: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient, and then inputting them into the intelligent grade identification model in parallel;

[0131] Wherein, using the intelligent grade identification model to intelligently identify the severity level of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient further comprises: the severity level of the rheumatoid arthritis of the current patient output by the intelligent grade identification model is a numerical representation of a normalized value;

[0132] The step of performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent grade identification model in parallel includes: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient using a numerical processing device;

[0133] And wherein, the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient are respectively subjected to numerical normalization processing and then input in parallel into the intelligent level identification model, which also includes: using a parallel control device connected to the numerical processing device to input the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient after the numerical normalization processing are respectively performed into the intelligent level identification model in parallel.

[0134] In addition, in a digitalized rheumatoid arthritis data modeling method and system according to the present invention:

[0135] Obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network is trained is monotonically positively correlated with the course of disease data of the current patient, including: using an information mapping function to represent an information mapping relationship of the monotonically positive correlation between the number of times the BP neural network is trained and the course of disease data of the current patient;

[0136] The information mapping relationship of using an information mapping function to represent a monotonically positive correlation between the number of BP neural network training times and the current patient's disease course data includes: in the information mapping function, the current patient's disease course data is the input information of the information mapping function, and the number of BP neural network training times corresponding to the current patient's disease course data is the output information of the information mapping function;

[0137] For example, the information mapping relationship of using the information mapping function to represent the monotonically positive correlation between the number of times the BP neural network is trained and the disease course data of the current patient further includes: optionally using a MATLAB toolbox to complete the testing and simulation of the implementation process of the information mapping function;

[0138] The step of performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent level identification model in parallel further includes: performing hexadecimal numerical conversion processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent level identification model in parallel;

[0139] And wherein, the severity level of rheumatoid arthritis of the current patient output by the intelligent level identification model is a numerical representation of a normalized value, including: the severity level of rheumatoid arthritis of the current patient output by the intelligent level identification model is a numerical representation of a hexadecimal value.

[0140] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0141] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device / electronic device / computer-readable storage medium / computer program product embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The above is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A digital data modeling method for rheumatoid arthritis, characterized in that: The method comprises: Input an image block occupied by a joint portion in an X-ray image of a current patient suffering from rheumatoid arthritis and an image block occupied by a joint portion in an MRI image of the current patient, and output them as a first image block and a second image block respectively; The brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block are used as the first visual data of the current patient, and the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block are used as the second visual data of the current patient; Extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking mark, drinking mark, age information and gender information as multiple correlation parameters of the current patient; Obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network has been trained is monotonically positively correlated with the disease course data of the current patient; intelligently identifying the severity level of rheumatoid arthritis of the current patient using an intelligent grade identification model based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient; The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution.

2. The digitalized rheumatoid arthritis data modeling method according to claim 1, wherein: Rheumatoid arthritis severity is classified into four levels: 1, 2, 3, and 4, corresponding to the four severity levels of mild functional impairment, moderate functional impairment, severe functional impairment, and extremely severe functional impairment, respectively. In each training of the BP neural network, the known severity level of a patient suffering from rheumatoid arthritis is used as a single output of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple associated parameters of the patient are used as outputs of the BP neural network item by item to complete the training of the BP neural network. For each pixel in an image block, the brightness values corresponding to each of the adjacent pixels of the pixel in the image block and the single brightness value of the pixel are combined to form a brightness value set corresponding to the pixel, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the pixel in the image block. Among them, for each component pixel point in the image block, the brightness values corresponding to each adjacent pixel point of the component pixel point in the image block and the single brightness value of the component pixel point are combined into a brightness value set corresponding to the component pixel point, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the component pixel point in the image block, including: for each component pixel point in the image block, the brightness values corresponding to each pixel point covered by a square pixel window centered on the component pixel point in the image block are combined into a brightness value set corresponding to the component pixel point, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel point is calculated as the brightness gradient value of the component pixel point in the image block.

3. The digital data modeling method for rheumatoid arthritis according to claim 2, wherein: After intelligently identifying the severity level of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient using the intelligent level identification model, the method further includes: The severity level of rheumatoid arthritis of the current patient is received, and the severity level of rheumatoid arthritis of the current patient is wirelessly transmitted to a remote medical care information server via a mobile communication link.

4. The digitalized rheumatoid arthritis data modeling method according to claim 2, wherein: After obtaining an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network is trained is monotonically positively correlated with the disease course data of the current patient, the method further includes: The intelligent level identification model corresponding to the current patient is received, and the model storage of the intelligent level identification model corresponding to the current patient is completed by storing various model data of the intelligent level identification model corresponding to the current patient.

5. The digitalized rheumatoid arthritis data modeling method according to any one of claims 2 to 4, characterized in that: For each pixel in the image block, the vertical coordinate value and the horizontal coordinate value of the pixel in the image block are used as the coordinate value corresponding to the pixel; The intelligent grade identification model is used to intelligently identify the severity grade of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient, including: the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution; Wherein, using the intelligent grade identification model to intelligently identify the severity level of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient further comprises: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient, and then inputting them into the intelligent grade identification model in parallel; Wherein, using the intelligent grade identification model to intelligently identify the severity level of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient further comprises: the severity level of the rheumatoid arthritis of the current patient output by the intelligent grade identification model is a numerical representation of a normalized value; The step of performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent grade identification model in parallel includes: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient using a numerical processing device; Among them, the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient are respectively subjected to numerical normalization processing and then input in parallel into the intelligent level identification model, which also includes: using a parallel control device connected to the numerical processing device to input the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient after the numerical normalization processing are respectively performed into the intelligent level identification model in parallel.

