Medical knowledge visual propagation method based on 3D medical image and AI interaction

Abstract

The invention discloses a medical knowledge visual propagation method based on 3D medical image and AI interaction. According to the method, firstly, multi-dimensional medical image data of human organs or tissues are obtained from medical image equipment, then the data are preprocessed, and a high-precision 3D medical image is generated through a three-dimensional reconstruction algorithm. A deep learning model is trained through manually labeled medical image data, the model can identify different tissues and lesions in the image, and related medical knowledge introduction is provided according to interactive operation of a user. And finally, spreading the 3D medical image and the intelligent interaction result thereof to a target user through a network platform. According to the invention, through visual display of the 3D image and the AI interaction technology, the visuality, interactivity and propagation efficiency of medical knowledge propagation are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data visualization, and particularly to a medical knowledge visualization dissemination method based on 3D medical images and AI interaction. Background Art

[0002] In the field of medical knowledge dissemination, traditional teaching and learning methods mainly rely on two-dimensional medical images (such as X-ray films, CT two-dimensional slice images) and text and chart explanations. For example, a CT device obtains tomographic image data by scanning the human body and converts it into two-dimensional images for display on a screen or printing on a film. Doctors or learners identify tissues and lesions by observing the gray-scale differences on the two-dimensional images. For example, in a lung CT image, white areas may represent bones or high-density lesions, and gray areas represent normal lung tissue.

[0003] However, this two-dimensional image method has obvious limitations: First, two-dimensional images cannot intuitively display the three-dimensional structure and spatial relationships of organs, and it is difficult for learners to comprehensively understand complex anatomical structures, such as the anterior-posterior and up-down positional relationships of blood vessels in the brain; Second, existing medical knowledge dissemination methods are mostly one-way outputs and lack intelligent interaction. Learners cannot obtain more detailed medical knowledge through interaction with the images, which is not conducive to in-depth understanding and memory of knowledge; In addition, the dissemination efficiency of medical information is low, usually relying on hospital internal systems or simple data sharing methods. For example, when a patient is transferred, the receiving hospital needs to spend a lot of time obtaining and interpreting the data. Summary of the Invention

[0004] Based on this, the embodiments of the present application provide a medical knowledge visualization dissemination method based on 3D medical images and AI interaction, which can improve the dissemination efficiency of medical information and facilitate telemedicine and patient transfer.

[0005] In a first aspect, a medical knowledge visualization dissemination method based on 3D medical images and AI interaction is provided, and the method includes:

[0006] Obtain multi-dimensional medical image data of human organs or tissues from a medical imaging device;

[0007] Preprocess the obtained multi-dimensional medical image data, and then perform volume rendering on the preprocessed data using a three-dimensional reconstruction algorithm to generate high-precision three-dimensional medical images;

[0008] Train a deep learning model with manually annotated medical image data, and process the three-dimensional medical images through the trained deep learning model to obtain intelligent interaction results; wherein, the deep learning model is used to identify different tissues and lesions in the images and obtain corresponding intelligent interaction results according to the interaction data provided by the user;

[0009] Disseminate the three-dimensional medical image and its intelligent interaction results to the target users in a visual form.

[0010] Optionally, the preprocessing includes denoising and artifact correction for multi-dimensional medical image data, including:

[0011] Apply a filtering algorithm to the acquired multi-dimensional medical image data to remove noise;

[0012] Correct the artifacts in the data through an image correction algorithm to improve the data quality.

[0013] Optionally, use a three-dimensional reconstruction algorithm to perform volume rendering on the preprocessed data to generate a high-precision three-dimensional medical image, specifically including:

[0014] Divide the preprocessed multi-dimensional medical image data into voxel grids;

[0015] Calculate the intersection points and normal vectors of the isosurfaces for each voxel grid;

[0016] According to the intersection point and normal vector information, use the Marching Cubes algorithm to generate the surface grid of the three-dimensional model.

[0017] Optionally, train a deep learning model with manually annotated medical image data;

[0018] Collect and organize a large amount of manually annotated medical image data, and the annotation information includes tissue type and lesion type;

[0019] Divide the annotated data into a training set and a validation set;

[0020] Use the training set to train the CNN model and optimize the model parameters through the backpropagation algorithm;

[0021] Use the validation set to evaluate and adjust the trained model.

