A control method and device of a medical dynamic X-ray collimator

Abstract

The specification discloses a control method and device of a medical dynamic X-ray collimator. First, continuous multiple frames of X-ray images collected by a medical dynamic X-ray acquisition device are acquired. For each frame of X-ray image, each region of interest is identified and labeled from the frame of X-ray image. The labeled frames of X-ray image are input into a pre-trained change analysis model. For each region of interest included in each frame of X-ray image, the image change amount corresponding to the region of interest is determined according to the image of the region of interest included in different frames of X-ray image. According to the image change amount corresponding to each region of interest, the region of interest with an image change amount less than a set threshold is determined as a covering region. The medical dynamic X-ray collimator is controlled according to the covering region, so that the medical dynamic X-ray collimator blocks the medical X-ray emitted by the medical dynamic X-ray device and irradiated in the covering region.

Description

Technical Field

[0001] This specification relates to the medical field, and in particular to a control method and device for a medical dynamic X-ray collimator. Background Technology

[0002] During surgery, various medical radiations (such as X-rays and gamma rays) are often used. Since medical radiation can cause damage to the human body, it is necessary to minimize the area of ​​the body exposed to medical radiation during surgery in order to reduce the harm caused by medical radiation.

[0003] Currently, to minimize the area exposed to medical radiation, it is first necessary to determine the body parts involved in the surgery and the areas through which surgical instruments will pass, and then use shielding materials to block medical radiation from irradiating areas outside these areas. For example, during heart surgery, shielding materials are used to cover areas other than the heart region to reduce the harmful effects of X-rays. Similarly, in heart surgery where surgical instruments need to pass from the lungs to the heart region, shielding materials can be used to cover both the lung area and areas outside the heart region.

[0004] While this method can reduce the area exposed to medical radiation to some extent, the surgical instruments are constantly moving during the procedure, and the areas where the instruments leave may not require X-ray exposure. For example, during heart surgery, the surgical instruments enter the heart area from the lungs, and the lung area may not require X-ray exposure when the instruments leave the lungs.

[0005] Therefore, how to further reduce the harm of medical radiation to the human body is an urgent problem to be solved. Summary of the Invention

[0006] This specification provides a control method and device for a medical dynamic X-ray collimator, which partially solves the above-mentioned problems existing in the prior art.

[0007] The following technical solution is adopted in this specification:

[0008] This specification provides a control method for a medical dynamic X-ray collimator, including:

[0009] Acquire continuous multi-frame X-ray images from medical dynamic X-ray equipment;

[0010] For each frame of X-ray image, identify and label the regions of interest from that frame of X-ray image;

[0011] The labeled X-ray images are input into a pre-trained change analysis model to determine the amount of image change corresponding to each region of interest contained in each X-ray image frame, based on the images of that region of interest contained in different X-ray images.

[0012] Based on the amount of image change corresponding to each region of interest, regions of interest with image change less than a set threshold are identified and designated as occlusion regions.

[0013] The medical dynamic X-ray collimator is controlled according to the covered area to block the medical rays emitted by the medical X-ray device that irradiate the covered area.

[0014] Optionally, regions of interest are identified and labeled from the X-ray image frame, specifically including:

[0015] For each frame of X-ray image, the frame of X-ray image is input into a pre-trained region recognition model to determine and label the regions of interest contained in the frame of X-ray image.

[0016] Optionally, the training region recognition model specifically includes:

[0017] Acquire the first sample X-ray image;

[0018] The first sample X-ray image is input into the region recognition model to be trained, and the region of interest involving the target object in the first sample X-ray image is marked.

[0019] The region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest identified by the region recognition model in the first sample X-ray image and the region of interest marked in the first sample X-ray image.

[0020] Optionally, the region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest identified by the region recognition model in the first sample X-ray image and the region of interest marked in the first sample X-ray image. Specifically, this includes:

[0021] If the number of regions of interest in the first sample X-ray image is less than the preset number of regions of interest, then at least one region in the first sample X-ray image that does not involve the target object is identified as an interference region.

[0022] The region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest (ROI) identified by the region recognition model in the first sample X-ray image and the ROI marked in the first sample X-ray image, and maximizing the deviation between the ROI identified by the region recognition model in the first sample X-ray image and the interference region.