6. A digital rheumatoid arthritis data modeling system, characterized in that: The system comprises: a block acquisition mechanism for recording image blocks occupied by joints in an X-ray image of a current patient suffering from rheumatoid arthritis and image blocks occupied by joints in an MRI image of the current patient, and outputting them as first image blocks and second image blocks, respectively; a visual recording mechanism connected to the block acquisition mechanism, configured to use the brightness gradient values and coordinate values corresponding to the respective pixel points in the first image block as the first visual data of the current patient, and use the brightness gradient values and coordinate values corresponding to the respective pixel points in the second image block as the second visual data of the current patient; A parameter extraction mechanism is used to extract the current patient's disease course data, C-reactive protein value, erythrocyte sedimentation rate value, smoking indicator, drinking indicator, age information and gender information as multiple related parameters of the current patient; A multi-layer training mechanism is used to obtain an intelligent level identification model corresponding to the current patient, wherein the intelligent level identification model is a BP neural network that has been trained multiple times, and the number of times the BP neural network is trained is monotonically positively correlated with the current patient's disease course data; a grade identification mechanism, connected to the visual input mechanism, the parameter extraction mechanism, and the multi-layer training mechanism, respectively, for using an intelligent grade identification model to intelligently identify the severity grade of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and a plurality of associated parameters of the current patient; The resolution of the X-ray images of each patient with rheumatoid arthritis is fixed and is the first resolution, and the resolution of the MRI images of each patient with rheumatoid arthritis is fixed and is the second resolution.

7. The digitalized rheumatoid arthritis data modeling system according to claim 6, characterized in that: Rheumatoid arthritis severity is classified into four levels: 1, 2, 3, and 4, corresponding to the four severity levels of mild functional impairment, moderate functional impairment, severe functional impairment, and extremely severe functional impairment, respectively. In each training of the BP neural network, the known severity level of a patient suffering from rheumatoid arthritis is used as a single output of the BP neural network, and the first resolution, the second resolution, the first visual data of the patient, the second visual data of the patient, and multiple associated parameters of the patient are used as outputs of the BP neural network item by item to complete the training of the BP neural network. For each pixel in an image block, the brightness values corresponding to each of the adjacent pixels of the pixel in the image block and the single brightness value of the pixel are combined to form a brightness value set corresponding to the pixel, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the pixel in the image block. Among them, for each component pixel point in the image block, the brightness values corresponding to each adjacent pixel point of the component pixel point in the image block and the single brightness value of the component pixel point are combined into a brightness value set corresponding to the component pixel point, and the brightness value standard deviation of the brightness value set is calculated as the brightness gradient value of the component pixel point in the image block, including: for each component pixel point in the image block, the brightness values corresponding to each pixel point covered by a square pixel window centered on the component pixel point in the image block are combined into a brightness value set corresponding to the component pixel point, and the standard deviation of all brightness values in the brightness value set corresponding to the component pixel point is calculated as the brightness gradient value of the component pixel point in the image block.

8. The digitalized rheumatoid arthritis data modeling system according to claim 2, wherein: The system further comprises: The mobile transmission mechanism is connected to the grade identification mechanism and is used for receiving the severity grade of the rheumatoid arthritis of the current patient and wirelessly transmitting the severity grade of the rheumatoid arthritis of the current patient to a remote medical care information server through a mobile communication link.

9. The digitalized rheumatoid arthritis data modeling system according to claim 2, wherein: The system further comprises: The object storage mechanism is connected to the level identification mechanism, and is used to receive the intelligent level identification model corresponding to the current patient, and complete the model storage of the intelligent level identification model corresponding to the current patient by storing various model data of the intelligent level identification model corresponding to the current patient.

10. The digitalized rheumatoid arthritis data modeling system according to any one of claims 7 to 9, characterized in that: For each pixel in the image block, the vertical coordinate value and the horizontal coordinate value of the pixel in the image block are used as the coordinate value corresponding to the pixel; The intelligent grade identification model is used to intelligently identify the severity grade of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient, including: the numerical value of the first resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the first resolution, and the numerical value of the second resolution is expressed as the product of the total number of pixel rows and the total number of pixel columns in the second resolution; Wherein, using the intelligent grade identification model to intelligently identify the severity level of rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient further comprises: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the multiple associated parameters of the current patient, and then inputting them into the intelligent grade identification model in parallel; Wherein, using the intelligent grade identification model to intelligently identify the severity level of the rheumatoid arthritis of the current patient based on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and multiple associated parameters of the current patient further comprises: the severity level of the rheumatoid arthritis of the current patient output by the intelligent grade identification model is a numerical representation of a normalized value; The step of performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient and then inputting the processing into the intelligent grade identification model in parallel includes: performing numerical normalization processing on the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient, and the plurality of associated parameters of the current patient using a numerical processing device; Among them, the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient are respectively subjected to numerical normalization processing and then input in parallel into the intelligent level identification model, which also includes: using a parallel control device connected to the numerical processing device to input the first resolution, the second resolution, the first visual data of the current patient, the second visual data of the current patient and multiple related parameters of the current patient after the numerical normalization processing are respectively performed into the intelligent level identification model in parallel.

Images