Medical image equipment and system, medical image processing method and equipment and medium
By integrating AI computing modules in medical imaging devices, AI device incompatibility and privacy leakage problems are solved, efficient AI computing and user-friendly image processing are achieved, and transformation costs are reduced.
Patent Information
- Application Number
- CN202410129702.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-30
- Publication Date
- 2025-08-01
AI Technical Summary
The separation of existing AI aid tools and medical imaging devices leads to incompatibility problems and has high cost of local area network transformation and risk of privacy leakage.
Integrate the AI computing module into medical imaging devices to realize the AI computing capabilities of the device itself, share hardware resources and route medical images to appropriate AI computing submodules through interfaces, use AI-specific computing devices and ensure independent functions through isolated containers.
Save the cost of local area network transformation, overcome device incompatibility problems, protect privacy, improve AI computing capabilities, and achieve convenient user control and report generation.
Smart Images
Figure CN120392129A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical devices, in particular to medical imaging devices, systems, and medical imaging processing methods, devices, and media. Background Art
[0002] Traditional medical imaging devices focus on high-quality imaging. Radiologists manually read and judge the images and then write a reading report. However, manual image reading has a large workload, is boring, and there is a possibility of omission and incorrect analysis. Recently, with the emergence of Artificial Intelligence (AI) assistance tools, AI-enhanced medical imaging devices have been applied.
[0003] Currently, AI assistance tools and medical imaging devices are separated from each other. AI assistance tools and medical imaging devices are manufactured by different device manufacturers respectively, and there is a lack of extensive integration testing between them, and compatibility problems often occur. Moreover, AI assistance tools and medical imaging devices are usually connected through a local area network (such as a local area network within a hospital) or the Internet (such as a medical cloud). However, the local area network-based solution requires network structure transformation and has cost problems. If the Internet solution is selected, image data needs to be transmitted to the cloud, which poses a risk of privacy leakage. Summary of the Invention
[0004] Embodiments of the present invention provide a medical imaging device, system, and medical imaging processing method, device, and medium.
[0005] A medical imaging device includes:
[0006] A medical image generation module for generating medical images;
[0007] An AI operation module for performing AI operations on the medical images to obtain operation results;
[0008] A sending module for sending the medical images and / or the operation results.
[0009] Therefore, by integrating the AI operation module into the medical imaging device, the medical imaging device itself has AI operation capabilities, and there is no need to access the AI operation capabilities through a local area network, saving the transformation cost of the local area network. Moreover, the medical imaging device integrated with the AI operation module is conducive to performing extensive integration testing and can overcome the compatibility problems between the AI operation module and the medical imaging device. In addition, by integrating the AI operation module into the medical imaging device, there is no need to send medical images to the cloud, which also protects privacy.
[0010] In one embodiment, the AI operation module includes N AI operation sub-modules, where N is a positive integer of at least 2, and the N AI operation sub-modules each have their own AI operation capabilities;
[0011] The medical imaging device includes:
[0012] An interface arranged between the medical imaging generation module and the N AI operation sub-modules for routing the medical image to the AI operation sub-module that processes the medical image based on the AI operation capability.
[0013] It can be seen that various AI operation capabilities are integrated in the medical imaging device, and the medical image is routed to the appropriate AI operation sub-module through the interface.
[0014] In one embodiment, the N AI operation sub-modules share the hardware resources of the medical imaging generation module.
[0015] Therefore, the N AI operation sub-modules share the hardware resources of the medical imaging generation module, saving costs.
[0016] In one embodiment, the AI operation module includes an AI dedicated operation device with hardware resources; N containers that run in isolation are in the AI dedicated operation device, and each container has its own container runtime environment; the N AI operation sub-modules are correspondingly integrated into their respective containers with the N container runtime environments.
[0017] It can be seen that using the AI dedicated operation device improves the AI operation ability. Moreover, through the isolated containers, the functional independence between the AI operation sub-modules is also ensured.
[0018] A medical imaging system includes:
[0019] A medical imaging device for generating a medical image; performing an AI operation on the medical image to obtain an operation result; sending the medical image and / or the operation result;
[0020] A Picture Archiving and Communication System (PACS) server for receiving the medical image and / or the operation result;
[0021] A workstation for obtaining the medical image and / or the operation result from the PACS server and displaying the medical image and / or the operation result.
