A medical imaging big data analysis method and analysis system

Abstract

The present application discloses an analysis method and analysis system for medical imaging big data, and relates to the technical field of medical imaging data analysis. The present application classifies and analyzes historical imaging images in various hospitals in a target area to obtain a prediction set of associated diseases of a target disease, and matches the medical imaging images of a target patient with the prediction set of associated diseases of the target disease to obtain the diagnosis probability of each associated disease of the patient. The present application successfully increases the prediction success probability of each associated disease of the patient by 70%. Doctors can formulate highly personalized treatment plans based on the diagnosis probability of each associated disease of the patient, preventively adjust the type and dosage of medication, and improve the treatment effect and the patient's recovery rate.

Description

Technical Field

[0001] The present application relates to the technical field of medical imaging data analysis, and specifically to an analysis method and analysis system for medical imaging big data. Background Art

[0002] In today's medical field, accurate diagnosis and effective treatment of diseases have always been the core pursuit. Medical imaging plays an extremely critical role in the diagnosis and treatment of various diseases. However, under the traditional medical model, the data of each hospital is relatively isolated and lacks an effective integration and sharing mechanism. This greatly limits the comprehensive research and accurate judgment of certain diseases. In addition, the current prediction of related diseases relies solely on the experience and judgment of doctors, which lacks perfection and rigor. Therefore, this application proposes an analysis method and analysis system for medical imaging big data.

[0003] Prior art, such as the invention patent application with announcement number: CN117876344A, discloses a method, device, and computer-readable medium for processing medical image data, the method comprising: acquiring medical image data; determining, based on the medical image data, first information of a first layer, second information of a second layer, and third information of a third layer; wherein the first information is used to represent the analysis result of the medical image data, the second information is used to represent the lesion segmentation result of lesion segmentation of the medical image data based on the analysis result; the third information is used to represent the manual annotation result of manual annotation of the medical image data based on the lesion segmentation result; the first information of the first layer, the second information of the second layer, and the third information of the third layer are fused to determine a target annotation image, which can at least be used to solve the technical problem that tiny lesions are easily missed during image analysis, resulting in low accuracy of diagnostic results based on medical image data.

[0004] Regarding the above scheme, there are the following technical problems: 1. The current technology mainly integrates and processes the information in each layer of medical imaging data, determines the target annotation object, and thus solves the problem that tiny lesions are easily missed. The current technology mainly focuses on the processing of single medical imaging images, and does not conduct a comprehensive analysis of various historical imaging data of the same disease. The same disease may have similar characteristics and patterns in different patients or different stages of the same patient. Comprehensive analysis of a large amount of historical imaging data can summarize these common characteristics and improve the accuracy and reliability of diagnosis. Medical research requires a large amount of data to explore the occurrence and development mechanism of the disease, as well as to evaluate the effectiveness of new diagnostic methods and treatment means. The value of a single imaging data is limited.

[0005] 2. Current technology mainly focuses on the analysis of medical imaging data of patients' confirmed diseases, and does not take into account the analysis of various related diseases of the confirmed diseases. There are correlations between many diseases. The current technology's neglect of this level may miss the early detection and prevention of various potential diseases, causing the patient's condition to further deteriorate. Summary of the Invention

[0006] The purpose of this application is to provide a medical imaging big data analysis method and analysis system to solve the problems existing in the background technology.

[0007] In order to solve the above technical problems, the present application adopts the following technical solutions: In the first aspect, the present application provides a method for analyzing medical imaging big data, including: Step 1, obtaining target disease information: obtaining each historical target image in each hospital in the target area, and then obtaining the medical information of each historical patient corresponding to each historical target image.

[0008] Step 2: Target disease information analysis: Classify each medical image based on the medical information of each historical patient, and then analyze and obtain the associated disease prediction set of the target disease.

[0009] Step 3: Prediction of associated disease information: Match the target patient's medical imaging images with the associated prediction set of the target disease, and predict the target patient's associated diseases and the probability of diagnosis of the associated diseases based on the matching results.

[0010] In a second aspect, the present application provides an analysis system for medical imaging big data, including: a target disease information acquisition module: used to obtain historical target imaging images in each hospital in the target area, and then obtain medical information of each historical patient corresponding to each historical target imaging image.

[0011] Target disease information analysis module: used to classify each medical image based on the medical information of each historical patient, and then analyze and obtain the associated disease prediction set of the target disease.

