An AI-based medical image intelligent teaching platform

Abstract

The application relates to the technical field of medical teaching, in particular to an AI-based medical image intelligent teaching platform, which comprises an image acquisition unit, a question storage unit, a statistical unit, a comparison unit and an analysis unit. Whether the storage data of a corresponding sub-class question is qualified is determined based on a contour difference representation value, and when it is determined that the storage data of a single sub-class question is abnormal, the storage data in the image acquisition unit is adjusted based on a disease detail quantitative parameter and an image transverse comparison parameter, including adjusting a preset acquisition period of the image acquisition unit to obtain medical images to a corresponding value, adjusting the storage amount of medical images for a single sub-class question to a corresponding value, or adjusting the definition of medical images for a single sub-class question to a corresponding value. Whether the storage data of each sub-class question is qualified is monitored based on the analysis result of abnormal return data of a user, and the storage data in the image acquisition unit is adjusted according to the analysis result, so that the management efficiency of teaching resources is improved.

Description

An AI-based intelligent teaching platform for medical imaging Technical Field

[0001] This invention relates to the field of medical teaching technology, and in particular to an AI-based intelligent teaching platform for medical imaging. Background Technology

[0002] In medical schools, students need to learn a large amount of medical imaging to master the diagnostic methods and characteristics of various diseases; within hospitals, newly hired doctors or doctors who need to improve their professional skills need to continuously learn the latest medical imaging knowledge to improve their diagnostic level; in distance medical education, medical learners in different regions can access unified and high-quality medical imaging teaching resources through this platform to improve their own medical literacy.

[0003] Traditional medical imaging teaching relies primarily on printed textbooks, PowerPoint presentations, and oral lectures. Teachers demonstrate typical medical imaging cases in class, explaining the characteristics of the diseases and diagnostic methods. Students learn by viewing these images and listening to explanations. Simultaneously, students also undertake internships in hospital radiology departments, observing actual medical imaging diagnostic processes.

[0004] With the development of computer technology, some medical imaging teaching software has emerged. This software can store a certain number of medical images and provide basic image browsing and annotation functions. Teachers can use this software for teaching demonstrations, and students can also operate the software independently for learning. Some software also has simple testing functions to assess students' learning outcomes.

[0005] Chinese Patent Publication No. CN109658765B discloses a digital medical imaging software teaching service system, including a server, a teacher terminal, and a student terminal connected via the Internet. The teacher terminal includes a human-computer operation module, a teaching video recording module, an image processing path mining module, an image mining module, and an image processing path integration module. The student terminal includes a human-computer operation module, a teaching video playback module, an image processing path playback module, and a problem recording module. It is evident that the above technical solution has the following problems: it does not consider monitoring the qualification of stored data in each subcategory based on the analysis results of abnormal user feedback data, nor does it consider adjusting the stored data in the image acquisition module based on the analysis results, thus affecting the management efficiency of teaching resources. Summary of the Invention

[0006] To address this, the present invention provides an AI-based intelligent teaching platform for medical imaging, which overcomes the problems in existing technologies that fail to consider monitoring the qualification of stored data in each subcategory based on the analysis results of abnormal user feedback data, and fail to consider adjusting the stored data in the image acquisition unit according to the analysis results, thus affecting the efficiency of teaching resource management.

[0007] To achieve the above objectives, the present invention provides an AI-based intelligent teaching platform for medical imaging, comprising:

[0008] The image acquisition unit is used to acquire medical images and classify and store them according to the annotation data of the medical images;

[0009] A question storage unit, which is connected to the image acquisition unit, is used to store medical image test questions;

[0010] A statistics unit, which is connected to the question storage unit, is used to count the abnormal feedback data of each user input for each medical imaging question;

[0011] The comparison unit is connected to the image acquisition unit, the question storage unit and the statistics unit respectively, and is used to determine the contour difference characterization value based on the abnormal receipt data and the standard medical images of the corresponding sub-category in the image acquisition unit.

[0012] An analysis unit, connected to the image acquisition unit, the question storage unit, and the comparison unit, is used to determine whether the stored data of the corresponding sub-category is qualified based on the contour difference characterization value. When the stored data of a single sub-category is determined to be abnormal, the analysis unit determines the symptom detail quantification parameter based on the answer data of the corresponding medical image question. The analysis unit then adjusts the stored data in the image acquisition unit based on the symptom detail quantification parameter and the image horizontal comparison parameter. This includes adjusting the preset acquisition cycle of the medical images acquired by the image acquisition unit to the corresponding value, adjusting the storage amount of medical images for a single sub-category to the corresponding value, or adjusting the clarity of medical images for a single sub-category to the corresponding value.