[0022] Therefore, the medical imaging device itself has AI computing capabilities, eliminating the need to access AI computing capabilities through a local area network and saving the transformation cost of the local area network. Moreover, a medical imaging device integrated with AI computing capabilities is conducive to performing extensive integration tests, which can overcome the incompatibility issues between AI computing capabilities and medical imaging devices. Additionally, by integrating AI computing capabilities into the medical imaging device, there is no need to send medical images to the cloud, thus protecting privacy.
[0023] In one embodiment, the medical imaging device includes a medical image generation module and N AI computing sub-modules, where N is a positive integer of at least 2, and the N AI computing sub-modules each have their own AI computing capabilities.
[0024] The N AI computing sub-modules share the hardware resources of the medical image generation module; or
[0025] The medical imaging device includes an AI dedicated computing device with hardware resources; N containers that run in isolation are in operation in the AI dedicated computing device, and each container has its own container runtime environment; the N AI computing sub-modules are correspondingly integrated into their respective containers with the N container runtime environments.
[0026] It can be seen that the N AI computing sub-modules sharing the hardware resources of the medical image generation module saves costs. Moreover, the use of an AI dedicated computing device improves AI computing capabilities. Through the isolated containers, the functional independence between the AI computing sub-modules is also ensured.
[0027] A medical image processing method includes:
[0028] Generating a medical image based on a medical imaging device;
[0029] Performing AI computing on the medical image in the medical imaging device to obtain a computing result;
[0030] Sending the medical image and / or the computing result.
[0031] Therefore, the medical imaging device itself has AI computing capabilities, eliminating the need to access AI computing capabilities through a local area network and saving the transformation cost of the local area network. Moreover, a medical imaging device integrated with AI computing capabilities is conducive to performing extensive integration tests, which can overcome the incompatibility issues between AI computing capabilities and medical imaging devices. Additionally, by integrating AI computing capabilities into the medical imaging device, there is no need to send medical images to the cloud, thus protecting privacy.
[0032] In one embodiment, the medical imaging device includes N AI operator modules each having a respective identifier, and each AI operator module has its own AI computing capability, where N is a positive integer of at least 2; the method includes:
[0033] Receiving a first user instruction, where the first user instruction includes the identifier of the AI operator module that is the destination node of the medical image;
[0034] Based on the identifier, routing the medical image to the AI operator module having the identifier.
[0035] Therefore, based on the user instruction containing the identifier, the medical image can be routed to the AI operator module having the identifier, implementing a routing mechanism based on the user's active control behavior.
[0036] In one embodiment, the medical imaging device includes N AI operator modules each having a respective identifier, and each AI operator module has its own AI computing capability, where N is a positive integer of at least 2; the method includes:
[0037] Establishing a correspondence between the type of the medical image and the N AI operator modules;
[0038] Based on the correspondence, routing the medical image to the AI operator module corresponding to the type of the medical image.
[0039] Therefore, based on the correspondence between the type of the medical image and the AI operator module, the medical image can be routed to the AI operator module corresponding to the type of the medical image, implementing a routing mechanism based on the correspondence.
[0040] In one embodiment, the sending of the medical image and / or the operation result includes:
[0041] Encapsulating the medical image into a first Digital Imaging and Communications in Medicine (DICOM) file;
[0042] Encapsulating the operation result into a second DICOM file, where the second DICOM file includes the medical image, a first region, and a second region, the first region showing the region of interest in the medical image determined based on the operation result and / or auxiliary information of the region of interest, and the second region showing a sign that differentiates from the medical image;
[0043] Based on the DICOM protocol, sending the first DICOM file and / or the second DICOM file.
[0044] It can be seen that the first region is displayed in the second DICOM file to provide the user with the region of interest and its auxiliary information, facilitating the user to view the film. Moreover, the second region is displayed in the second DICOM file to facilitate the user to distinguish the original medical image from the medical image after AI operation and processing.
[0045] In one embodiment, it includes:
[0046] Display the first DICOM file and / or the second DICOM file;
[0047] Receive a second user instruction triggered during the process of the user browsing the first DICOM file and / or the second DICOM file;
[0048] Generate a film viewing report based on the second user instruction;
[0049] Send the film viewing report to the PACS server.
[0050] Therefore, based on the user instruction triggered during the process of browsing the first DICOM file and / or the second DICOM file, a film viewing report can be conveniently generated.