[0012] Related disease information prediction module: used to match the medical imaging images of the target patient with the related amount set of the target disease, and predict the related diseases of the target patient and the diagnosis probability of the related diseases based on the matching results.

[0013] The beneficial effects of the present application are: 1. The present application provides a medical imaging big data analysis method and analysis system, which classifies and analyzes historical imaging images in various hospitals in the target area to obtain a set of associated disease predictions for the target disease, and matches the medical imaging images of the target patient with the set of associated disease predictions for the target disease to obtain the diagnosis probability of each associated disease of the patient. The present application successfully increases the prediction success probability of each associated disease of the patient by 70%. Doctors can formulate highly personalized treatment plans based on the diagnosis probability of each associated disease of the patient, preventively adjust the type and dosage of medication, and improve the treatment effect and the patient's recovery rate.

[0014] 2. This application integrates massive multi-source data by acquiring historical target images and corresponding patient medical information from hospitals in the target area, which is conducive to breaking the data silo phenomenon and providing a strong data basis for subsequent analysis and prediction of related diseases.

[0015] 3. This application classifies each medical image based on the medical information of each historical patient, and then analyzes the related diseases of each type of medical image to obtain the disease diagnosis probability and the illness diagnosis probability of each related disease. The accurate disease diagnosis probability and illness diagnosis probability allow doctors to no longer rely solely on experience and fuzzy judgment when predicting related diseases. This application successfully increases the success rate of predicting each related disease of patients by 70%. Doctors can decide the direction of subsequent examinations based on the diagnosis probability of each related disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 The figure is a flowchart of the steps for implementing the application method.

[0018] Figure 2 This is a schematic diagram of the system structure connection for this application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0020] Reference Figure 1 As shown, the present application provides a method for analyzing medical imaging big data in the first aspect, comprising the following steps: Step 1, obtaining target disease information: obtaining each historical target image in each hospital in the target area, and then obtaining the medical information of each historical patient corresponding to each historical target image.

[0021] In a specific example, the specific process of obtaining target images in each hospital in the target area is as follows: obtain an arbitrary area from a satellite remote sensing system, record it as the target area, and then obtain each hospital in the target area, record it as each target hospital, and any disease diagnosed with the assistance of medical images is recorded as the target disease, obtain each historical medical image of the target disease from the electronic medical record system of each target hospital, record it as each target medical image, and obtain each related disease of the target disease and each medical image of each related disease from the electronic medical record system of each target hospital.

[0022] It should be noted that any area must include at least 3 hospitals.

[0023] It should be noted that the target disease may be associated with various diseases, for example, brain tumors may be associated with epilepsy and cerebral edema, and bone tumors may be associated with fractures and osteoarthritis.

[0024] In a specific example, the medical information of each historical patient includes diagnosis and treatment information and personal information. The diagnosis and treatment information includes each related disease of each historical patient, each medical imaging image of each related disease and the diagnosis timestamp of each related disease; the personal information includes gender and age type; the diagnosis timestamp of each related disease is the length of time between the diagnosis time of each related disease and the diagnosis time of the target disease.

[0025] It should be noted that the associated disease types of each historical patient are the associated disease types of the target disease suffered by each historical patient. For example, the associated diseases of brain tumors are epilepsy and cerebral edema, and the associated diseases of bone tumors are fractures and osteoarthritis.

[0026] It should be noted that patients under the age of 18 were recorded as juvenile patients, patients aged 18-30 were recorded as young patients, patients aged 30-50 were recorded as middle-aged patients, and patients aged 50 and above were recorded as elderly patients, based on which the age type of each historical patient was obtained.

[0027] Step 2: Target disease information analysis: Classify each medical image based on the medical information of each historical patient, and then analyze and obtain the associated disease prediction set of the target disease.

[0028] In a specific example, the medical imaging images are classified based on the medical information of each historical patient. The specific process is as follows: the personal information of each historical patient is obtained based on the medical information of each historical patient, the gender and age type of each historical patient is obtained based on the personal information of each historical patient, and each target medical imaging image is classified based on the gender and age type of each historical patient to obtain various types of target medical imaging images.

