Image video acquisition data intelligent classification management system based on artificial intelligence

Abstract

The invention discloses an image video acquisition data intelligent classification management system based on artificial intelligence, and belongs to the technical field of medical image processing, the system comprises a medical image acquisition module, and the medical image acquisition module comprises an image acquisition unit and a preprocessing unit; the image acquisition unit is used for acquiring medical image data; the preprocessing unit responds to the image acquisition unit and is used for preprocessing the acquired medical image data, and preprocessing of the medical image data comprises denoising, contrast enhancement and format conversion; and the feature extraction module is used for performing feature extraction on the preprocessed medical image data, and feature extraction comprises extraction of feature vectors of the medical image data. Through the functions of automatic feature extraction, intelligent classification, dynamic optimization, data fusion management and the like, the classification efficiency and accuracy of the medical image data are improved, the data management capability is enhanced, and powerful support is provided for diagnosis and prognosis of diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to an intelligent classification and management system for image and video acquisition data based on artificial intelligence. Background Art

[0002] With the rapid development of information technology, the amount of image and video data collected has exploded. This data is widely used in security monitoring, intelligent transportation, medical imaging, entertainment and other fields, and is also classified and managed in the corresponding fields.

[0003] Regarding this research, application document CN202311389989.5 provides an AI-based image classification system. This technical solution includes a first image acquisition unit that acquires uploaded mammography images to be classified; a first image preprocessing unit that divides the mammography images to be classified into target regions to obtain target region images corresponding to the mammography images; and an AI model unit that inputs the mammography images to be classified and the target region images into a trained AI model to obtain classification results for the mammography images to be classified. This technical solution addresses the technical problem of low classification accuracy when classifying mammography images using high-level features extracted from mammography images via convolutional neural networks.

[0004] Another application document with application number CN202011158352.1 provides an artificial intelligence-based medical image classification and processing system. This technical solution obtains the image information of the current patient's lesion through a camera device, labels the uploaded image data through this system, and uploads it to the terminal server to achieve data expansion and sharing. At the same time, the patient's image information is compared with the image information stored in the database, and matched with similar case image information. The medical imaging data taken by the patient can be accurately analyzed and a medical examination report can be made, thereby assisting doctors to improve the efficiency and accuracy of medical examinations on patients.

[0005] However, the above technical solutions ignore the noise, artifacts and other uncertainties that may exist in medical images. These factors will affect doctors' annotation of patients' disease types, lesion sites, and textures, shapes and densities associated with disease types and lesion sites, leading to errors in the annotation of medical images, which in turn makes it difficult to subsequently classify and retrieve massive amounts of medical images, affecting clinical diagnosis efficiency. Summary of the Invention

[0006] In view of the above problems existing in the technical field of existing medical image processing, the present invention is proposed.

[0007] Therefore, one of the purposes of the present invention is to provide an intelligent classification and management system for image and video acquisition data based on artificial intelligence, which improves the classification efficiency and accuracy of medical imaging data through functions such as automated feature extraction, intelligent classification, dynamic optimization and data fusion management, enhances data management capabilities, and provides strong support for disease diagnosis and prognosis.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] The present invention provides an artificial intelligence-based intelligent classification and management system for image and video acquisition data, comprising:

[0010] A medical image acquisition module, comprising an image acquisition unit and a preprocessing unit;

[0011] The image acquisition unit is used to acquire medical image data;

[0012] The preprocessing unit responds to the image acquisition unit and is used to preprocess the acquired medical image data, wherein the preprocessing of the medical image data includes denoising, contrast enhancement and format conversion;

[0013] A feature extraction module is used to extract features from the preprocessed medical image data, wherein the feature extraction includes extracting feature vectors of the medical image data, and the feature vectors include texture features, shape features, and grayscale features of the image in the medical image data;

[0014] a data processing module, the data processing module being responsive to the feature vectors, for constructing a database based on the extracted feature vectors, classifying the medical imaging data in the database based on different feature vectors, and generating a classification model;

[0015] A data fusion management module, comprising a training unit, a classification unit, and a learning unit;

[0016] The training unit is used to train the classification model according to different feature vectors, and the training method includes labeling the medical image data in the classification model, and the information labeling includes labeling the disease type and lesion location of the medical image data;

[0017] The classification unit is used to input the extracted feature vector into the trained classification model to automatically classify the medical imaging data;

[0018] The learning unit is used to optimize the classification model based on the doctor's feedback and newly collected medical imaging data.