[0051] An electronic device includes:
[0052] A processor;
[0053] A memory for storing the executable instructions of the processor;
[0054] The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the medical image processing method described in any one of the above.
[0055] A computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the medical image processing method described in any one of the above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] The following will make the above and other features and advantages of the present invention clearer to those of ordinary skill in the art by referring to the accompanying drawings in detail. In the drawings:
[0057] Figure 1 is the first exemplary structural diagram of a medical imaging system with AI capabilities in the prior art.
[0058] Figure 2 is the second exemplary structural diagram of a medical imaging system with AI capabilities in the prior art.
[0059] Figure 3It is the first exemplary structural diagram of a medical imaging device according to an embodiment of the present invention.
[0060] Figure 4 It is the second exemplary structural diagram of a medical imaging device according to an embodiment of the present invention.
[0061] Figure 5 It is the exemplary structural diagram of a medical imaging system with AI capabilities according to an embodiment of the present invention.
[0062] Figure 6 It is the schematic diagram of a medical imaging processing procedure according to an embodiment of the present invention.
[0063] Figure 7A It is the schematic diagram of a medical image according to an embodiment of the present invention.
[0064] Figure 7B It is the schematic diagram of a medical image after AI operation according to an embodiment of the present invention.
[0065] Figure 8 It is the exemplary flowchart of a medical imaging processing method according to an embodiment of the present invention.
[0066] Figure 9 It is the exemplary structural diagram of an electronic device according to an embodiment of the present invention.
[0067] Among them, the reference numerals are as follows:
[0068]
[0069] Detailed implementation manners
[0070] To make the objectives, technical solutions, and advantages of the present invention clearer, the following examples are given to further elaborate on the present invention in detail. In this patent application, nouns and pronouns related to people are not limited to specific genders.
[0071] For the sake of simplicity and intuitiveness in description, the following elaborates on the solutions of the present invention by describing several representative embodiments. A large number of details in the embodiments are only used to help understand the solutions of the present invention. However, it is obvious that the technical solutions of the present invention can be implemented without being limited to these details. To avoid unnecessarily obscuring the solutions of the present invention, some embodiments are not described in detail but only present the frameworks. Hereinafter, "including" means "including but not limited to", and "according to..." means "at least according to..., but not limited to only according to...". Due to the language habits of Chinese, when the quantity of a component is not specifically indicated hereinafter, it means that the component can be one or more, or can be understood as at least one.
[0072] In the medical field, the development of AI-assisted technologies has brought convenience to doctors and patients. AI-assisted technologies mainly rely on deep learning and big data analysis. By training on a large amount of medical data, the AI module identifies data features and patterns, thus providing doctors with more comprehensive auxiliary information. Currently, AI-assisted technologies have achieved remarkable results in multiple fields, such as identifying regions of interest (such as lesions) from medical images, stitching multiple medical images (such as long bone stitching), and adjusting the angles of medical images, and so on.
[0073] In the prior art, AI-assisted tools and medical imaging devices are physically and logically separated.
[0074] Figure 1 is the first exemplary structural diagram of a prior art medical imaging system.
[0075] In Figure 1 : M medical imaging devices 11, 12... 1M are arranged in a local area network (such as a hospital's local area network), where M is a positive integer of at least 1. N AI computing nodes 21, 22... 2N are arranged in the same local area network, where N is a positive integer of at least 2. The M medical imaging devices 11, 12... 1M are respectively connected to the N AI computing nodes 21, 22... 2N through a router 20 in the local area network. Each of the M medical imaging devices 11, 12... 1M can access any of the N AI computing nodes 21, 22... 2N to obtain the AI computing power of the accessed AI computing node.
[0076] However, this processing method requires the transformation of the local area network (such as: expanding the capabilities of the router 20, arranging AI computing nodes in the local area network, and wiring for the AI computing nodes), which has cost issues and is also difficult to implement.
[0077] Figure 2 is the second exemplary structural diagram of a prior art medical imaging system.
[0078] In Figure 2 : M medical imaging devices 11, 12... 1M are connected to an AI computing module library 31 located in the cloud 30 via the Internet, where M is a positive integer of at least 1. The AI computing module library 31 contains various types of AI computing capabilities. Each of the M medical imaging devices 11, 12... 1M can access any of the AI computing capabilities in the AI computing module library 31.