[0029] Based on the medical information of each historical patient, the diagnosis and treatment information of each historical patient is obtained. Based on the diagnosis and treatment information of each historical patient, each related disease of each historical patient and the confirmed timestamp of each related disease are obtained. Then, the minimum confirmed timestamp of the related disease of each historical patient is obtained, thereby obtaining the related disease corresponding to the minimum confirmed timestamp of the related disease of each historical patient. The related disease corresponding to the minimum confirmed timestamp of the related disease of a historical patient is recorded as the first-level related disease of the historical patient. Then, the minimum confirmed timestamp of the related disease among the confirmed timestamps of the remaining related diseases of the historical patient is obtained, thereby obtaining the related disease corresponding to the minimum timestamp of the related disease, and recording the related disease as the second-level related disease of the historical patient. Based on this, the related diseases of each historical patient at all levels are obtained. The medical imaging images of each related disease are classified based on the related diseases of each historical patient at all levels to obtain the related medical imaging images of each target medical imaging image at all levels.

[0030] In a specific example, the analysis obtains a set of associated disease predictions for the target disease. The specific process is as follows: the categories of various target medical image images are recorded as index information of the target medical image images, and each target medical image image is stored in the corresponding index information to obtain an index information set of the target disease.

[0031] It should be noted that the index information includes gender, age type and target medical imaging images of historical patients of the same gender and age type.

[0032] Obtain each association level of each associated disease in each index in the index information set, analyze each association level of each associated disease of each target medical image in each index to obtain the diagnosis probability of each associated disease and the diagnosis probability of each level in each index information set, and comprehensively record the diagnosis probability of each associated disease, the probability of disease and the corresponding associated medical image corresponding to each index as a subset of each index.

[0033] It should be noted that one index information includes multiple target medical images, and therefore one index information includes multiple associated diseases and associated disease levels.

[0034] The associated disease prediction set of the target disease is constructed by integrating the subsets of each index and the index information set of the target disease.

[0035] In a specific example, the analysis of each correlation level of each associated disease based on each target medical image in each index obtains the diagnosis probability of each associated disease and the diagnosis probability of each level in each index information set. The specific process is as follows: the total number of historical patients in each index information set is counted, recorded as , where i represents the number of each index information set, i is a positive integer, and the number of confirmed cases of each associated disease at each associated level in each index information set is counted, recorded as , where j represents the number of each associated disease, j is a positive integer, and k is the number of the level of the associated disease, k is a positive integer, according to the calculation formula: The analysis obtains the diagnosis probability of each associated disease at each level in each index information set.

[0036] Count the total number of confirmed cases of each associated disease in each index information set, recorded as , according to the calculation formula: Analyze and obtain the probability of diagnosis of each associated disease in each index information set , where J is the total number of associated diseases.

[0037] Step 3: Prediction of related disease information: Match the medical imaging images of the target patient with the related disease prediction set of the target disease, and predict the related diseases of the target patient and the probability of diagnosis of the related diseases based on the matching results.

[0038] In a specific example, the medical imaging image of the target patient is matched with the associated disease prediction set of the target disease. The specific process is as follows: the patient diagnosed with the target disease is recorded as the target patient, the medical information of the target patient is obtained from the hospital electronic medical record system, the personal information and diagnosis and treatment information of the target patient are obtained based on the medical information of the target patient, and then the index information of the target patient is obtained, and the index information of the target patient is matched with each index information in the associated disease prediction set of the target disease, thereby obtaining the associated disease prediction subset of the target disease.

[0039] It should be noted that, in addition to gender and age type, the consistent matching also means that the medical image of the target patient is consistent with each medical image in the index information.

[0040] The medical imaging images of the target patient are obtained based on the medical information of the target patient. When the medical imaging images of the target patient are all medical imaging images of the target disease, it indicates that the target patient has not been diagnosed with the related diseases.

[0041] When the medical imaging images of the target patient include the medical imaging image of the target disease and the medical imaging images of the associated diseases, it indicates that the target patient has been diagnosed with the associated diseases.

[0042] In a specific example, the associated diseases and the probability of diagnosis of associated diseases of the target patient are predicted based on the matching results. The specific process is as follows: when the target patient has not yet been diagnosed with each associated disease, the first-level diagnosis probability and the probability of diagnosis of illness of each associated disease in the associated disease prediction subset of the target patient are obtained based on the associated disease prediction set of the target disease, and the result obtained by multiplying the first-level diagnosis probability of each associated disease by the probability of diagnosis of illness is recorded as the probability of illness of each associated infectious disease of the target patient.