[0019] As a preferred solution of the present invention, wherein: in the learning unit, the doctor's feedback includes the accuracy of the classification result, and the accuracy of the classification result includes correctness feedback and misclassification feedback;

[0020] The correctness feedback indicates to the doctor whether the classification result of the classification model is correct;

[0021] The misclassification feedback points out the misclassification of the classification model to the doctor;

[0022] The doctor's feedback also includes identification of key areas, which includes feedback on lesion areas and feature importance.

[0023] The lesion area feedback indicates to the doctor whether the classification model accurately identifies the lesion area;

[0024] The feature importance feedback is for doctors to point out key features and redundant features of medical imaging data in the classification model, and the key features include texture, shape and density.

[0025] As a preferred solution of the present invention, wherein: in the data processing module, the medical image data is classified according to different feature vectors, including dividing the medical image data into in, Representing the nth category of divided medical imaging data, obtaining the difference in the feature vector in different categories of medical imaging data, and extracting the feature vector of the newly collected medical imaging data in the future time period, comparing the feature vector with the feature vector in the classified medical imaging data, and inputting the corresponding feature vector into the classification model according to the comparison result to classify the collected medical imaging data, and when retrieving medical imaging data in the classification model in the future time period, inputting the feature vector into the classification model to obtain the medical imaging data corresponding to the feature vector.

[0026] As a preferred solution of the present invention, wherein: the characteristic vectors of the medical imaging data are distinguished in the classification model, and the distinguishing method includes distinguishing the characteristic vectors into early characteristic vectors, mid-term characteristic vectors and late characteristic vectors according to the acquisition time of the medical imaging data; wherein the characteristic change from the early characteristic vector to the mid-term characteristic vector is marked as θ1 stage change, and the characteristic change from the mid-term characteristic vector to the late characteristic vector is marked as θ2 stage change; and the disease type, lesion site and symptoms corresponding to the early characteristic vector are obtained in the classification model, and in the medical imaging data newly collected in the future time period, if the characteristic vector corresponding to the medical imaging data is consistent with the disease type, lesion site and symptoms in the early characteristic vector, the system determines that the characteristic vector of the newly collected medical imaging data will change toward the mid-term characteristic vector; otherwise, it will not be judged.

[0027] As a preferred solution of the present invention, wherein: the characteristic vectors of the medical imaging data are distinguished in the classification model, and the distinguishing method also includes distinguishing the characteristic vectors into mild characteristic vectors, moderate characteristic vectors and severe characteristic vectors according to the degree of the lesion; wherein the characteristic change of the mild characteristic vector to the moderate characteristic vector is marked as a δ1 stage change, and the characteristic change of the mid-term characteristic vector to the late characteristic vector is marked as a δ2 stage change; and the disease type, lesion site and symptoms corresponding to the mild characteristic vector are obtained in the classification model. In the newly collected medical imaging data in the future time period, if the characteristic vector corresponding to the medical imaging data meets the disease type, lesion site and symptoms in the mild characteristic vector, the system determines that the characteristic vector of the newly collected medical imaging data will change toward the moderate characteristic vector; otherwise, it will not be judged.

[0028] As a preferred solution of the present invention, according to the result of the system judgment, if the feature vector of the newly collected medical imaging data in the future time period does not change towards the medium-term feature vector or the moderate feature vector, the newly collected medical imaging data is newly classified in the classification model; otherwise, it is not performed.

[0029] As a preferred embodiment of the present invention, when the system determines that the feature vector of the newly acquired medical image data will change toward the mid-term feature vector, the degree of change of the early feature vector at different time points during the period of change toward the mid-term feature vector is calculated according to the following formula:

[0030] Δ v =v 中期 -v 前期 ; Among them, v 前期 represents the early feature vector;

[0031] Where, v 中期 represents the mid-term eigenvector, Δv represents the change vector of the eigenvector;

[0032] The change amplitude of the eigenvector is calculated based on the change vector. The change amplitude of the eigenvector is used to measure the change size of the eigenvector at different time points, as follows:

[0033] ||Δ v ||=||Δ 中期 -Δ 前期 ||;

[0034] Where, ||Δ v || represents the Euclidean norm of the change of the eigenvector, which is used to quantify the magnitude of the change.