[0079] However, this processing method requires sending the medical images provided by each of the M medical devices 11, 12... 1M to the cloud 30 via the Internet, which has security risks.
[0080] The applicant has found that: InFigure 1 the local area network solution shown and Figure 2 in the Internet solution shown, the AI computing power and the medical imaging device are physically and logically separated from each other. If the AI computing module capable of providing AI computing power is integrated into the medical imaging device, there is no need to modify the local area network, and the security risk of sending medical images to the cloud is also overcome.
[0081] The above disclosure details the technical defects existing in the prior art, the reasons for such technical defects, and the thinking and analysis process for overcoming such technical defects. In fact, the recognition of the above technical defects is not common knowledge in this field, but a novel discovery by the applicant in the research. In addition, the tracing of the reasons for such technical defects and the thinking and analysis process for overcoming such technical defects are also the step-by-step analysis results of the applicant in the actual research process, and are not common knowledge in this field.
[0082] Figure 3 is the first exemplary structural diagram of a medical imaging device according to an embodiment of the present invention. In Figure 3 it, the medical imaging device 100 includes: a medical image generation module 101 for generating medical images; an AI computing module 102 for performing AI computing on the medical images to obtain a computing result; and a sending module 103 for sending the medical images and / or the computing result.
[0083] The medical imaging device 100 can provide medical images of internal tissues and organs through means such as X-rays, electromagnetic fields, and ultrasonic waves. For example, the medical imaging device 100 may include:
[0084] (1) X-ray device: The X-ray device is one of the common medical imaging devices, usually used to obtain two-dimensional images of the skeletal system and certain tissue structures. It generates images of bones and tissues by emitting X-rays and receiving the reflected X-rays through a sensor. The X-ray device may include a traditional X-ray machine and a digital X-ray device.
[0085] (2) Computed tomography (CT) device: The CT device uses a rotating X-ray source and a sensor for multi-angle in-vivo imaging, and can generate cross-sectional digital images. CT scans can provide more detailed anatomical structure and tissue information, and are widely used in fields such as diagnosis and surgical planning.
[0086] (3) Magnetic resonance imaging (MRI) device: The MRI device obtains high-resolution internal images by using a strong magnetic field and radio waves, can provide detailed anatomical structure information, has a high resolution for soft tissues, and can obtain medical images of multiple planes.
[0087] (4) Ultrasonic device: The ultrasonic device uses ultrasonic waves to generate real-time internal images. It forms medical images by sending ultrasonic waves into the human body and receiving their echoes, based on the propagation and reflection of sound waves in different tissues.
[0088] The medical image generation module 101 in the medical imaging device 100 is used to generate medical images and is the main module in the medical imaging device 100. For example, when the medical imaging device 100 is implemented as an X-ray device, the medical image generation module 101 may include an X-ray generating component, a chest radiograph stand (Bucky-wall-stand, BWS) component, an examination table (table) component, a flat panel detector, and a control host located remotely. Details regarding the generation of medical images by the medical image generation module 101 can be referred to the known technologies in the art and will not be elaborated here.
[0089] In the medical imaging device 100, an AI operation module 102 is integrated. The AI operation module 102 is used to perform AI operations on the medical images generated by the medical image generation module 101 to obtain operation results. The AI operation module 102 can be arranged in any available physical space of the medical image generation module 101 and is connected to the medical image generation module 101 by means of cables. For example, the AI operation module 102 can be arranged at the bottom of the bed board of the examination table component, etc.
[0090] In the medical imaging device 100, a sending module 103 is also integrated. The sending module 103 is used to send medical images and / or operation results. Preferably, the sending module 103 sends medical images and / or operation results by wire to ensure privacy and security.
[0091] For example, the AI operation module 102 can provide multiple AI operation capabilities. For example, the AI operation capabilities provided by the AI operation module 102 may include:
[0092] (1) Identifying regions of interest from medical images (such as diseases like tuberculosis regions, cancer cell regions, or fracture regions, etc.);
[0093] (2) Stitching multiple medical images;
[0094] (3) Performing angle adjustment on medical images, etc.
[0095] In one embodiment, the AI computing module 102 can share the hardware resources of the medical image generation module 101 and utilize the hardware resources of the medical image generation module 101 to provide AI computing capabilities. For example, when the medical imaging device 100 is implemented as an X-ray device, the AI computing module 102 can call the graphics card in the control host to perform AI computing. It can be seen that the AI computing module 102 sharing the hardware resources of the medical image generation module 101 saves costs. In another embodiment, the AI computing module 102 can also be implemented as an AI dedicated computing device. Using an AI dedicated computing device can improve AI computing capabilities.