[0013] Furthermore, the comparison unit is used to periodically determine contour difference characterization values ​​based on abnormal feedback data and standard medical images of the corresponding subcategories in the image acquisition unit, including:

[0014] This is used to classify each abnormal receipt data based on the subcategories of the medical imaging test questions corresponding to each user's abnormal receipt data;

[0015] Used to identify abnormal comparison images based on disease information in each abnormal receipt data corresponding to a single subcategory in statistics;

[0016] The abnormal comparison images and the standard medical images corresponding to the medical image test questions are compared and overlapped to determine the contour difference characterization value of the corresponding sub-category.

[0017] Furthermore, the analysis unit is used to determine whether the stored data for a single subcategory is qualified based on the contour difference characterization value, including:

[0018] If the contour difference characterization value is less than or equal to the preset contour difference characterization value, the stored data of a single subcategory is determined to be abnormal. Based on the answer data of the corresponding medical image test question, the symptom detail quantification parameter is determined, and the stored data in the image acquisition unit is adjusted based on the symptom detail quantification parameter.

[0019] Furthermore, the analysis unit is used to determine the quantitative parameters of disease details based on the answer data of medical imaging test questions, including determining the quantitative parameters of disease details based on the detailed description keywords in the answer data of medical imaging test questions.

[0020] Furthermore, the analysis unit is used to adjust the stored data within the image acquisition unit based on quantification parameters of symptom details, including:

[0021] If the symptom detail quantification parameter is less than or equal to the preset symptom detail quantification parameter, the stored data in the image acquisition unit is adjusted based on the image horizontal comparison parameter.

[0022] Furthermore, the analysis unit is used to determine image lateral comparison parameters for a single subcategory, including:

[0023] Used to determine the lateral contour comparison coefficient based on the feature extraction maps of each stored medical image in a single subcategory;

[0024] Used to determine the lateral texture comparison coefficient based on the surface texture map of each stored medical image in a single subcategory;

[0025] It is used to determine the image lateral comparison parameters based on the lateral contour comparison coefficient and the lateral texture comparison coefficient.

[0026] Furthermore, the analysis unit is used to adjust the stored data within the image acquisition unit based on image lateral comparison parameters, including:

[0027] If the image horizontal comparison parameter is less than or equal to the preset horizontal comparison parameter, the storage capacity of the image acquisition unit for the medical image of a single subcategory will be adjusted to the corresponding value based on the contour difference characterization value.

[0028] The increase in the storage volume of medical images for a single subcategory is negatively correlated with the contour difference characterization value.

[0029] Furthermore, if the image horizontal comparison parameter is greater than the preset horizontal comparison parameter, the preset acquisition cycle of the image acquisition unit for acquiring medical images is adjusted to the corresponding value based on the image horizontal comparison parameter.

[0030] The reduction in the preset acquisition period is positively correlated with the image horizontal comparison parameters.

[0031] Furthermore, if the symptom detail quantification parameter is greater than the preset symptom detail quantification parameter, the clarity of each medical image corresponding to a single subcategory will be adjusted to the corresponding value based on the symptom detail quantification parameter.

[0032] The increase in the clarity of each medical image corresponding to a single subcategory is positively correlated with the quantitative parameter of disease details.

[0033] Furthermore, if the contour difference characterization value is greater than the preset contour difference characterization value, the stored data of a single subcategory is deemed qualified, and the image acquisition unit is controlled to continue running using the current operating parameters.

[0034] Compared with existing technologies, the beneficial effect of this invention lies in the statistical analysis of abnormal feedback data input by each user for each medical image test question. This abnormal feedback data reflects errors or mismatches between the user's answers and the actual answers. A contour difference characterization value is periodically determined based on the abnormal feedback data and the standard medical images of the corresponding subcategories in the image acquisition unit. By classifying and statistically analyzing the abnormal feedback data, disease information within the user error set is identified, and its corresponding standard medical images are compared with the standard images of the test questions to obtain the contour difference characterization value. This value reflects the degree of difference between the standard medical images of user errors and the standard medical images corresponding to the actual questions. The validity of the stored data for a single subcategory is determined based on the contour difference characterization value. When the contour difference characterization value is greater than a preset value, the difference between the image corresponding to the user's error and the standard image is greater, indicating that the user made the mistake due to insufficient actual knowledge. When the contour difference characterization value is less than or equal to the preset value, the image data is too similar, and the user is confused due to insufficient clarity of the image corresponding to the question. In this case, the stored data in the image acquisition unit is further adjusted based on the symptom detail quantification parameter. By promptly identifying problems in the stored data and making targeted adjustments, the efficiency of managing teaching resources has been improved.