[0035] As a preferred solution of the present invention, the change rate of the characteristic vector is calculated according to the change size, and the change rate is the speed of change of the characteristic vector per unit time, as follows:

[0036]

[0037] Where, Δ t =t 中期 -t 前期 represents the time interval, and r represents the rate of change of the feature vector.

[0038] As a preferred solution of the present invention, a change threshold is preset according to the degree of change of the calculated early feature vector toward the mid-term feature vector at different time points, and a first time point, a second time point and a third time point are preset within the period based on the change threshold. At any time point, if the degree of change of the early feature vector toward the mid-term feature vector exceeds the change threshold, the system performs a new classification on the medical imaging data corresponding to the data exceeding the change threshold; otherwise, it does not perform the classification.

[0039] Beneficial effects:

[0040] 1. The feature extraction module automatically extracts the texture, shape, and grayscale features of medical images, reducing the workload of manual annotation and improving data processing efficiency. The classification model automatically classifies the extracted feature vectors, enabling rapid and accurate classification of large amounts of medical image data, thus improving diagnostic efficiency.

[0041] 2. The learning unit optimizes the classification model based on doctors' feedback (such as the accuracy of classification results, identification of key areas, etc.) and newly collected medical imaging data, enabling the model to continuously adapt to new data and clinical needs, further improving classification accuracy;

[0042] 3. The data processing module classifies medical imaging data according to feature vectors and builds a database, making it easier for doctors to quickly retrieve and access specific types of imaging data. This improves the convenience and efficiency of data management. Furthermore, by comparing feature vectors, it can quickly classify newly collected medical imaging data, ensuring orderly storage and management of data.

[0043] 4. The system can dynamically adjust the classification criteria according to the preset change threshold to ensure timely reclassification of imaging data during disease progression, thereby improving the flexibility and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0045] Figure 1 Schematic diagram of the modular structure of an intelligent classification and management system for image and video acquisition data based on artificial intelligence according to an embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of the process structure of an embodiment of the present invention;

[0047] Numbers in the figure: 110 - medical image acquisition module; 1101 - image acquisition unit; 1102 - preprocessing unit; 120 - feature extraction module; 130 - data processing module; 140 - data fusion management module; 1401 - training unit; 1402 - classification unit; 1403 - learning unit. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0049] Because existing technologies ignore the noise, artifacts and other uncertainties that may exist in medical images, these factors will affect doctors' annotation of patients' disease types, lesion locations, and the textures, shapes and densities associated with disease types and lesion locations, leading to errors in the annotation of medical images, which in turn makes it difficult to subsequently classify and retrieve massive amounts of medical images, affecting clinical diagnosis efficiency.

[0050] Based on this, the present invention proposes an artificial intelligence-based intelligent classification and management system for image and video acquisition data. Through functions such as automated feature extraction, intelligent classification, dynamic optimization, and data fusion management, it improves the classification efficiency and accuracy of medical imaging data, enhances data management capabilities, and provides strong support for disease diagnosis and prognosis.

[0051] The present invention will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0052] Reference Figures 1 to 2 , is an embodiment of the present invention, which provides an intelligent classification and management system for image and video acquisition data based on artificial intelligence, including:

[0053] The medical image acquisition module 110 includes an image acquisition unit 1101 and a pre-processing unit 1102;

[0054] The image acquisition unit 1101 is used to acquire medical image data;

[0055] The pre-processing unit 1102 responds to the image acquisition unit and is used to pre-process the acquired medical image data. The pre-processing of the medical image data includes denoising, contrast enhancement and format conversion.

[0056] A feature extraction module 120 is used to extract features from the preprocessed medical image data, wherein the feature extraction includes extracting feature vectors of the medical image data, and the feature vectors include texture features, shape features, and grayscale features of the image in the medical image data;

[0057] The data processing module 130 is responsive to the feature vectors and is used to construct a database based on the extracted feature vectors, classify the medical imaging data in the database based on different feature vectors, and generate a classification model;

[0058] The data fusion management module 140 includes a training unit 1401, a classification unit 1402 and a learning unit 1403;

[0059] The training unit 1401 is used to train the classification model according to different feature vectors. The training method includes labeling the medical image data in the classification model. The information labeling includes labeling the disease type and lesion location of the medical image data.