[0043] When the target patient has been diagnosed with various related diseases, the diagnosis level of each confirmed related disease of the target patient is obtained based on the target patient's medical information analysis, and the next diagnosis level of each confirmed related disease of the target patient is recorded as the target prediction level, and the related diseases included in the target prediction level in the target patient's related disease prediction subset are obtained, and then the related diseases in which the target patient is diagnosed among the related diseases included in the target prediction level are obtained, and recorded as target predicted related diseases, and the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease are obtained, and the product of the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease is recorded as the disease probability of each target predicted related disease.

[0044] It should be noted that based on the predicted probability of each related disease, a personalized treatment plan can be formulated for the target patient. For example, when the target patient is diagnosed with heart disease, it is analyzed that the probability of suffering from heart failure is 40% and the probability of suffering from arrhythmia is 30%. Based on the basic medication for treating heart disease, the doctor can adjust the patient's fluid management strategy in advance based on the risk of heart failure, and preventively adjust the type and dosage of medication based on the risk of arrhythmia, so as to achieve accurate customization of the treatment plan, improve the treatment effect and the patient's recovery rate.

[0045] It should be noted that the analysis of the target patient's medical information to obtain the confirmed levels of the target patient's various related diseases is the same as the analysis of the confirmed levels of the various related diseases of each historical patient, so it will not be repeated here.

[0046] Reference Figure 2 As shown, in the second aspect, the present application provides an analysis system for medical imaging big data, including the following modules: a target disease information acquisition module: used to obtain historical target imaging images in each hospital in the target area, and then obtain the medical information of each historical patient corresponding to each historical target imaging image.

[0047] Target disease information analysis module: used to classify each medical image based on the medical information of each historical patient, and then analyze and obtain the associated disease prediction set of the target disease.

[0048] Related disease information prediction module: used to match the medical imaging images of the target patient with the related disease prediction set of the target disease, and predict the related diseases of the target patient and the probability of diagnosis of the related diseases based on the matching results.

[0049] The present application provides a method and system for analyzing medical imaging big data. By classifying and analyzing historical imaging images in various hospitals in a target area, a prediction set of associated diseases of a target disease is obtained. By matching the medical imaging images of a target patient with the prediction set of associated diseases of the target disease, the diagnosis probability of each associated disease of the patient is obtained. The present application successfully increases the prediction success probability of each associated disease of the patient by 70%. Doctors can formulate highly personalized treatment plans based on the diagnosis probability of each associated disease of the patient, preventively adjust the type and dosage of medication, and improve the treatment effect and the patient's recovery rate.

[0050] The above content is merely an example and explanation of the concept of the present application. Technicians in this technical field may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this application, they should all fall within the scope of protection of this application.