[0035] Furthermore, the stored data within the image acquisition unit is adjusted based on the symptom detail quantification parameter, which measures the degree of description of symptom details in the answer data. When anomalies are identified in the stored data, further analysis reveals that when the symptom detail quantification parameter exceeds the preset parameter, the learning demand for symptom details in the corresponding question is high, requiring a detailed understanding of the symptom details. In this case, the clarity of each medical image corresponding to a single subcategory is adjusted to the corresponding value to refine the image information of the corresponding symptom. Targeted determination of the clarity of each medical image ensures that users can understand medical information in detail while improving the utilization of storage space and further enhancing the management efficiency of teaching resources.

[0036] Furthermore, the analysis unit adjusts the stored data program based on the image horizontal comparison parameters to determine the image horizontal comparison parameters. The horizontal contour comparison coefficient reflects the similarity of the contours of the extracted medical image features within a single subcategory. The horizontal texture comparison coefficient reflects the similarity of the surface texture of each medical image within a single subcategory. The image horizontal comparison parameters integrate the similarity of contours and textures, comprehensively reflecting the similarity between medical images within a single subcategory. When the image horizontal comparison parameters are less than or equal to the preset horizontal comparison parameters, the similarity between medical images within a single subcategory is low. In this case, the disease has many variant morphologies, so the storage capacity is increased to enrich the database. When the image horizontal comparison parameters are greater than the preset horizontal comparison parameters, the similarity between medical images within a single subcategory is high. In this case, based on the image horizontal comparison parameters, the preset acquisition cycle of the image acquisition unit for acquiring medical images is adjusted to the corresponding value to acquire more diverse images, further optimizing the medical image resources in the database. Reasonable adjustment of the stored data according to the image horizontal comparison parameters increases the diversity of medical images and enriches the database resources.

[0037] Furthermore, the increase in the storage capacity of medical images for a single subcategory is negatively correlated with the contour difference characterization value. The contour difference characterization value reflects the degree of difference between the abnormal comparison image and the standard medical image. The smaller the contour difference characterization value, the more similar the abnormal comparison image is to the standard image corresponding to the test question, and the greater the need to increase storage capacity to enrich teaching resources and increase the reserve of image information. The storage capacity of medical images is precisely adjusted based on user feedback. This avoids the waste of resources caused by blindly increasing storage capacity while timely supplementing teaching resources, thus improving the utilization efficiency of teaching resources. Attached Figure Description

[0038] Figure 1 is a block diagram of the AI-based intelligent teaching platform for medical imaging according to an embodiment of the present invention;

[0039] Figure 2 is a logic diagram of the analysis unit of the present invention determining whether the stored data of a single subcategory is qualified based on the contour difference characterization value;

[0040] Figure 3 is a logic decision diagram of the stored data in the image acquisition unit based on the parameter adjustment of disease details in the analysis unit of the embodiment of the present invention.

[0041] Figure 4 is a logic decision diagram of the image acquisition unit based on the image horizontal comparison parameter adjustment in the analysis unit of the embodiment of the present invention, which is a logic decision diagram of the stored data in the image acquisition unit. Detailed Implementation

[0042] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0043] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0044] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0045] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0046] Please refer to Figure 1, which is a block diagram of the AI-based intelligent medical imaging teaching platform according to an embodiment of the present invention. The intelligent medical imaging teaching platform of the present invention includes:

[0047] The image acquisition unit is used to acquire medical images and classify and store them according to the annotation data of the medical images;

[0048] A question storage unit, which is connected to the image acquisition unit, is used to store medical image test questions;

[0049] A statistics unit, which is connected to the question storage unit, is used to count the abnormal feedback data of each user input for each medical imaging question;

[0050] The comparison unit is connected to the image acquisition unit, the question storage unit and the statistics unit respectively, and is used to determine the contour difference characterization value based on the abnormal receipt data and the standard medical images of the corresponding sub-category in the image acquisition unit.