[0060] The classification unit 1402 is used to input the extracted feature vector into the trained classification model to automatically classify the medical imaging data;

[0061] In this embodiment, the classification result can be a specific disease type or a disease probability distribution;

[0062] The learning unit 1403 is used to optimize the classification model based on the doctor's feedback and the newly collected medical imaging data;

[0063] In this embodiment, a complete AI-based intelligent classification and management system for medical imaging data is constructed, covering the entire process from image acquisition, preprocessing, feature extraction to data processing and model optimization, which can meet the needs of medical imaging data from acquisition to classification management;

[0064] In the learning unit, the doctor's feedback includes the accuracy of the classification results, and the accuracy of the classification results includes correctness feedback and misclassification feedback;

[0065] Correctness feedback, indicating to doctors whether the classification results of the classification model are correct;

[0066] For example, the model classifies an image as "normal", but the doctor finds that the image actually has a lesion through diagnosis;

[0067] Misclassification feedback, which points out misclassifications made by the classification model to doctors;

[0068] For example, misclassifying a benign lesion as malignant, or vice versa, this feedback is particularly important for adjusting the decision boundary of the classification model;

[0069] Physician feedback also includes identification of key areas, which includes feedback on lesion areas and feature importance.

[0070] Lesion area feedback indicates to doctors whether the classification model accurately identifies the lesion area;

[0071] For example, in lung CT images, whether the model accurately identifies the location and extent of lung nodules;

[0072] Feature importance feedback, which helps doctors identify key and redundant features of medical imaging data in classification models. Key features include texture, shape, and density.

[0073] In this embodiment, the professional feedback of doctors is incorporated into the optimization process of the classification model, so that the model can be adjusted based on clinical experience and actual needs;

[0074] Physicians provide feedback on the correctness and misclassification of classification results, which helps to correct model errors in a timely manner and improve classification accuracy;

[0075] Doctors also point out the importance of lesion areas and key features, enabling the model to more accurately identify and focus on imaging features that are important for diagnosis, avoiding interference from redundant information;

[0076] In the data processing module, the medical imaging data is classified according to different feature vectors, including dividing the medical imaging data into in, Representing the nth category of medical image data, obtaining differences in feature vectors between different categories of medical image data, extracting feature vectors from newly collected medical image data in a future time period, comparing the feature vectors with feature vectors in the classified medical image data, and inputting corresponding feature vectors into a classification model based on the comparison results to classify the collected medical image data. When retrieving medical image data from the classification model in a future time period, inputting the feature vectors into the classification model to obtain medical image data corresponding to the feature vectors.

[0077] This embodiment clarifies the specific method of classifying medical imaging data based on feature vectors, as well as the method for processing newly collected data in the future. By dividing medical imaging data into different categories and obtaining the feature vector differences between different categories, the data classification is made clearer and more specific, making it easier to manage and retrieve.

[0078] By extracting and comparing feature vectors of newly collected medical imaging data in the future, it can be promptly classified into appropriate categories, ensuring the timeliness and accuracy of the classification model;

[0079] During retrieval, the corresponding medical imaging data can be obtained by inputting the feature vector, which improves the retrieval efficiency and facilitates doctors to quickly access the required imaging data;

[0080] It should be emphasized in this embodiment that the feature vectors of the medical imaging data are differentiated in the classification model. The differentiation method includes distinguishing the feature vectors into early feature vectors, mid-term feature vectors, and late feature vectors according to the acquisition time of the medical imaging data; wherein the feature change from the early feature vector to the mid-term feature vector is marked as a θ1 stage change, and the feature change from the mid-term feature vector to the late feature vector is marked as a θ2 stage change; and the disease type, lesion site, and symptoms corresponding to the early feature vector are obtained in the classification model. In the medical imaging data newly acquired in the future time period, if the feature vector corresponding to the medical imaging data meets the disease type, lesion site, and symptoms in the early feature vector, the system determines that the feature vector of the newly acquired medical imaging data will change toward the mid-term feature vector; otherwise, no determination is made;

[0081] In this embodiment, by introducing the time dimension, the feature vector is distinguished into early, middle and late stages, which can better reflect the development process of the disease;

[0082] By marking the stage changes, the evolution of the disease from the early stage to the middle stage and then to the late stage can be clearly seen, providing doctors with dynamic information about the disease;

[0083] Furthermore, the system can predict whether newly acquired imaging data will change toward the mid-term eigenvector based on the disease type, lesion location, and symptoms corresponding to the early eigenvector, thereby making an early judgment on the development trend of the disease and facilitating early intervention and treatment.