[0022] Optionally, process the three-dimensional medical image through the trained deep learning model to obtain intelligent interaction results, specifically including:

[0023] Use the trained deep learning model to process the three-dimensional medical image to achieve intelligent recognition of each region in the image;

[0024] When the user clicks on a certain region of the three-dimensional medical image or asks a question through voice, the deep learning model can identify the tissue type and lesion information of that region;

[0025] The model retrieves the corresponding medical knowledge from the preset knowledge base according to the recognition results and displays the retrieved medical knowledge to the user in text or voice form.

[0026] Optionally, disseminating the 3D model and the analyzed target medical information in a visual form to the target user includes:

[0027] Disseminating the 3D medical image and its intelligent interaction results through a network platform, and users can access it through a browser or a dedicated client;

[0028] Or displaying the 3D medical image and its intelligent interaction results through AR / VR devices.

[0029] In a second aspect, a medical knowledge visual dissemination system based on the interaction between 3D medical images and AI is provided. The system includes:

[0030] An acquisition module for acquiring multi-dimensional medical image data of human organs or tissues from medical imaging devices;

[0031] A rendering module for preprocessing the acquired multi-dimensional medical image data, and then performing volume rendering on the preprocessed data using a 3D reconstruction algorithm to generate a high-precision 3D medical image;

[0032] A processing module for training a deep learning model with manually annotated medical image data, and processing the 3D medical image through the trained deep learning model to obtain intelligent interaction results; wherein, the deep learning model is used to identify different tissues and lesions in the image and obtain corresponding intelligent interaction results according to the interaction data provided by the user;

[0033] A dissemination module for disseminating the 3D medical image and its intelligent interaction results in a visual form to the target user.

[0034] In a third aspect, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the medical knowledge visual dissemination method according to any one of the first aspects is implemented.

[0035] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the medical knowledge visual dissemination method according to any one of the first aspects is implemented.

[0036] In a fifth aspect, a computer program product is provided, including computer programs / instructions. When the computer programs / instructions are executed by a processor, the medical knowledge visual dissemination method according to any one of the first aspects is implemented.

[0037] The beneficial effects brought by the technical solutions provided in the embodiments of the present application at least include:

[0038] Improve spatial comprehension: 3D medical images can fully display the three-dimensional structure and spatial relationship of organs. Learners can observe from different angles and understand medical knowledge more intuitively. For example, when learning about the structure of the heart, 3D images can clearly show the spatial position and connection relationship of the atria, ventricles, and valves, which helps to improve the understanding of the physiological function of the heart.

[0039] Enhanced interactivity and learning effect: AI interactive technology enables learners to actively acquire knowledge, and through asking questions and clicking, they can gain a deeper understanding of the details in medical images, making learning more interesting and memorable. Studies have shown that students who learn in this interactive way have an 80% higher level of mastery of medical knowledge than those who learn in traditional ways.

[0040] Improved communication efficiency: Rapid dissemination of medical knowledge is achieved through the Internet and AR / VR technology. For example, when a patient is referred, the receiving hospital can quickly obtain visual medical data through the Internet platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0042] Figure 1 A flowchart of the steps of a method for visualizing and disseminating medical knowledge based on 3D medical images and AI interaction provided in an embodiment of the present application;

[0043] Figure 2 A block diagram of a medical knowledge visualization dissemination system based on 3D medical images and AI interaction provided in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] In the description of the present invention, the terms "including", "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units that are clearly listed, but may also include other steps or units that are inherent to these processes, methods, products or devices although not clearly listed, or steps or units added based on further optimization schemes conceived in the present invention.

[0047] The dissemination of existing medical knowledge mostly relies on two-dimensional images (such as X-ray films, CT two-dimensional slice images) and traditional text and chart explanations. Taking two-dimensional CT images as an example, tomographic image data is obtained by scanning the human body with a CT device, and these data are converted into two-dimensional images and displayed on a screen or printed on a film. Doctors or learners identify tissues and lesions by observing the gray-scale differences on the two-dimensional images. For example, in lung CT images, white areas may represent bones or high-density lesions, and gray areas represent normal lung tissue. However, this method has limitations. Two-dimensional images cannot intuitively display the three-dimensional structure and spatial relationships of organs.