[0051] An analysis unit, connected to the image acquisition unit, the question storage unit, and the comparison unit, is used to determine whether the stored data of the corresponding sub-category is qualified based on the contour difference characterization value. When the stored data of a single sub-category is determined to be abnormal, the analysis unit determines the symptom detail quantification parameter based on the answer data of the corresponding medical image question. The analysis unit then adjusts the stored data in the image acquisition unit based on the symptom detail quantification parameter and the image horizontal comparison parameter. This includes adjusting the preset acquisition cycle of the medical images acquired by the image acquisition unit to the corresponding value, adjusting the storage amount of medical images for a single sub-category to the corresponding value, or adjusting the clarity of medical images for a single sub-category to the corresponding value.

[0052] Specifically, the labeled data for medical images includes pathological diagnosis results, feature extraction maps, and surface texture maps;

[0053] The pathological diagnosis results can be encoded using the International Classification of Diseases (ICD) to indicate the specific disease.

[0054] The specific methods for obtaining feature extraction maps include labeling disease features in medical image information, and extracting the labeled disease features to obtain feature extraction maps.

[0055] The surface texture diagram is a selected area showing the surface details used to represent the characteristics of the disease.

[0056] Specifically, the image acquisition unit classifies medical images based on labeled data by mapping the medical images to the corresponding subcategories of the corresponding categories, and then mapping the medical images to specific diseases in specific departments according to the medical discipline classification system, with categories corresponding to departments and subcategories corresponding to specific diseases.

[0057] Specifically, the answer data for each medical imaging question in the question storage unit contains specific disease information linked by ICD codes, so that a single medical imaging question can be randomly matched with multiple corresponding medical images.

[0058] Specifically, abnormal receipt data refers to data where the user's input does not match the answer data for the medical imaging test questions.

[0059] Specifically, the comparison unit is used to periodically determine contour difference characterization values ​​based on abnormal feedback data and standard medical images of the corresponding subcategories in the image acquisition unit, including:

[0060] Obtain abnormal receipt data for each user within a preset comparison period, and classify the abnormal receipt data based on the subcategories of the medical imaging test questions corresponding to each abnormal receipt data.

[0061] The disease information in each abnormal receipt data corresponding to a single subcategory is statistically analyzed, and the standard medical images corresponding to the disease information with the largest proportion are identified as abnormal comparison images.

[0062] The abnormal comparison images and the standard medical images corresponding to the medical image test questions are compared by overlap, and the ratio of the area of ​​the overlapping abnormal region to the total area of ​​the abnormal comparison images is determined as the contour difference characterization value of the corresponding sub-category.

[0063] Specifically, the image acquisition unit stores standard medical images corresponding to each subcategory. The standard medical images are feature extraction maps of clear medical image information corresponding to specific diseases.

[0064] Specifically, the image acquisition unit periodically acquires medical images of each subcategory according to a preset acquisition cycle.

[0065] Specifically, there are no restrictions on the source of medical images; they can be clinical medical images.

[0066] Please refer to Figure 2, which is a logic diagram of the analysis unit of this invention determining whether the stored data of a single subcategory is qualified based on the contour difference characterization value. The analysis unit of this invention is used to determine whether the stored data of a single subcategory is qualified based on the contour difference characterization value, including:

[0067] If the contour difference characterization value is greater than the preset contour difference characterization value, the stored data of a single subcategory is deemed qualified, and the image acquisition unit is controlled to continue to run using the current operating parameters.

[0068] If the contour difference characterization value is less than or equal to the preset contour difference characterization value, the stored data of a single subcategory is determined to be abnormal. Based on the answer data of the corresponding medical image test question, the symptom detail quantification parameter is determined, and the stored data in the image acquisition unit is adjusted based on the symptom detail quantification parameter.

[0069] Specifically, the preset contour difference characterization value is selected within the range [0.2, 0.3]. Those skilled in the art can select and determine the preset contour difference characterization value according to the actual situation. A large amount of historical medical image test questions and user answer data can be analyzed to statistically determine the reasons for user errors under different contour difference characterization values. Combined with the actual requirements for data discrimination, the preset contour difference characterization value can be determined. It is understood that this can be used to classify whether two standard medical images are easily confused. In this embodiment, preferably, the preset contour difference characterization value is 0.3.