[0096] The above exemplary description presents typical examples of medical imaging devices and AI computing modules. Those skilled in the art can realize that such a description is merely exemplary and is not used to limit the protection scope of the embodiments of the present invention.
[0097] Figure 4 It is the second exemplary structural diagram of a medical imaging device according to an embodiment of the present invention. In Figure 4 it, the medical imaging device 100 includes: a medical image generation module 101 for generating medical images; an AI computing module 102 for performing AI computing on the medical images to obtain a computing result; and a sending module 103 for sending the medical images and / or the computing result.
[0098] The AI computing module 102 includes N AI computing sub-modules 401, 402... 40N, where N is a positive integer of at least 2. The N AI computing sub-modules 401, 402... 40N each have their own AI computing capabilities. For example, the AI computing sub-module 401 has the ability to identify regions of interest from medical images; the AI computing sub-module 402 has the ability to stitch multiple medical images; the AI computing sub-module 40N has the ability to perform angle adjustment on medical images, and so on.
[0099] The medical imaging device 100 includes: an interface 104 arranged between the medical image generation module 101 and the N AI computing sub-modules 401, 402... 40N for routing the medical images to the AI computing sub-module that processes the medical images based on the AI computing capabilities.
[0100] In one embodiment, a first user instruction is received via the user interface of the medical image generation module 101, and the first user instruction includes the identifier of the AI computing sub-module that is the destination node of the medical image. Then, based on the identifier, the interface 104 routes the medical image to the AI computing sub-module with the identifier. Therefore, based on the first user instruction containing the identifier, the medical image can be routed to the AI computing sub-module with the identifier, realizing a routing mechanism based on the user's active control behavior.
[0101] Example: Assume that the number of AI operator modules is 3, and their identifiers are AAA, BBB, and CCC respectively. After the medical image generation module 101 generates a medical image, a first user instruction is further received via the user interface of the medical image generation module 101. The first user instruction is used to indicate the specific AI operator module for performing AI operations on the medical image. Assume that the identifier AAA is included in the first user instruction, that is, it is indicated that the AI operator module AAA specifically performs the AI operation. The medical image generation module 101 sends the medical image and the first user instruction to the interface 104. The interface 104 parses the identifier AAA from the first user instruction and sends the medical image to the AI operator module AAA for the AI operator module AAA to perform AI operations on the medical image.
[0102] In one embodiment, a correspondence relationship between the type of medical image and the AI operator module is established and saved in the interface 104 in advance. Then, based on the correspondence relationship, the medical image is routed to the AI operator module corresponding to the type of the medical image. Therefore, based on the correspondence relationship between the type of medical image and the AI operator module, the medical image can be routed to the AI operator module corresponding to the type of the medical image, and a routing mechanism based on the correspondence relationship is implemented.
[0103] Example: Assume that the types of medical images include palm images, chest images, and leg bone images, and the number of AI operator modules is 3, namely: AI operator module 1 for identifying regions of interest from palm images, AI operator module 2 for identifying regions of interest from chest images, and AI operator module 3 for identifying regions of interest from leg bone images. A correspondence relationship between the type of medical image and the AI operator module is established and saved in the interface 104, that is, palm images correspond to AI operator module 1, chest images correspond to AI operator module 2, and leg bone images correspond to AI operator module 3. Then, when the interface 104 receives a medical image from the medical image generation module 101, based on the correspondence relationship, the medical image is routed to the AI operator module corresponding to the type of the medical image.
[0104] In one embodiment, N artificial intelligence operator modules 401, 402... 40N share the hardware resources of the medical image generation module 101. In another embodiment, the medical imaging device 100 includes an AI dedicated computing device 109 with hardware resources. N containers running in isolation are provided in the AI dedicated computing device 109, and each container has its own container runtime environment 501, 402... 50N. The N AI operator modules 401, 402... 40N are correspondingly integrated into their respective containers with the N container runtime environments 501, 402... 50N. For example, the AI operator module 401 and the container runtime environment 501 are integrated into the first container; the AI operator module 402 and the container runtime environment 502 are integrated into the second container... The AI operator module 40N and the container runtime environment 50N are integrated into the Nth container. For example, the AI dedicated computing device 109 can be implemented as an AI computing board with a Graphics Processing Unit (GPU). It can be seen that the use of the AI dedicated computing device improves the AI computing power. Moreover, through the isolated containers, the functional independence between the AI operator modules is also ensured.