[0048] Typical application scenarios: In medical education, learners understand complex anatomical structures through two-dimensional images, but it is difficult to intuitively master three-dimensional spatial relationships.

[0049] The existing technologies have the following disadvantages:

[0050] Lack of spatial information: It is difficult for two-dimensional images to enable learners to comprehensively understand the three-dimensional structure and spatial positional relationships of organs. For example, when learning complex brain vascular structures, two-dimensional images cannot clearly present the front-back, up-down positional relationships of blood vessels, resulting in difficult understanding.

[0051] Lack of intelligent interaction: The existing dissemination methods are mostly one-way outputs, and learners cannot interact with the images intelligently. For example, they cannot obtain detailed medical knowledge explanations by clicking on a certain area of the image, nor can they ask in real time about the functions of specific structures in the image, which is not conducive to in-depth understanding and memory of knowledge. The present invention can effectively solve these problems by constructing 3D medical images and introducing AI interaction technology.

[0052] Low dissemination efficiency: The dissemination methods rely on hospital internal systems or simple data sharing methods and lack efficient visual dissemination means. For example, when a patient is transferred, the receiving hospital needs to spend a lot of time obtaining and interpreting the data.

[0053] Please refer to Figure 1 , which shows a flowchart of a medical knowledge visual dissemination method based on 3D medical images and AI interaction provided by an embodiment of the present application, and may include the following steps:

[0054] S1. Obtain multi-dimensional medical image data of human organs or tissues from a medical imaging device.

[0055] In this step, the acquisition of multi-dimensional medical image data is achieved through an interface with a medical imaging device (such as CT, MRI). These devices can generate tomographic image data of human organs or tissues, which are stored in a multi-dimensional form and contain rich anatomical structure information.

[0056] S2. Preprocess the obtained multi-dimensional medical image data, and then use a 3D reconstruction algorithm to perform volume rendering on the preprocessed data to generate a high-precision 3D medical image.

[0057] In this step, preprocessing the obtained multi-dimensional medical image data mainly aims to remove noise and artifacts to improve data quality. The preprocessing steps include applying a filtering algorithm (such as Gaussian filtering or median filtering) to remove image noise, and correcting artifacts through an image correction algorithm. The preprocessed data is divided into a voxel grid, and then volume rendering is performed using a 3D reconstruction algorithm (such as the Marching Cubes algorithm). This algorithm generates the surface grid of the 3D model by calculating the intersection points and normal vectors of the isosurfaces, thus constructing a high-precision 3D medical image. For example, when constructing a 3D model of the liver, by performing the above processing on the tomographic images of the liver obtained by CT scanning, the three-dimensional morphology of the liver, the distribution of internal blood vessels and bile ducts can be clearly presented.

[0058] In the exemplary embodiments of the present application, in addition to the marching cubes algorithm, a raycasting algorithm can also be used for 3D reconstruction. The ray casting algorithm generates a 3D image by emitting rays from the viewpoint through the volume data and calculating the intersection point colors and transparencies of the rays with the voxels. Although the principles of the two algorithms are different, both can achieve the construction of 3D medical images, and each has its own advantages in different scenarios.

[0059] S3. Train a deep learning model with manually annotated medical image data, and process the 3D medical image through the trained deep learning model to obtain an intelligent interaction result.

[0060] Among them, the deep learning model is used to identify different tissues and lesions in images and obtain corresponding intelligent interaction results according to the interaction data provided by the user. In this step, the core of this step is to use deep learning technology to achieve intelligent recognition and interaction of 3D medical images. First, a large amount of manually annotated medical image data is collected, and the annotation information includes tissue types, lesion types, etc. These data are divided into a training set and a validation set for training a convolutional neural network (CNN) model. During the training process, the model parameters are optimized through the backpropagation algorithm so that it can accurately identify different tissues and lesions in the images. The trained model is applied to three-dimensional medical images. When the user clicks on an image area or asks a question by voice, the model can quickly identify the tissue type of the area and retrieve the corresponding medical knowledge from a preset knowledge base and display it to the user in text or voice form. For example, when the user asks "What are the main functions of the liver?", the AI will highlight the liver on the 3D liver image and give a detailed medical knowledge explanation.