[0070] Specifically, the system collects abnormal feedback data from each user's input of medical image test questions. This abnormal feedback data reflects errors or mismatches between user answers and actual responses. Contour difference representation values ​​are periodically determined based on the abnormal feedback data and the standard medical images for the corresponding subcategories in the image acquisition unit. By classifying and statistically analyzing the abnormal feedback data, disease information within user error clusters is identified. The corresponding standard medical images are then compared with the standard images for the test questions to obtain contour difference representation values. These values ​​reflect the degree of difference between the standard medical images used for user errors and the standard medical images corresponding to the actual questions. The validity of stored data for a single subcategory is determined based on these contour difference representation values. When the value exceeds a preset value, the difference between the user's incorrect image and the standard image is greater, indicating insufficient user knowledge. When the value is less than or equal to the preset value, the image data is too similar, leading to user confusion due to insufficient image clarity corresponding to the questions. In this case, the stored data within the image acquisition unit is further adjusted based on symptom detail quantification parameters. Timely identification and targeted adjustments of problems in the stored data improve the efficiency of teaching resource management.

[0071] Specifically, the analysis unit is used to determine quantitative parameters of disease details based on the answer data of medical imaging test questions, including,

[0072] Used to extract detailed descriptive keywords based on answer data of medical imaging test questions;

[0073] The ratio of the number of bytes of the keywords describing the detailed information in the calculation to the total number of bytes of the corresponding answer data is used to determine the parameter for quantifying the details of the illness.

[0074] Specifically, there are no restrictions on the method for determining the keywords for detailed descriptions. Adjectives extracted from the answer data of medical imaging test questions can be used as keywords for detailed descriptions, which will not be elaborated further.

[0075] Please refer to Figure 3, which is a logic decision diagram of the analysis unit adjusting the stored data in the image acquisition unit based on the quantification parameter of disease details in an embodiment of the present invention. The analysis unit of the present invention is used to adjust the stored data in the image acquisition unit based on the quantification parameter of disease details, including:

[0076] If the symptom detail quantification parameter is less than or equal to the preset symptom detail quantification parameter, the stored data in the image acquisition unit is adjusted based on the image horizontal comparison parameter.

[0077] If the symptom detail quantification parameter is greater than the preset symptom detail quantification parameter, the clarity of each medical image corresponding to a single subcategory will be adjusted to the corresponding value based on the symptom detail quantification parameter.

[0078] Specifically, the preset quantification parameter for symptom details is selected within the range [0.32, 0.44]. Those skilled in the art can select and determine the preset quantification parameter for symptom details according to the actual situation. This can be achieved by analyzing a large number of medical imaging test questions and answers, statistically analyzing the demand for medical imaging details under different levels of detail description, and combining this with the emphasis on symptom details in teaching. In this embodiment, preferably, the preset quantification parameter for symptom details is 0.32.

[0079] Specifically, the stored data within the image acquisition unit is adjusted based on the symptom detail quantification parameter, which measures the degree of description of symptom details in the answer data. When anomalies are identified in the stored data, further analysis of the symptom detail quantification parameter reveals that when it exceeds the preset parameter, the learning demand for symptom details in the corresponding question is high, requiring a detailed understanding of the symptom details. In this case, the clarity of each medical image corresponding to a single subcategory is adjusted to the corresponding value to refine the image information of the corresponding symptom. Targeted determination of the clarity of each medical image ensures that users can understand medical information in detail while improving the utilization of storage space and further enhancing the management efficiency of teaching resources.

[0080] Specifically, the analysis unit is used to determine image lateral comparison parameters for a single subcategory, including:

[0081] This is used to obtain the stored medical images of a single subcategory, extract the feature maps of each medical image, and perform two-class group comparisons on the extracted feature maps.

[0082] For a single comparison group, contour comparison is performed on two feature extraction images;

[0083] The ratio of the overlapping contour length of a single alignment group to the total contour length of the two feature extraction maps is calculated to obtain the contour alignment value for a single alignment group.

[0084] The average value of the comparison values ​​of each comparison group is calculated to obtain the horizontal contour comparison coefficient for a single subcategory.

[0085] The surface texture maps of a single comparison group are compared by overlap, and the ratio of the area of ​​the overlapping area to the total area of ​​the two surface texture maps is determined as the texture comparison value for a single comparison group.