[0023] Optionally, the target object includes surgical instruments and human organs.

[0024] Optionally, training a change analysis model includes:

[0025] Acquire a second sample X-ray image and a third sample X-ray image, wherein the second sample X-ray image and the third sample X-ray image are images of the same region;

[0026] The second sample X-ray image and the third sample X-ray image are input into the change analysis model to obtain the image change between the second sample X-ray image and the third sample X-ray image, which is used as the image change to be optimized.

[0027] Based on the changes in the image to be optimized, the second sample X-ray image is transformed to obtain a transformed X-ray image;

[0028] The change analysis model is trained with the optimization objective of minimizing the deviation between the converted X-ray image and the third sample X-ray image.

[0029] Optionally, the method further includes:

[0030] If the detected change in the entire X-ray image exceeds a set threshold, the irradiation range of the medical X-ray emitted by the medical X-ray device is expanded by controlling the medical dynamic X-ray collimator.

[0031] This manual provides a control device for a medical dynamic X-ray collimator, including:

[0032] The acquisition module is used to acquire multiple consecutive X-ray images collected by the medical dynamic X-ray equipment;

[0033] The recognition module is used to identify and label the regions of interest in each frame of X-ray image.

[0034] The input module is used to input the labeled X-ray images of each frame into a pre-trained change analysis model, so as to determine the image change corresponding to each region of interest contained in each X-ray image frame based on the images of that region of interest contained in different X-ray images.

[0035] The determination module is used to determine the regions of interest whose image changes are less than a set threshold based on the image change amount corresponding to each region of interest, and to use these regions as occlusion areas.

[0036] The control module is used to control the medical dynamic X-ray collimator according to the covered area, so as to block the medical rays emitted by the medical X-ray device that irradiate the covered area.

[0037] This specification provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described control method for a medical dynamic X-ray collimator.

[0038] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the control method of the above-described medical dynamic X-ray collimator.

[0039] The above-mentioned technical solutions adopted in this specification can achieve the following beneficial effects:

[0040] This specification provides a control method for a medical dynamic X-ray collimator. First, multiple consecutive frames of X-ray images acquired by the medical dynamic X-ray device are obtained. For each frame, regions of interest (ROIs) are identified and labeled. The labeled X-ray images are then input into a pre-trained variation analysis model. For each ROI contained in each X-ray image frame, the corresponding image variation is determined based on the images of that ROI in different X-ray images. Based on the image variation corresponding to each ROI, ROIs with image variation less than a set threshold are identified as masking regions. The medical dynamic X-ray collimator is then controlled according to these masking regions to block the medical radiation emitted by the medical X-ray device from irradiating the masked regions.

[0041] As can be seen from the above method, by using the aforementioned control method for a medical dynamic X-ray collimator, the image change amount of each frame of X-ray image can be determined using a pre-trained change analysis model, and based on the image change amount, the shielding area with smaller changes can be determined. Since areas with smaller image changes can be considered as areas not involved in the current surgical procedure, the medical dynamic X-ray collimator can be controlled to block the medical rays that would otherwise irradiate the shielded area, thereby effectively reducing the harm of medical rays to the human body. Attached Figure Description

[0042] The accompanying drawings, which are included to provide a further understanding of this specification and form part of this specification, illustrate exemplary embodiments and are used to explain this specification, but do not constitute an undue limitation thereof. In the drawings:

[0043] Figure 1 This is a flowchart illustrating a control method for a medical dynamic X-ray collimator provided in this specification;

[0044] Figure 2 This is a schematic diagram illustrating the process of training a change analysis model as provided in this specification.

[0045] Figure 3 This is a schematic diagram of an embodiment of a control method for a medical dynamic X-ray collimator provided in this specification;

[0046] Figure 4 A schematic diagram of a control device for a medical dynamic X-ray collimator provided in this specification;

[0047] Figure 5 This specification provides a corresponding Figure 1 A schematic diagram of an electronic device. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0049] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0050] Figure 1 This is a flowchart illustrating a control method for a medical dynamic X-ray collimator provided in this specification, including the following steps:

[0051] S101: Acquire multiple consecutive X-ray images from a medical dynamic X-ray device.