[0105] Figure 5 is a schematic structural diagram of a medical imaging system with AI capabilities according to an embodiment of the present invention. In Figure 5 the medical imaging system includes: a medical imaging device 100 for generating medical images, performing AI operations on the medical images to obtain operation results; and sending the medical images and / or operation results. A PACS server 106 for receiving the medical images and / or operation results. A workstation 107 for obtaining the medical images and / or operation results from the PACS server 106, displaying the medical images and / or operation results; generating a reading report 108, and sending the reading report 108 to the PACS server 106. Wherein: the medical imaging device 100 can be specifically implemented as a medical imaging device 100 having Figure 3 or Figure 4 structure.
[0106] In one embodiment, the medical imaging device 100 includes a medical image generation module 101 and N AI operator modules 401, 402... 40N, where N is a positive integer of at least 2, and the N AI operator modules 401, 402... 40N each have their own AI capabilities; the N AI operator modules 401, 402... 40N share the hardware resources of the medical image generation module 101.
[0107] In one embodiment, the medical imaging device 100 includes a medical imaging generation module 101 and N AI operator modules 401, 402... 40N, where N is a positive integer of at least 2, and the N AI operator modules 401, 402... 40N each have their own AI computing capabilities; the medical imaging device 100 is an AI dedicated computing device 109 with hardware resources; N containers running in isolation are provided in the AI dedicated computing device 109, and each container has its own container runtime environment 501, 402... 50N; the N artificial intelligence operator modules 401, 402... 40N are correspondingly integrated into their respective containers with the N container runtime environments 501, 402... 50N.
[0108] Taking the specific implementation of the medical imaging device as an X-ray imaging device as an example, the medical image processing process of the embodiment of the present invention will be described below.
[0109] Figure 6 It is a schematic diagram of the medical image processing process according to the embodiment of the present invention. The X-ray imaging device 40 performs X-ray imaging on the imaging target to obtain an X-ray image 41. An AI computing module 42 is integrated in the X-ray imaging device 40. For example, the computing capability of the AI computing module 42 is to identify the region of interest (such as a fracture region) from the X-ray image. The AI computing module 42 may include an AI computing capability 422 and a DICOM protocol encapsulation capability 421. The AI computing module 42, through the AI computing capability 422, identifies the region of interest from the X-ray image 41 (preferably, the confidence level of the region of interest can also be calculated) to obtain the AI computing result 43 of the X-ray image. In the AI computing result 43 of the X-ray image, it includes the X-ray image 41, the region of interest identified from the X-ray image 41, and the confidence level of the region of interest. Moreover, the AI computing module 42, through the DICOM protocol encapsulation capability 421, encapsulates the AI computing result 43 of the X-ray image into the DICOM format.
[0110] The X-ray imaging device 40 encapsulates the X-ray image 41 into the DICOM format, and sends the X-ray image 41 encapsulated in the DICOM format and the AI computing result 43 encapsulated in the DICOM file format as the upload content 45 to the PACS server 44. The user at the workstation 46 can obtain the upload content 45 from the PACS server 44.
[0111] The uploaded content 45 is displayed in the PACS server 44. For example, in the user interface of the PACS server 44, the X-ray image 41 and the AI operation result 43 are displayed side by side, facilitating the user to view the film. The workstation 46 receives the second user instruction triggered during the process of the user browsing the uploaded content 45, and generates a film viewing report 47 based on the second user instruction. The workstation 46 sends the film viewing report 47 to the PACS server 44.
[0112] The above takes the medical imaging device specifically implemented as an X-ray imaging device as an example to demonstrate the embodiments of the present invention. Those skilled in the art can realize that this description is only exemplary and is not used to limit the protection scope of the embodiments of the present invention.
[0113] Figure 7A It is a schematic diagram of a medical image according to an embodiment of the present invention. Figure 7B It is a schematic diagram of a medical image after AI operation according to an embodiment of the present invention.