[0061] In an alternative embodiment of the present application, in addition to the convolutional neural network, recurrent neural networks (RNNs) or Transformer architectures can also be tried for medical image recognition and interaction. RNNs are suitable for processing sequential data and may be more advantageous for analyzing the changes in medical images over time; the Transformer architecture has performed well in the field of natural language processing and may be able to better understand and handle the complex problems of learners.

[0062] S4. Propagate the three-dimensional medical image and its intelligent interaction result to the target user in a visual form.

[0063] In this step, the generated three-dimensional medical image and its intelligent interaction result are propagated to the target user in a visual form. The propagation methods include remote access through a network platform, and users can view the 3D medical image and perform interactive operations in real time through a browser or a dedicated client. In addition, an immersive experience can be achieved through AR / VR devices. Users can wear a head-mounted device or use a mobile terminal to observe the 3D medical image and interact with the image through gestures, voice, etc. This visual propagation method not only improves the propagation efficiency of medical knowledge but also provides an efficient and convenient solution for scenarios such as medical education, telemedicine, and patient rehabilitation. In addition to click and voice interaction, gesture recognition or eye movement tracking technology can also be introduced to achieve more natural interaction.

[0064] In an alternative embodiment of the present application, the specific solution can also be as follows:

[0065] Input: Obtain multi-dimensional medical data generated by medical imaging devices (such as CT, MRI).

[0066] 3D Medical Image Construction: Using medical imaging data (such as CT, MRI data), high-precision 3D medical images are constructed through 3D reconstruction algorithms. First, the tomographic image data obtained by medical imaging equipment is preprocessed to remove noise and artifacts. Then, classic algorithms such as marching cubes are used to perform volume rendering on the processed data to generate a 3D model. For example, when constructing a 3D model of the liver, by performing the above processing on the tomographic images of the liver obtained by CT scanning, the three-dimensional shape of the liver, the distribution of internal blood vessels and bile ducts can be clearly presented.

[0067] AI Algorithm for Image Intelligent Recognition and Interaction: Applying deep learning algorithms to 3D medical images. A convolutional neural network (CNN) is trained with a large amount of labeled medical image data so that the model can identify different tissues and lesions in the images. When a learner clicks on a certain area in the 3D image, the AI algorithm can quickly identify the tissue type of that area and provide relevant medical knowledge introductions, such as the functions of the tissue, common lesion types, etc. At the same time, learners can also ask questions by voice, and the AI will locate and give answers according to the questions in the 3D image. For example, when a learner asks "What are the main functions of the liver", the AI will highlight the liver in the 3D liver image and give text and voice explanations.

[0068] Visualization and Dissemination: The 3D model and intelligent interaction interface are disseminated to learners through a network platform or AR / VR devices. For example, in medical education, learners can observe a 3D heart model through AR devices and obtain relevant knowledge in real time.

[0069] In summary, the differences between this application and the prior art include:

[0070] (1) Differences in 3D Image Construction:

[0071] The prior art mainly relies on 2D images and can only display the information of a certain cross-section of an organ or tissue;

[0072] This invention constructs 3D medical images using medical imaging data, comprehensively presenting the three-dimensional structure and spatial relationship of organs.

[0073] For example, in the display of brain structure, only a certain layer of brain tissue can be seen in 2D images, while 3D images can clearly show different regions, neural connections and blood vessel distributions of the whole brain, greatly improving learners' understanding of complex structures.

[0074] Improvement Effect: Significantly improve learners' understanding of complex structures. For example, in the display of brain structure, 3D images can clearly show neural connections and blood vessel distributions.

[0075] (2) Differences in AI Interaction:

[0076] Existing medical knowledge dissemination methods lack intelligent interaction and are mainly one-way knowledge output. This invention introduces AI algorithms to achieve intelligent recognition and interaction of 3D images. Learners can not only obtain detailed knowledge by clicking on 3D image areas, but also ask questions by voice. For example, when asking "What are the common diseases of the kidneys?", the AI will associate relevant knowledge on the 3D kidney image and present it, which significantly enhances the initiative of learning and the convenience of knowledge acquisition.