[0084] This distinction has important clinical significance for subsequent judgment of the patient's disease course;

[0085] Early feature vectors: In the early stages of the disease, imaging manifestations may be mild or atypical. For example, in early lung cancer, CT images may only show tiny nodules. The feature vectors mainly reflect the small size and edge clarity of the nodules.

[0086] Mid-stage eigenvectors: As the disease progresses, imaging manifestations become increasingly apparent. For example, in mid-stage lung cancer, CT images may show enlarged nodules, irregular margins, and ground-glass opacities. The eigenvectors will reflect these changes.

[0087] Late-stage feature vectors: In the late stages of the disease, the imaging manifestations are more severe. For example, in advanced lung cancer, CT images may show large tumors, lymph node metastases, pleural effusions, etc. The feature vectors will contain these complex features;

[0088] On the basis of the above, the feature vectors of the medical imaging data are distinguished in the classification model. The distinguishing method also includes distinguishing the feature vectors into mild feature vectors, moderate feature vectors and severe feature vectors according to the degree of the lesion; wherein the feature change from the mild feature vector to the moderate feature vector is marked as a δ1 stage change, and the feature change from the intermediate feature vector to the late feature vector is marked as a δ2 stage change; and the disease type, lesion site and symptoms corresponding to the mild feature vector are obtained in the classification model. In the medical imaging data newly collected in the future time period, if the feature vector corresponding to the medical imaging data meets the disease type, lesion site and symptoms in the mild feature vector, the system determines that the feature vector of the newly collected medical imaging data will change towards the moderate feature vector; otherwise, no determination is made;

[0089] In this embodiment, the classification dimension of the feature vector is further refined, and the feature vector is differentiated from the perspective of the degree of lesion, which enriches the description of the disease state;

[0090] It helps to classify and manage patients with different disease severity, improve the utilization efficiency of medical resources, and optimize the disease management process;

[0091] Mild eigenvectors: When the lesion is mild, the image features are relatively simple. For example, in early cerebral infarction, MRI images may only show mild edema of local brain tissue, and the eigenvectors mainly reflect the range and density of the edema area.

[0092] Moderate eigenvectors: As the lesion becomes more severe, the imaging features become more obvious. For example, in mid-term cerebral infarction, MRI images may show a large area of ​​brain tissue edema, local brain atrophy, etc., and the eigenvectors will reflect these changes;

[0093] Severe eigenvectors: When the lesion is severe, the image features are complex and diverse. For example, in late-stage cerebral infarction, MRI images may show large areas of brain tissue necrosis, ventricular deformation, widened cerebral sulci, etc. The eigenvectors will contain these complex features;

[0094] Distinguishing eigenvectors has advantages in judging disease trends, including dynamic monitoring of disease progression, predicting disease prognosis, assisting clinical decision-making, and improving diagnostic accuracy;

[0095] Furthermore, according to the result of the system determination, if the feature vector of the newly collected medical image data in the future period does not change towards the medium-term feature vector or the moderate feature vector, the newly collected medical image data is newly classified in the classification model; otherwise, no classification is performed;

[0096] In this embodiment, the processing method when the newly collected medical imaging data does not conform to the expected change trend is clarified, ensuring the accuracy and stability of the classification model;

[0097] If the newly collected feature vector does not change towards the expected medium or moderate feature vector, the data will be reclassified, avoiding classification errors caused by misjudgment and ensuring the reliability of the classification results;

[0098] This mechanism enables the classification model to correct the impact of abnormal data in a timely manner, maintain the stability and adaptability of the model, and avoid the degradation of model performance due to interference from individual data;

[0099] Based on the above, when the system determines that the eigenvector of the newly acquired medical imaging data will change towards the mid-term eigenvector, the degree of change of the early eigenvector towards the mid-term eigenvector at different time points is calculated according to the following formula:

[0100] Δ v =v 中期 -v 前期 ; Among them, v 前期 represents the early feature vector;