[0086] The average value of the texture comparison values ​​of each comparison group is calculated to obtain the horizontal texture comparison coefficient for a single subcategory.

[0087] The average value of the horizontal contour comparison coefficient and the horizontal texture comparison coefficient is calculated to obtain the horizontal comparison parameters of the image.

[0088] Please refer to Figure 4, which is a logic decision diagram of the analysis unit adjusting the stored data in the image acquisition unit based on image lateral comparison parameters in an embodiment of the present invention. The analysis unit of the present invention is used to adjust the stored data in the image acquisition unit based on image lateral comparison parameters, including:

[0089] If the image horizontal comparison parameter is less than or equal to the preset horizontal comparison parameter, the storage capacity of the image acquisition unit for the medical image of a single subcategory will be adjusted to the corresponding value based on the contour difference characterization value.

[0090] If the image horizontal comparison parameter is greater than the preset horizontal comparison parameter, the preset acquisition cycle of the image acquisition unit for acquiring medical images will be adjusted to the corresponding value based on the image horizontal comparison parameter.

[0091] Specifically, the preset horizontal comparison parameters are selected within the range [0.68, 0.74]. Those skilled in the art can select and determine the preset horizontal comparison parameters according to the actual situation. This allows for horizontal comparison analysis of a large number of medical images. The preset horizontal comparison parameters are determined in conjunction with the requirements for diversity in medical images in teaching. In this embodiment, preferably, the preset horizontal comparison parameter is 0.7.

[0092] Specifically, the analysis unit adjusts the stored data program based on image lateral comparison parameters to determine these parameters. The lateral contour comparison coefficient reflects the similarity of the contours of the extracted medical image features within a single subcategory. The lateral texture comparison coefficient reflects the similarity of the surface texture of each medical image within a single subcategory. The image lateral comparison parameters integrate the similarity of contours and textures, comprehensively reflecting the similarity between medical images within a single subcategory. When the image lateral comparison parameters are less than or equal to the preset lateral comparison parameters, the similarity between medical images within a single subcategory is low. In this case, the disease has many variations, requiring increased storage to enrich the database. When the image lateral comparison parameters are greater than the preset lateral comparison parameters, the similarity between medical images within a single subcategory is high. In this case, based on the image lateral comparison parameters, the preset acquisition cycle of the image acquisition unit for acquiring medical images is adjusted to the corresponding value to acquire more diverse images, further optimizing the medical image resources in the database. Reasonable adjustment of the stored data according to the image lateral comparison parameters increases the diversity of medical images and enriches the database resources.

[0093] Specifically, the analysis unit is used to adjust the storage capacity of the image acquisition unit for a single subcategory of medical images to a corresponding value based on the contour difference characterization value, wherein,

[0094] The increase in the storage volume of medical images for a single subcategory is negatively correlated with the contour difference characterization value.

[0095] In this embodiment, optionally,

[0096] Compare the contour difference characterization value with the first contour comparison value and the second contour comparison value;

[0097] If the contour difference characterization value is less than or equal to the first contour comparison value, the storage size of the medical images of a single subcategory will be adjusted to 1.28 times the initial storage size.

[0098] If the contour difference characterization value is less than or equal to the second contour comparison value and greater than the first contour comparison value, then the storage size of the medical images of a single subcategory will be adjusted to 1.21 times the initial storage size.

[0099] If the contour difference characterization value is greater than the second contour comparison value, the storage size of the medical images of a single subcategory will be adjusted to 1.09 times the initial storage size.

[0100] The first contour comparison value is set to 0.64L0, and the second contour comparison value is set to 0.76L0, where L0 is a preset contour difference characterization value.

[0101] Specifically, the image acquisition unit allocates corresponding storage space for each subcategory. When the storage space reaches the required amount, historical medical images are deleted based on the storage time.

[0102] Specifically, the increase in the storage capacity of medical images for a single subcategory is negatively correlated with the contour difference characterization value. The contour difference characterization value reflects the degree of difference between the abnormal comparison image and the standard medical image. The smaller the contour difference characterization value, the more similar the abnormal comparison image is to the standard image corresponding to the test question, and the greater the need to increase storage capacity to enrich teaching resources and increase the reserve of image information. The storage capacity of medical images is precisely adjusted based on user feedback. This avoids the waste of resources caused by blindly increasing storage capacity while timely supplementing teaching resources, thus improving the utilization efficiency of teaching resources.