[0114] It can be seen that the AI operation result 60 of the X-ray image includes: (1) the X-ray image 50; (2) the first region 61, which highlights the region of interest (for example, the fracture part and the credibility of the fracture part); (3) the second region 62, which highlights the signs that distinguish from the X-ray image 50, such as the text signs "AI map", "AI content", and so on.
[0115] Figure 8 It is a schematic flowchart of a medical image processing method according to an embodiment of the present invention. As Figure 8 shown, the medical image processing method includes:
[0116] Step 701: Generate a medical image based on a medical imaging device.
[0117] Step 702: Perform an AI operation on the medical image in the medical imaging device to obtain an operation result.
[0118] Step 703: Send the medical image and / or the operation result.
[0119] In one embodiment, the medical imaging device includes N AI operation sub-modules with their respective identifiers, and each AI operation sub-module has its own AI operation ability, where N is a positive integer of at least 2. The method includes: receiving a first user instruction, where the first user instruction includes the identifier of the AI operation sub-module that is the destination node of the medical image; routing the medical image to the AI operation sub-module with the identifier based on the identifier.
[0120] In one embodiment, a medical imaging device includes N AI operator modules each having a respective identifier, and each AI operator module has its own AI computing capability, where N is a positive integer of at least 2. The method includes: establishing a correspondence between the types of medical images and the N AI operator modules; and routing the medical images to the AI operator module corresponding to the type of the medical images based on the correspondence.
[0121] In one embodiment, sending the medical images and / or the computing results includes: encapsulating the medical images into a first DICOM file; encapsulating the computing results into a second DICOM file, where the second DICOM file includes the medical images, a first region, and a second region, the first region shows the region of interest in the medical images and / or the auxiliary information (such as confidence) of the region of interest determined based on the computing results, and the second region shows a sign that differentiates from the medical images; and sending the first DICOM file and / or the second DICOM file based on the DICOM protocol.
[0122] In one embodiment, it includes: displaying the first DICOM file and / or the second DICOM file; receiving a second user instruction triggered during the user's browsing of the first DICOM file and / or the second DICOM file; generating a film reading report based on the second user instruction; and sending the film reading report to a PACS server.
[0123] An embodiment of the present invention also proposes an electronic device having a processor-memory architecture. Figure 9 It is a schematic structural diagram of the electronic device according to the embodiment of the present invention. As Figure 9 shown, the electronic device 800 includes a processor 801, a memory 802, and a computer program stored on the memory 802 and executable on the processor 801. When the computer program is executed by the processor 801, it implements any one of the above medical image processing methods. Among them, the memory 802 can be specifically implemented as various storage media such as an electrically erasable programmable read-only memory (EEPROM), a flash memory, and a programmable read-only memory (PROM). The processor 801 can be implemented as including one or more central processing units or one or more field programmable gate arrays, where the field programmable gate array integrates one or more central processing unit cores. Specifically, the central processing unit or the central processing unit core can be implemented as a CPU, an MCU, a GPU, or a DSP, and so on.
[0124] It should be noted that not all steps and modules in the above-mentioned processes and structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted according to needs. The division of each module is only for the convenience of description and is a functional division. In actual implementation, one module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device or in different devices.
[0125] The hardware modules in each embodiment can be implemented mechanically or electronically. For example, a hardware module can include a specially designed permanent circuit or logic device (such as a dedicated processor, such as an FPGA or ASIC) for performing specific operations. A hardware module can also include a programmable logic device or circuit (such as including a general-purpose processor or other programmable processors) temporarily configured by software for performing specific operations. As for whether to specifically adopt a mechanical method, a dedicated permanent circuit, or a temporarily configured circuit (such as configured by software) to implement the hardware module, it can be determined according to cost and time considerations.
[0126] The above is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A medical imaging device (100), characterized in that, Comprising; A medical image generation module (101) for generating medical images; An artificial intelligence operation module (102) for performing artificial intelligence operations on the medical images to obtain operation results; A sending module (103) for sending the medical images and / or the operation results.
2. The medical imaging device (100) according to claim 1, characterized in that, The artificial intelligence operation module (102) includes N artificial intelligence operation sub-modules (401, 402... 40N), where N is a positive integer of at least 2, and the N artificial intelligence operation sub-modules (401, 402... 40N) each have their own artificial intelligence operation capabilities; The medical imaging device (100) includes: An interface (104) arranged between the medical image generation module (101) and the N artificial intelligence operation sub-modules (401, 402... 40N) for routing the medical images to the artificial intelligence operation sub-module that processes the medical images based on the artificial intelligence operation capabilities.