[0052] In this manual, the subject implementing the control method of the medical dynamic X-ray collimator can be a desktop computer, a medical dynamic X-ray acquisition device, or other designated equipment. For ease of description, this manual will only use designated equipment as the subject to describe the control method of the medical dynamic X-ray collimator in detail.

[0053] During surgery, various medical radiations (such as X-rays and gamma rays) are often used to irradiate the human body to obtain images needed for the procedure. However, because irradiation with these medical radiations can cause harm, it is essential to minimize the area of ​​the body exposed to them during surgery.

[0054] Currently, the main way to reduce the area of ​​the human body exposed to medical radiation is to use a medical dynamic X-ray collimator to shield parts of the body not involved in the surgery and areas that surgical instruments do not pass through during the operation.

[0055] While this method can reduce the area of ​​the body exposed to medical radiation to some extent, surgical instruments are constantly moving during surgery, and areas where the instruments have moved away may not require exposure to medical radiation. This increases the area of ​​unnecessary exposure to medical radiation.

[0056] To address the aforementioned technical problems, this specification provides a control method for a medical dynamic X-ray collimator. First, a designated device acquires multiple consecutive frames of X-ray images from the medical dynamic X-ray acquisition device. Each acquired X-ray image involves at least one target object, which may include: surgical instruments (e.g., balloon markers, guiding catheters, guide lines, intravascular ultrasound transducers, blood vessels, ablation catheter electrodes, peripheral mapping catheter electrodes, coronary sinus catheter electrodes, etc.) and human organs (e.g., bone tissue, heart, head, limbs, abdomen, neck, etc.).

[0057] S102: For each frame of X-ray image, identify and label each region of interest from that frame of X-ray image.

[0058] After acquiring an X-ray image, the designated device can identify and label each region of interest in each frame of the X-ray image. The region of interest can be a region in the X-ray image that involves at least one target object.

[0059] It should be noted that, in order to improve the efficiency of identifying regions of interest from X-ray images and the efficiency of annotation, for each frame of X-ray image, the designated device can input the frame of X-ray image into a pre-trained region recognition model to determine and annotate the regions of interest contained in the frame of X-ray image.

[0060] Regarding the process of training the aforementioned region recognition model, a designated device can acquire a first sample X-ray image, which can be a historical X-ray image acquired by the X-ray device or a sample of a constructed X-ray image.

[0061] After acquiring the first sample X-ray image, the designated device determines the region of interest corresponding to the target object marked in the first sample X-ray image, and inputs the first sample X-ray image into the region recognition model to be trained, so as to obtain the region of interest contained in the first sample X-ray image recognized by the region recognition model.

[0062] The designated device can train the region recognition model with the optimization objective of minimizing the deviation between the region of interest (ROI) identified by the region recognition model in the first sample X-ray image and the ROI labeled in the first sample X-ray image. This allows the trained region recognition model to accurately identify the ROI (i.e., the image region containing the specified target object) in X-ray images in practical applications.

[0063] In actual training, to further improve the accuracy of the region recognition model in identifying regions of interest (ROIs), if the number of ROIs identified by the designated device in the first sample X-ray image after labeling them is less than the preset maximum number of ROIs, then at least one region that does not involve the target object is identified in the first sample X-ray image as an interference region. For example, if the preset maximum number of ROIs is 10, and the number of ROIs identified in the first sample X-ray image is 8, then the designated device can identify 2 regions that do not involve the target object in the first sample X-ray image as interference regions. It should be noted that the sum of the ROIs and interference regions identified by the designated device in the first sample X-ray image cannot exceed the preset maximum number of ROIs.

[0064] Based on this, after the designated device acquires the region of interest and interference regions contained in the first sample X-ray image identified by the region recognition model, it can train the region recognition model with the optimization objective of minimizing the deviation between the region of interest contained in the first sample X-ray image identified by the region recognition model and the region of interest marked in the first sample X-ray image, and maximizing the deviation between the region of interest contained in the first sample X-ray image identified by the region recognition model and the interference regions.

[0065] In this specification, various loss functions can be used to train the aforementioned region recognition model on the specified device. For example, cross-entropy loss can be used to train the region recognition model. This specification does not restrict the specific loss function used.