[0101] Where, v 中期 represents the mid-term eigenvector, Δ v represents the change vector of the eigenvector;

[0102] The change amplitude of the eigenvector is calculated based on the change vector. The change amplitude of the eigenvector is a measure of the change size of the eigenvector at different time points, as follows:

[0103] ||Δ v ||=||Δ 中期 -Δ 前期 ||;

[0104] Where, ||Δ v || represents the Euclidean norm of the change in the eigenvector, used to quantify the magnitude of the change;

[0105] In this embodiment, a specific calculation method is provided to quantify the degree of change of the early stage characteristic vector to the mid-term characteristic vector, so that the change process can be quantified and compared;

[0106] By calculating the change vector and change amplitude of the characteristic vector, the characteristic change from the early stage to the middle stage is expressed in numerical form, which is convenient for quantitative analysis of the change process;

[0107] The magnitude of changes in different patients or different diseases can be compared, which helps to discover the patterns and differences in disease progression and provide data support for clinical research;

[0108] Furthermore, the change rate of the eigenvector is calculated based on the change size. The change rate is the speed at which the eigenvector changes in unit time, as follows:

[0109]

[0110] Where, Δ t =t 中期 -t 前期 represents the time interval, r represents the rate of change of the eigenvector;

[0111] In this embodiment, the speed of the characteristic vector change is further quantified, providing a more accurate indicator for the assessment of the disease progression rate;

[0112] Presetting a change threshold value based on the calculated degree of change of the early stage feature vector toward the mid-term feature vector at different time points, and presetting a first time point, a second time point, and a third time point within the period based on the change threshold value. At any time point, if the degree of change of the early stage feature vector toward the mid-term feature vector exceeds the change threshold value, the system performs a new classification on the medical imaging data corresponding to the exceeded change threshold value; otherwise, no new classification is performed;

[0113] In this embodiment, by introducing a change threshold and a preset time point, dynamic monitoring and classification adjustment of the degree of change of the feature vector are achieved;

[0114] By presetting change thresholds and time points, the system can monitor changes in characteristic vectors in real time. Once the degree of change exceeds the threshold, it will immediately issue an early warning and make a new classification, thereby improving sensitivity and response speed to disease changes.

[0115] In summary, this application improves the classification efficiency and accuracy of medical imaging data through functions such as automated feature extraction, intelligent classification, dynamic optimization, and data fusion management, enhances data management capabilities, and provides strong support for disease diagnosis and prognosis.

[0116] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent classification and management system for image and video acquisition data based on artificial intelligence, characterized by: include: A medical image acquisition module, comprising an image acquisition unit and a preprocessing unit; The image acquisition unit is used to acquire medical image data; The preprocessing unit responds to the image acquisition unit and is used to preprocess the acquired medical image data, wherein the preprocessing of the medical image data includes denoising, contrast enhancement and format conversion; A feature extraction module is used to extract features from the preprocessed medical image data, wherein the feature extraction includes extracting feature vectors of the medical image data, and the feature vectors include texture features, shape features, and grayscale features of the image in the medical image data; a data processing module, the data processing module being responsive to the feature vectors, for constructing a database based on the extracted feature vectors, classifying the medical imaging data in the database based on different feature vectors, and generating a classification model; A data fusion management module, comprising a training unit, a classification unit, and a learning unit; The training unit is used to train the classification model according to different feature vectors, and the training method includes labeling the medical image data in the classification model, and the information labeling includes labeling the disease type and lesion location of the medical image data; The classification unit is used to input the extracted feature vector into the trained classification model to automatically classify the medical imaging data; The learning unit is used to optimize the classification model based on the doctor's feedback and newly collected medical imaging data.

2. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 1, characterized in that: In the learning unit, the doctor's feedback includes the accuracy of the classification results, and the accuracy of the classification results includes correctness feedback and misclassification feedback; The correctness feedback indicates to the doctor whether the classification result of the classification model is correct; The misclassification feedback points out the misclassification of the classification model to the doctor; The doctor's feedback also includes identification of key areas, which includes feedback on lesion areas and feature importance. The lesion area feedback indicates to the doctor whether the classification model accurately identifies the lesion area; The feature importance feedback is for doctors to point out key features and redundant features of medical imaging data in the classification model, and the key features include texture, shape and density.

3. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 1, characterized in that: In the data processing module, the medical image data is classified according to different feature vectors, including dividing the medical image data into ▽1, ▽2, ..., ▽ n , where ▽ n Representing the nth category of divided medical imaging data, obtaining the difference in the feature vector in different categories of medical imaging data, and extracting the feature vector of the newly collected medical imaging data in the future time period, comparing the feature vector with the feature vector in the classified medical imaging data, and inputting the corresponding feature vector into the classification model according to the comparison result to classify the collected medical imaging data, and when retrieving medical imaging data in the classification model in the future time period, inputting the feature vector into the classification model to obtain the medical imaging data corresponding to the feature vector.

4. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 3, characterized in that: The characteristic vectors of the medical imaging data are distinguished in the classification model, and the distinguishing method includes distinguishing the characteristic vectors into early characteristic vectors, mid-term characteristic vectors and late characteristic vectors according to the acquisition time of the medical imaging data; wherein the characteristic change from the early characteristic vector to the mid-term characteristic vector is marked as θ1 stage change, and the characteristic change from the mid-term characteristic vector to the late characteristic vector is marked as θ2 stage change; and the disease type, lesion site and symptoms corresponding to the early characteristic vector are obtained in the classification model. In the medical imaging data newly collected in the future time period, if the characteristic vector corresponding to the medical imaging data is consistent with the disease type, lesion site and symptoms in the early characteristic vector, the system determines that the characteristic vector of the newly collected medical imaging data will change toward the mid-term characteristic vector; otherwise, no determination is made.

5. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 4, characterized in that: The characteristic vectors of the medical imaging data are distinguished in the classification model, and the distinguishing method also includes distinguishing the characteristic vectors into mild characteristic vectors, moderate characteristic vectors and severe characteristic vectors according to the degree of the lesion; wherein the characteristic change from the mild characteristic vector to the moderate characteristic vector is marked as a δ1 stage change, and the characteristic change from the mid-term characteristic vector to the late characteristic vector is marked as a δ2 stage change; and the disease type, lesion site and symptoms corresponding to the mild characteristic vector are obtained in the classification model. In the newly collected medical imaging data in the future time period, if the characteristic vector corresponding to the medical imaging data meets the disease type, lesion site and symptoms in the mild characteristic vector, the system determines that the characteristic vector of the newly collected medical imaging data will change toward the moderate characteristic vector; otherwise, no determination is made.

6. The intelligent classification and management system for image and video acquisition data based on artificial intelligence according to any one of claims 4 to 5, characterized in that: According to the result of the system judgment, if the feature vector of the newly collected medical imaging data in the future period does not change towards the medium-term feature vector or the moderate feature vector, the newly collected medical imaging data will be newly classified in the classification model; otherwise, it will not be performed.

7. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 4, characterized in that: When the system determines that the feature vector of the newly acquired medical image data will change toward the mid-term feature vector, the degree of change of the early feature vector at different time points during the period of change toward the mid-term feature vector is calculated according to the following formula: Δ v =v 中期 -v 前期 ; Among them, v 前期 represents the early feature vector; Where, v 中期 represents the mid-term eigenvector, Δ v represents the change vector of the eigenvector; The change amplitude of the eigenvector is calculated based on the change vector. The change amplitude of the eigenvector is used to measure the change size of the eigenvector at different time points, as follows: ||D v ||=||D 中期 -D 前期 ||; Where, ||Δ v || represents the Euclidean norm of the change of the eigenvector, which is used to quantify the magnitude of the change.

8. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 7, characterized in that: The change rate of the feature vector is calculated based on the change size. The change rate is the speed at which the feature vector changes in unit time, as follows: Where, Δ t =t 中期 -t 前期 represents the time interval, and r represents the rate of change of the feature vector.

9. The image and video acquisition data intelligent classification and management system based on artificial intelligence according to claim 7, characterized in that: A change threshold is preset according to the degree of change of the calculated early feature vector toward the mid-term feature vector at different time points, and a first time point, a second time point and a third time point are preset within the period based on the change threshold. At any time point, if the degree of change of the early feature vector toward the mid-term feature vector exceeds the change threshold, the system will perform a new classification on the medical imaging data corresponding to the data exceeding the change threshold; otherwise, it will not be performed.

Images