[0103] Specifically, the analysis unit is used to adjust the preset acquisition cycle of the medical images acquired by the image acquisition unit to a corresponding value based on the image lateral comparison parameters, wherein,

[0104] The reduction in the preset acquisition period is positively correlated with the image horizontal comparison parameters.

[0105] In this embodiment, optionally,

[0106] The image horizontal comparison parameters are compared with the first preset horizontal comparison value and the second preset horizontal comparison value;

[0107] If the image horizontal comparison parameter is less than or equal to the first preset horizontal comparison value, the preset acquisition cycle of the image acquisition unit for acquiring medical images will be adjusted to 0.91 times the initial preset acquisition cycle adjustment.

[0108] If the image horizontal comparison parameter is less than or equal to the second preset horizontal comparison value and greater than the first preset horizontal comparison value, then the preset acquisition cycle of the image acquisition unit for acquiring medical images is adjusted to 0.81 times the initial preset acquisition cycle adjustment.

[0109] If the image horizontal comparison parameter is greater than the second preset horizontal comparison value, the preset acquisition cycle of the image acquisition unit for acquiring medical images will be adjusted to 0.69 times the initial preset acquisition cycle.

[0110] The first preset horizontal comparison value is 1.33C0, and the second preset horizontal comparison value is 1.61C0, where C0 is the preset horizontal comparison parameter.

[0111] When medical images within a single subcategory have high similarity and correspond to similar disease information, the acquisition frequency should be increased to enrich the medical image samples in the database. Medical images from different times and from different patients will vary; increasing the acquisition frequency will yield more diverse images, thereby improving the comprehensiveness of the medical images.

[0112] Specifically, the analysis unit is used to adjust the clarity of each medical image corresponding to a single subcategory to a corresponding value based on the symptom detail quantification parameter, wherein,

[0113] The increase in the clarity of each medical image corresponding to a single subcategory is positively correlated with the quantitative parameter of disease details.

[0114] In this embodiment, optionally,

[0115] The quantification parameters of symptom details are compared with the first preset quantification comparison value and the second preset quantification comparison value;

[0116] If the symptom detail quantification parameter is less than or equal to the first preset detail quantification comparison value, then the clarity of each medical image corresponding to a single subcategory will be adjusted to 1.11 times the initial clarity.

[0117] If the symptom detail quantization parameter is less than or equal to the second preset detail quantization comparison value and greater than the first preset detail quantization comparison value, then the clarity of each medical image corresponding to a single subcategory will be adjusted to 1.23 times the initial clarity.

[0118] If the symptom detail quantification parameter is greater than the second preset detail quantification comparison value, the clarity of each medical image corresponding to a single subcategory will be adjusted to 1.31 times the initial clarity.

[0119] The first preset detail quantization comparison value is 1.31X0, and the second preset detail quantization comparison value is 1.56X0, where X0 is the preset symptom detail quantization parameter.

[0120] Specifically, the increase in the clarity of medical images corresponding to a single subcategory is positively correlated with the quantification parameter of symptom detail. The quantification parameter of symptom detail measures the richness of the symptom detail description in the answer data. The higher the quantification parameter of symptom detail, the greater the need for detailed symptom understanding, and the greater the need to improve the clarity of medical images so that users can more clearly observe detailed image features, ensuring that image clarity matches the symptom detail description. This improves the management efficiency of teaching resources.