[0077] Improvement effect: Learners can obtain relevant knowledge by clicking and asking questions by voice. For example, when asking "What are the common diseases of the kidneys?", the AI will associate relevant information on the 3D kidney image.

[0078] (3) Differences in dissemination methods:

[0079] Existing technology: Depends on hospital internal systems or simple data sharing methods.

[0080] This application: Achieves rapid sharing of visual medical data through network dissemination technology.

[0081] Improvement effect: Improves the dissemination efficiency of medical information and facilitates telemedicine and patient referral.

[0082] In summary, this application effectively makes up for the deficiencies of the existing technology in lacking spatial information and intelligent interaction by constructing 3D medical images and introducing AI interaction technology. From the perspective of learning effects, through experimental comparison, the depth of understanding and memory persistence of medical knowledge by those who use the method of this invention to learn medical knowledge are 80% higher than those using traditional two-dimensional images and text explanations, and it has great application value and promotion potential in scenarios such as medical education and doctor training.

[0083] The innovation points of this application are as follows:

[0084] (1) High-precision construction technology of 3D medical images: This is the basis for realizing visual dissemination and determines the quality and detail display ability of the images. High-quality 3D medical images can accurately restore the true shape and spatial structure of organs, providing an accurate data basis for subsequent AI recognition and knowledge dissemination. For example, in liver surgery simulation, a high-precision 3D liver model can help doctors clearly see the positional relationship between the blood vessels, bile ducts and diseased tissues inside the liver, so as to formulate a more precise surgical plan.

[0085] (2)AI and 3D medical image fusion and interaction technology: The realization of intelligent recognition and interaction through AI is the core innovation of the present invention, which greatly improves the dissemination effect. With the powerful computing power and deep learning ability of AI, two-way interaction between 3D medical images and users is achieved. Users can interact with the images naturally, such as voice inquiries and click queries, and the system can quickly and accurately give corresponding medical knowledge answers, making the learning and dissemination of medical knowledge more efficient and convenient.

[0086] (3)Data preprocessing and optimization algorithms: Ensure the quality of medical imaging data and improve the accuracy of 3D reconstruction and AI recognition. During the acquisition of medical imaging data, it is inevitable to introduce noise and artifacts, which will affect the reconstruction quality of 3D images and the accuracy of AI recognition. Through data preprocessing and optimization algorithms, noise can be effectively removed and the images can be corrected, providing high-quality data for subsequent 3D reconstruction and AI analysis, and indirectly improving the performance of the entire system.

[0087] Such as Figure 2 , the embodiments of the present application also provide a block diagram of a medical knowledge visualization dissemination system based on the interaction between 3D medical images and AI. The system may include:

[0088] An acquisition module, configured to acquire multi-dimensional medical imaging data of human organs or tissues from medical imaging devices;

[0089] A rendering module, configured to preprocess the acquired multi-dimensional medical imaging data, and then perform volume rendering on the preprocessed data using a three-dimensional reconstruction algorithm to generate high-precision three-dimensional medical images;

[0090] A processing module, configured to train a deep learning model with manually labeled medical image data, and process the three-dimensional medical images through the trained deep learning model to obtain intelligent interaction results; wherein, the deep learning model is used to identify different tissues and lesions in the images and obtain corresponding intelligent interaction results according to the interaction data provided by the user;

[0091] A dissemination module, configured to disseminate the three-dimensional medical images and their intelligent interaction results to target users in a visual form.

[0092] For the specific limitations of the medical knowledge visualization and dissemination system based on 3D medical image and AI interaction, reference can be made to the limitations of the medical knowledge visualization and dissemination method based on 3D medical image and AI interaction in the above text, which will not be elaborated here. Each module in the above medical knowledge visualization and dissemination system based on 3D medical image and AI interaction can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0093] In one embodiment, an electronic device is provided. The electronic device can be a computer, and its internal structural diagram can be as Figure 3 shown. The electronic device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the device is used to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used for medical knowledge visualization and dissemination data based on 3D medical image and AI interaction. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes a medical knowledge visualization and dissemination method based on 3D medical image and AI interaction.