Claims

1. A method for analyzing medical image big data, characterized in that: include: Step 1: Obtain target disease information: Obtain historical target images from hospitals in the target area, and then obtain medical information of historical patients corresponding to each historical target image; The specific process of obtaining target images in each hospital in the target area is as follows: Obtain any area from a satellite remote sensing system, record it as a target area, then obtain each hospital in the target area, record it as each target hospital, and record any disease diagnosed with the assistance of medical images as a target disease, obtain each historical medical image of the target disease from the electronic medical record system of each target hospital, record it as each target medical image, and obtain each related disease of the target disease and each medical image of each related disease from the electronic medical record system of each target hospital; The medical information of each historical patient includes diagnosis and treatment information and personal information. The diagnosis and treatment information includes each associated disease of each historical patient, each medical image of each associated disease, and the diagnosis timestamp of each associated disease. The personal information includes gender and age type. The diagnosis timestamp of each associated disease is the time between the diagnosis time of each associated disease and the diagnosis time of the target disease. Step 2: Target disease information analysis: Classify each medical image based on the historical medical information of each patient, and then analyze and obtain the associated disease prediction set of the target disease; The specific process of classifying each medical image based on the medical information of each historical patient is as follows: Obtaining personal information of each historical patient based on the medical information of each historical patient, obtaining the gender and age type of each historical patient based on the personal information of each historical patient, and classifying each target medical image based on the gender and age type of each historical patient to obtain various target medical image images; Based on the medical information of each historical patient, the diagnosis and treatment information of each historical patient is obtained; based on the diagnosis and treatment information of each historical patient, each related disease of each historical patient and the diagnosis timestamp of each related disease are obtained; and then the minimum diagnosis timestamp of the related disease of each historical patient is obtained, thereby obtaining the related disease corresponding to the minimum diagnosis timestamp of the related disease of each historical patient; the related disease corresponding to the minimum diagnosis timestamp of the related disease of a certain historical patient is recorded as the first-level related disease of the historical patient; and then the minimum diagnosis timestamp of the related disease among the diagnosis timestamps of the remaining related diseases of the historical patient is obtained, thereby obtaining the related disease corresponding to the minimum timestamp of the related disease, and recording the related disease as the second-level related disease of the historical patient; accordingly, the related diseases of each historical patient at all levels are obtained; and the medical imaging images of each related disease are classified based on the related diseases of each historical patient at all levels to obtain related medical imaging images of each level of each target medical imaging image; Step 3: Prediction of related disease information: Match the medical imaging images of the target patient with the related disease prediction set of the target disease, and predict the related diseases of the target patient and the diagnosis probability of the related diseases based on the matching results; The specific process of matching the medical imaging images of the target patient with the associated disease prediction set of the target disease is as follows: Record the patient diagnosed with the target disease as the target patient, obtain the target patient's medical information from the hospital's electronic medical record system, obtain the target patient's personal information and diagnosis and treatment information based on the target patient's medical information, and then obtain the target patient's index information, match the target patient's index information with each index information in the target disease's associated disease prediction set, and thereby obtain the target disease's associated disease prediction subset; Obtaining various medical imaging images of the target patient based on the target patient's medical information. When the various medical imaging images of the target patient are all medical imaging images of the target disease, it indicates that the target patient has not been diagnosed with various related diseases. When the medical imaging images of the target patient include the medical imaging image of the target disease and the medical imaging images of the associated diseases, it indicates that the target patient has been diagnosed with the associated diseases; The specific process of predicting the target patient's associated diseases and the probability of diagnosis of the associated diseases based on the matching results is as follows: When the target patient has not yet been diagnosed with each associated disease, the first-level confirmed probability and the confirmed probability of the disease in the target patient's associated disease prediction subset are obtained based on the associated disease prediction set of the target disease, and the result obtained by multiplying the first-level confirmed probability of each associated disease by the confirmed probability of the disease is recorded as the probability of the target patient suffering from each associated infectious disease; When the target patient has been diagnosed with various related diseases, the diagnosis level of each confirmed related disease of the target patient is obtained based on the target patient's medical information analysis, and the next diagnosis level of each confirmed related disease of the target patient is recorded as the target prediction level, and the related diseases included in the target prediction level in the target patient's related disease prediction subset are obtained, and then the related diseases in which the target patient is diagnosed among the related diseases included in the target prediction level are obtained, and recorded as target predicted related diseases, and the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease are obtained, and the product of the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease is recorded as the disease probability of each target predicted related disease.

2. The method for analyzing medical image big data according to claim 1, characterized in that: The analysis obtains a set of associated disease predictions for the target disease, and the specific process is as follows: Recording the categories of various target medical image images as index information of target medical image images, and storing each target medical image image in the corresponding index information to obtain an index information set of the target disease; Obtaining each correlation level of each associated disease in each index in the index information set, analyzing each correlation level of each associated disease of each target medical image in each index to obtain the diagnosis probability of each associated disease and the diagnosis probability of each level in each index information set, and comprehensively recording the diagnosis probability of each level of each associated disease, the probability of the disease, and the corresponding associated medical image corresponding to each index as a subset of each index; The associated disease prediction set of the target disease is constructed by integrating the subsets of each index and the index information set of the target disease.

3. The method for analyzing medical image big data according to claim 2, characterized in that: The analysis of the association levels of the associated diseases of the target medical images in the indexes is based on the following steps to obtain the diagnosis probability of the associated diseases and the diagnosis probability of each level in the index information set: Count the total number of historical patients in each index information set, recorded as , where i represents the number of each index information set, i is a positive integer, and the number of confirmed cases of each associated disease at each associated level in each index information set is counted, recorded as , where j represents the number of each associated disease, j is a positive integer, and k is the number of the level of the associated disease, k is a positive integer, according to the calculation formula: The analysis obtains the diagnosis probability of each associated disease at each level in each index information set; Count the total number of confirmed cases of each associated disease in each index information set, recorded as , according to the calculation formula: Analyze and obtain the probability of diagnosis of each associated disease in each index information set , where J is the total number of associated diseases.