3. The medical imaging device (100) according to claim 2, characterized in that, The N artificial intelligence operation sub-modules (401, 402... 40N) share the hardware resources of the medical image generation module (101).
4. The medical imaging device (100) according to claim 2, characterized in that, The artificial intelligence operation module (102) includes a dedicated artificial intelligence operation device (109) with hardware resources; N isolated containers are running in the dedicated artificial intelligence operation device (109), and each container has its own container runtime environment (501, 402... 50N); the N artificial intelligence operation sub-modules (401, 402... 40N) are correspondingly integrated into their respective containers with the N container runtime environments (501, 402... 50N).
5. A medical imaging system, characterized in that, Including: A medical imaging device (100) for generating medical images; performing artificial intelligence operations on the medical images to obtain operation results; sending the medical images and / or the operation results; An image archiving and communication system server (106) for receiving the medical images and / or the operation results; A workstation (107) for obtaining the medical images and / or the operation results from the image archiving and communication system server (106) and displaying the medical images and / or the operation results.
6. The medical imaging system according to claim 5, wherein, The medical imaging device (100) includes a medical image generation module (101) and N artificial intelligence operation sub-modules (401, 402... 40N), where N is a positive integer of at least 2, and the N artificial intelligence operation sub-modules (401, 402... 40N) each have their own artificial intelligence operation capabilities; The N artificial intelligence operation sub-modules (401, 402... 40N) share the hardware resources of the medical image generation module (101); or The medical imaging device (100) includes an artificial intelligence dedicated computing device (109) having hardware resources; N containers running in isolation are provided in the artificial intelligence dedicated computing device (109), and each container has its own container runtime environment (501, 402... 50N); the N artificial intelligence computing sub-modules (401, 402... 40N) are correspondingly integrated into their respective containers with the N container runtime environments (501, 402... 50N).
7. A medical image processing method, characterized in that, Including: Based on the medical imaging device, generate a medical image (701); In the medical imaging device, perform artificial intelligence operations on the medical image to obtain an operation result (702); Send the medical image and / or the operation result (703).
8. The method according to claim 7, wherein The medical imaging device includes N artificial intelligence computing sub-modules each having its own identifier, and each artificial intelligence computing sub-module has its own artificial intelligence computing ability, where N is a positive integer of at least 2; the method includes: Receive a first user instruction, where the first user instruction includes the identifier of the artificial intelligence computing sub-module that is the destination node of the medical image; Based on the identifier, route the medical image to the artificial intelligence computing sub-module having the identifier.
9. The method according to claim 7, characterized in that The medical imaging device includes N artificial intelligence computing sub-modules each having its own identifier, and each artificial intelligence computing sub-module has its own artificial intelligence computing ability, where N is a positive integer of at least 2; the method includes: Establish a correspondence relationship between the type of the medical image and the N artificial intelligence computing sub-modules; Based on the correspondence relationship, route the medical image to the artificial intelligence computing sub-module corresponding to the type of the medical image.
10. The method according to claim 7, wherein The sending the medical image and / or the operation result (703) includes: Encapsulate the medical image into a first Digital Imaging and Communications in Medicine (DICOM) file; Encapsulate the operation result into a second DICOM file, where the second DICOM file includes the medical image, a first region, and a second region, the first region shows the region of interest in the medical image determined based on the operation result and / or auxiliary information of the region of interest, and the second region shows a flag that differentiates from the medical image; Based on the DICOM protocol, send the first DICOM file and / or the second DICOM file.
11. The method according to any one of claims 7-10, characterized in that, Including: Display the first DICOM file and / or the second DICOM file; Receive a second user instruction triggered during the user's browsing of the first DICOM file and / or the second DICOM file; Generate a radiology report based on the second user instruction; Send the radiology report to a Picture Archiving and Communication System (PACS) server.
12. An electronic device, characterized in that, Including: A processor (801); A memory (802) for storing executable instructions of the processor (801); The processor (801) is configured to read the executable instructions from the memory (802) and execute the executable instructions to implement the medical image processing method according to any one of claims 7-11.
13. A computer-readable storage medium having computer instructions stored thereon, characterized in that, When the computer instructions are executed by a processor, the medical image processing method according to any one of claims 7-11 is implemented.