[0094] Those skilled in the art can understand that the structure shown in Figure 3 is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0095] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, covering all or part of the processes in the method of the above embodiment.

[0096] In one embodiment, a computer program product is also provided, including a computer program / instructions, covering all or part of the processes in the method of the above embodiment.

[0097] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in M forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (SyMchliMk) DRAM (SLDRAM), memory bus (RaMbus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0098] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0099] The above embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it cannot be understood as a limitation to the scope of the patent application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A medical knowledge visualization and dissemination method based on the interaction between 3D medical images and AI, characterized in that, The method includes: Obtaining multi-dimensional medical image data of a human organ or tissue from a medical imaging device; Preprocessing the obtained multi-dimensional medical image data, and then performing volume rendering on the preprocessed data using a 3D reconstruction algorithm to generate a high-precision 3D medical image; Training a deep learning model with manually annotated medical image data, and processing the 3D medical image with the trained deep learning model to obtain an intelligent interaction result; wherein, the deep learning model is used to identify different tissues and lesions in the image, and obtain a corresponding intelligent interaction result according to the interaction data provided by the user; Disseminating the 3D medical image and its intelligent interaction result to the target user in a visual form.

2. The medical knowledge visualization and dissemination method according to claim 1, wherein The preprocessing includes denoising and artifact correction of the multi-dimensional medical image data, including: Applying a filtering algorithm to the obtained multi-dimensional medical image data to remove noise; Correcting the artifacts in the data through an image correction algorithm to improve the data quality.

3. The medical knowledge visualization and dissemination method according to claim 1, characterized in that Performing volume rendering on the preprocessed data using a 3D reconstruction algorithm to generate a high-precision 3D medical image, specifically including: Dividing the preprocessed multi-dimensional medical image data into voxel grids; Calculating the intersection points and normal vectors of the isosurfaces for each voxel grid; Generating the surface grid of the 3D model based on the intersection point and normal vector information using the Marching Cubes algorithm.

4. The medical knowledge visualization and dissemination method according to claim 1, wherein Training a deep learning model with manually annotated medical image data Collecting and organizing a large amount of manually annotated medical image data, and the annotation information includes tissue type and lesion type; Dividing the annotated data into a training set and a validation set; Training a CNN model using the training set, and optimizing the model parameters through the backpropagation algorithm; Evaluating and adjusting the trained model using the validation set.

5. The medical knowledge visualization and dissemination method according to claim 4, wherein Processing the 3D medical image with the trained deep learning model to obtain an intelligent interaction result, specifically including: Processing the 3D medical image with the trained deep learning model to achieve intelligent recognition of each region in the image; When the user clicks on a certain region of the 3D medical image or asks a question by voice, the deep learning model can identify the tissue type and lesion information of that region; The model retrieves the corresponding medical knowledge from a preset knowledge base according to the recognition result, and displays the retrieved medical knowledge to the user in text or voice form.

6. The medical knowledge visualization and dissemination method according to claim 1, characterized in that Disseminating the 3D model and the analyzed target medical information to the target user in a visual form includes: Disseminating the 3D medical image and its intelligent interaction result through a network platform, and users can access it through a browser or a dedicated client; Or displaying the 3D medical image and its intelligent interaction result through AR / VR devices.

7. A medical knowledge visualization and dissemination system based on the interaction between 3D medical images and AI, characterized in that, The system includes: An acquisition module for obtaining multi-dimensional medical image data of a human organ or tissue from a medical imaging device; A rendering module for preprocessing the obtained multi-dimensional medical image data, and then performing volume rendering on the preprocessed data using a 3D reconstruction algorithm to generate a high-precision 3D medical image; A processing module, configured to train a deep learning model with manually annotated medical image data, and process the three-dimensional medical image with the trained deep learning model to obtain an intelligent interaction result; wherein, the deep learning model is used to identify different tissues and lesions in the image, and obtain a corresponding intelligent interaction result according to the interaction data provided by the user. A propagation module, configured to propagate the three-dimensional medical image and its intelligent interaction result to a target user in a visual form.

8. An electronic device, characterized in that, It includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images