4. A medical image big data analysis system that executes the medical image big data analysis method according to any one of claims 1 to 3, characterized in that: include: Target disease information acquisition module: used to obtain each historical target image in each hospital in the target area, and then obtain the medical information of each historical patient corresponding to each historical target image; The specific process of obtaining target images in each hospital in the target area is as follows: Obtain any area from a satellite remote sensing system, record it as a target area, then obtain each hospital in the target area, record it as each target hospital, and record any disease diagnosed with the assistance of medical images as a target disease, obtain each historical medical image of the target disease from the electronic medical record system of each target hospital, record it as each target medical image, and obtain each related disease of the target disease and each medical image of each related disease from the electronic medical record system of each target hospital; The medical information of each historical patient includes diagnosis and treatment information and personal information. The diagnosis and treatment information includes each associated disease of each historical patient, each medical image of each associated disease, and the diagnosis timestamp of each associated disease. The personal information includes gender and age type. The diagnosis timestamp of each associated disease is the time between the diagnosis time of each associated disease and the diagnosis time of the target disease. Target disease information analysis module: used to classify each medical image based on the medical information of each historical patient, and then analyze and obtain the associated disease prediction set of the target disease; The specific process of classifying each medical image based on the medical information of each historical patient is as follows: Obtaining personal information of each historical patient based on the medical information of each historical patient, obtaining the gender and age type of each historical patient based on the personal information of each historical patient, and classifying each target medical image based on the gender and age type of each historical patient to obtain various target medical image images; Based on the medical information of each historical patient, the diagnosis and treatment information of each historical patient is obtained; based on the diagnosis and treatment information of each historical patient, each related disease of each historical patient and the diagnosis timestamp of each related disease are obtained; and then the minimum diagnosis timestamp of the related disease of each historical patient is obtained, thereby obtaining the related disease corresponding to the minimum diagnosis timestamp of the related disease of each historical patient; the related disease corresponding to the minimum diagnosis timestamp of the related disease of a certain historical patient is recorded as the first-level related disease of the historical patient; and then the minimum diagnosis timestamp of the related disease among the diagnosis timestamps of the remaining related diseases of the historical patient is obtained, thereby obtaining the related disease corresponding to the minimum timestamp of the related disease, and recording the related disease as the second-level related disease of the historical patient; accordingly, the related diseases of each historical patient at all levels are obtained; and the medical imaging images of each related disease are classified based on the related diseases of each historical patient at all levels to obtain related medical imaging images of each level of each target medical imaging image; Related disease information prediction module: used to match the medical imaging images of the target patient with the related disease prediction set of the target disease, and predict the related diseases of the target patient and the diagnosis probability of the related diseases based on the matching results; The specific process of matching the medical imaging images of the target patient with the associated disease prediction set of the target disease is as follows: Record the patient diagnosed with the target disease as the target patient, obtain the target patient's medical information from the hospital's electronic medical record system, obtain the target patient's personal information and diagnosis and treatment information based on the target patient's medical information, and then obtain the target patient's index information, match the target patient's index information with each index information in the target disease's associated disease prediction set, and thereby obtain the target disease's associated disease prediction subset; Obtaining various medical imaging images of the target patient based on the target patient's medical information. When the various medical imaging images of the target patient are all medical imaging images of the target disease, it indicates that the target patient has not been diagnosed with various related diseases. When the medical imaging images of the target patient include the medical imaging image of the target disease and the medical imaging images of the associated diseases, it indicates that the target patient has been diagnosed with the associated diseases; The specific process of predicting the target patient's associated diseases and the probability of diagnosis of the associated diseases based on the matching results is as follows: When the target patient has not yet been diagnosed with each associated disease, the first-level confirmed probability and the confirmed probability of the disease in the target patient's associated disease prediction subset are obtained based on the associated disease prediction set of the target disease, and the result obtained by multiplying the first-level confirmed probability of each associated disease by the confirmed probability of the disease is recorded as the probability of the target patient suffering from each associated infectious disease; When the target patient has been diagnosed with various related diseases, the diagnosis level of each confirmed related disease of the target patient is obtained based on the target patient's medical information analysis, and the next diagnosis level of each confirmed related disease of the target patient is recorded as the target prediction level, and the related diseases included in the target prediction level in the target patient's related disease prediction subset are obtained, and then the related diseases in which the target patient is diagnosed among the related diseases included in the target prediction level are obtained, and recorded as target predicted related diseases, and the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease are obtained, and the product of the level diagnosis probability and the disease diagnosis probability of the target prediction level of each target predicted related disease is recorded as the disease probability of each target predicted related disease.

Images