[0121] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0122] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An AI-based intelligent teaching platform for medical imaging, characterized in that, include: The image acquisition unit is used to acquire medical images and classify and store them according to the annotation data of the medical images; A question storage unit, connected to the image acquisition unit, is used to store medical imaging questions; a statistics unit, connected to the question storage unit, is used to count the abnormal feedback data input by each user for each medical imaging question. The error receipt data is used to report errors or discrepancies between the user's answers and the actual answers during the question-answering process. The comparison unit is connected to the image acquisition unit, the question storage unit and the statistics unit respectively, and is used to determine the contour difference characterization value based on the abnormal receipt data and the standard medical images of the corresponding sub-category in the image acquisition unit. The contour difference characterization value is used to reflect the degree of difference between the incorrect medical images in the user set and the standard medical images corresponding to the actual questions; An analysis unit, connected to the image acquisition unit, the question storage unit, and the comparison unit, is used to determine whether the stored data for a single subcategory is qualified based on the contour difference characterization value, and to determine that the stored data for a single subcategory is abnormal when the contour difference characterization value is less than or equal to a preset contour difference characterization value. It also determines the symptom detail quantification parameter based on the answer data of the corresponding medical image question, and adjusts the stored data within the image acquisition unit based on the symptom detail quantification parameter and the image horizontal comparison parameter. This includes adjusting the preset acquisition cycle of medical images acquired by the image acquisition unit to a corresponding value, adjusting the storage amount of medical images for a single subcategory to a corresponding value, or adjusting the clarity of medical images for a single subcategory to a corresponding value. The symptom detail quantification parameter is used to measure the richness of symptom detail descriptions in the answer data. The analysis unit is used to determine whether the stored data for a single subcategory is qualified based on the symptom detail quantification parameter. The symptom detail quantification parameter adjusts the stored data in the image acquisition unit, including: if the symptom detail quantification parameter is less than or equal to a preset symptom detail quantification parameter, then the stored data in the image acquisition unit is adjusted based on the image horizontal comparison parameter; the analysis unit is used to determine the image horizontal comparison parameter for a single subcategory, including: determining the horizontal contour comparison coefficient based on the feature extraction map of each stored medical image in the acquired single subcategory; determining the horizontal texture comparison coefficient based on the surface texture map of each stored medical image in the single subcategory; determining the image horizontal comparison parameter based on the horizontal contour comparison coefficient and the horizontal texture comparison coefficient; the horizontal contour comparison coefficient is used to reflect the similarity of the contours of the feature extraction maps of each medical image in a single subcategory; the horizontal texture comparison coefficient is used to reflect the similarity of the surface texture of each medical image in a single subcategory.

2. The AI-based intelligent teaching platform for medical imaging according to claim 1, characterized in that, The comparison unit is used to determine the contour difference characterization value based on the abnormal receipt data and the standard medical images of the corresponding subcategories in the image acquisition unit, including: classifying each abnormal receipt data according to the subcategories of the medical image test questions corresponding to each user's abnormal receipt data; determining abnormal comparison images based on the disease information in each abnormal receipt data corresponding to a single subcategory; and performing an overlap comparison between the abnormal comparison images and the standard medical images corresponding to the medical image test questions to determine the contour difference characterization value of the corresponding subcategory.

3. The AI-based intelligent teaching platform for medical imaging according to claim 2, characterized in that, The analysis unit is used to determine the quantitative parameters of disease details based on the answer data of the corresponding medical imaging test questions, including: determining the quantitative parameters of disease details based on the detailed description keywords in the answer data of the medical imaging test questions.

4. The AI-based intelligent teaching platform for medical imaging according to claim 3, characterized in that, The analysis unit is used to adjust the stored data in the image acquisition unit based on the image horizontal comparison parameters, including: if the image horizontal comparison parameters are less than or equal to a preset horizontal comparison parameters, the storage amount of medical images for a single subcategory in the image acquisition unit is adjusted to the corresponding value based on the contour difference characterization value; the increase in the storage amount of medical images for a single subcategory is negatively correlated with the contour difference characterization value; the image horizontal comparison parameters are used to comprehensively reflect the similarity between medical images in a single subcategory.

5. The AI-based intelligent teaching platform for medical imaging according to claim 4, characterized in that, If the image horizontal comparison parameter is greater than the preset horizontal comparison parameter, the preset acquisition cycle of the image acquisition unit for acquiring medical images is adjusted to the corresponding value based on the image horizontal comparison parameter; the reduction of the preset acquisition cycle is positively correlated with the image horizontal comparison parameter.

6. The AI-based intelligent teaching platform for medical imaging according to claim 5, characterized in that, If the quantification parameter for symptom details is greater than the preset quantification parameter for symptom details, the clarity of each medical image corresponding to a single subcategory will be adjusted to the corresponding value based on the quantification parameter for symptom details; the increase in clarity of each medical image corresponding to a single subcategory is positively correlated with the quantification parameter for symptom details.

7. The AI-based intelligent teaching platform for medical imaging according to claim 6, characterized in that, If the contour difference characterization value is greater than the preset contour difference characterization value, the stored data of a single subcategory is deemed qualified, and the image acquisition unit is controlled to continue running using the current operating parameters.

Images