[0066] In practical applications, region recognition models can identify all image regions contained in an X-ray image and output the probability that these image regions belong to regions of interest. The sensitivity of the region recognition model can be controlled by setting a probability threshold for a specific device; that is, image regions with a probability greater than the probability threshold are considered as regions of interest ultimately identified by the region recognition model.

[0067] S103: Input the labeled X-ray images of each frame into a pre-trained change analysis model to determine the image change amount corresponding to each region of interest contained in each X-ray image frame, based on the images of that region of interest contained in different X-ray images.

[0068] The designated device can input the labeled X-ray images of each frame into a pre-trained change analysis model. For each region of interest (ROI) within each X-ray image frame, the model determines the image change corresponding to that ROI based on the images of that ROI in different X-ray images. This image change can represent the variation in differences between different X-ray images. Specifically, it can be the change in pixel values ​​between different X-ray images, or it can be obtained from the change matrix of the X-ray image. This change matrix can be understood as the change in pixel values ​​at each location in the X-ray image. Its specific function is to obtain the changed image by superimposing this change matrix onto the X-ray image. Alternatively, the optical flow (i.e., the instantaneous velocity of pixel movement at each location in the X-ray image) can be used as the image change of that X-ray image.

[0069] The following is combined Figure 2 The training of the aforementioned change analysis model will be explained.

[0070] Figure 2 This document provides a schematic diagram illustrating the process of training a change analysis model.

[0071] The designated device can input the second and third sample X-ray images into a change analysis model after acquiring them, and obtain the image change between the second and third sample X-ray images as the image change to be optimized. The second and third sample X-ray images are images of the same area.

[0072] The specified device can transform the second sample X-ray image based on the amount of change in the image to be optimized, resulting in a transformed X-ray image. The specified device can then train the change analysis model with the optimization objective of minimizing the deviation between the transformed X-ray image and the third sample X-ray image.

[0073] It should be noted that the first, second, and third sample X-ray images mentioned above are mainly used to distinguish the data used in different model training processes. In practical applications, the first, second, and third sample X-ray images can all come from the same dataset. That is, the same data source can be used to train the region recognition model and the change analysis model.

[0074] Of course, the second and third sample X-ray images can belong to the same dataset, while the first sample X-ray image can belong to another dataset. That is, different data sources can be used to train the region recognition model and the change analysis model. The terms "first," "second," and "third" themselves do not have any other special meaning.

[0075] S104: Based on the image change amount corresponding to each region of interest, determine the regions of interest whose image change amount is less than a set threshold, and use them as occlusion regions.

[0076] During surgery, certain procedures may not involve areas previously irradiated with medical radiation. For example, in heart surgery, surgical instruments may need to enter through the lungs, requiring medical radiation to reach the lungs during this process. However, when the surgical instruments are performing their function within the heart, the lungs may not require further radiation.

[0077] To reduce exposure to medical radiation, the designated equipment, after acquiring the image changes corresponding to each region of interest, identifies regions of interest where the image changes are less than a set threshold, and designates these regions as masked areas. These masked areas can be understood as regions that do not show significant changes in the image over a period of time; they can be considered areas that do not require surgery temporarily. Therefore, in subsequent procedures, the medical radiation that would otherwise irradiate these areas can be blocked using a medical dynamic X-ray collimator.

[0078] S105: Based on the covered area, control the medical dynamic X-ray collimator to block the medical rays emitted by the medical X-ray device that irradiate the covered area.

[0079] The designated device can control the medical dynamic X-ray collimator according to the coverage area, so as to block the medical rays emitted by the medical X-ray equipment that irradiate the coverage area. The medical dynamic X-ray collimator can be a component mounted on the head of the equipment emitting medical rays, and its main function is to limit the range of electron radiation.

[0080] In practical applications, since the area outside the region of interest cannot be irradiated by medical X-rays, and the operation may involve the area outside the region of interest, after a preset time period, the designated device can expand the irradiation range of medical X-rays by controlling the medical dynamic X-ray collimator and redetermine the region of interest to repeat the above method.

[0081] In practical applications, during surgery, surgical instruments may move too fast and go outside the region of interest. Therefore, to improve surgical safety, if the image change of the X-ray image detected by the designated equipment exceeds a set threshold, the irradiation range of the medical X-ray emitted by the medical X-ray equipment can be expanded by controlling the medical dynamic X-ray collimator.

[0082] The "detection of X-ray image changes exceeding a set threshold" mentioned here can specifically refer to the current X-ray image frame having an image change greater than the previous frame, or the N X-ray images (N being a positive integer greater than 1) preceding the current X-ray image frame having an image change greater than the set threshold.

[0083] Specifically, expanding the irradiation range of medical radiation can be achieved by having the designated equipment control a dynamic medical X-ray collimator to make the irradiation range larger than before. Alternatively, the designated equipment can use the entire X-ray image as the region of interest, and by controlling the dynamic medical X-ray collimator, ensure that medical radiation irradiates the entire area covered by the X-ray image.

[0084] The following describes a control method for a medical dynamic X-ray collimator provided in this specification, with reference to an embodiment.

[0085] Figure 3 A schematic diagram illustrating an embodiment of a control method for a medical dynamic X-ray collimator provided in this specification:

[0086] First, the designated device can acquire multiple consecutive frames of X-ray images collected by the X-ray device. Then, the acquired X-ray images are input into the region recognition model trained according to the above method to determine the region of interest contained in each X-ray image.

[0087] The designated device can input multiple consecutive X-ray images of the region of interest into a change analysis model trained using the above method to determine the amount of image change. The designated device can then determine the obscured area based on the image change and control the medical dynamic X-ray collimator accordingly to block the medical X-rays emitted by the medical X-ray equipment from irradiating the obscured area.

[0088] The designated device will then reacquire multiple consecutive frames of X-ray images and repeat the above process.

[0089] As can be seen from the above method, the designated equipment can use a pre-trained change analysis model to determine the amount of image change in each frame of X-ray images, and based on the amount of image change, determine the shielding area with smaller changes. Since areas with smaller image changes can be considered as areas not involved in the current surgical procedure, the medical dynamic X-ray collimator can be controlled to block medical rays that would normally irradiate the shielded area from irradiating that area, thereby effectively reducing the harm of medical rays to the human body.

[0090] The above describes the control methods for one or more medical dynamic X-ray collimators described in this manual. Based on the same concept, this manual also provides corresponding control devices for medical dynamic X-ray collimators, such as... Figure 4 As shown.

[0091] Figure 4 This specification provides a schematic diagram of a control device for a medical dynamic X-ray collimator, including:

[0092] The acquisition module 401 is used to acquire multiple consecutive X-ray images collected by the medical dynamic X-ray equipment;

[0093] The recognition module 402 is used to identify and label each region of interest from each frame of X-ray image;

[0094] The input module 403 is used to input the labeled X-ray images of each frame into a pre-trained change analysis model, so as to determine the image change amount corresponding to each region of interest contained in each X-ray image frame based on the images of the region of interest contained in different X-ray images.

[0095] The determination module 404 is used to determine the regions of interest whose image changes are less than a set threshold based on the image change amount corresponding to each region of interest, and to use them as occlusion regions.

[0096] The control module 405 is used to control the medical dynamic X-ray collimator according to the covered area, so as to block the medical rays emitted by the medical X-ray device that irradiate the covered area.

[0097] Optionally, the recognition module 402 is specifically used to input each frame of X-ray image into a pre-trained region recognition model to determine and label the regions of interest contained in the frame of X-ray image.

[0098] Optionally, the device further includes:

[0099] The first model training module 406 is used to acquire a first sample X-ray image; mark the region of interest (ROI) involving the target object in the first sample X-ray image, and input the marked first sample X-ray image into a region recognition model to be trained, so as to obtain the ROI contained in the first sample X-ray image recognized by the region recognition model; and train the region recognition model with the optimization objective of minimizing the deviation between the ROI contained in the first sample X-ray image recognized by the region recognition model and the marked ROI in the first sample X-ray image.

[0100] Optionally, the first model training module 406 is specifically configured to: if the number of regions of interest (ROIs) identified in the first sample X-ray image is less than a preset number of ROIs, then determine at least one region in the first sample X-ray image that does not involve the target object as an interference region; and train the region recognition model with the optimization objective of minimizing the deviation between the ROIs identified by the region recognition model in the first sample X-ray image and the ROIs marked in the first sample X-ray image, and maximizing the deviation between the ROIs identified by the region recognition model in the first sample X-ray image and the interference region.

[0101] Optionally, the target object includes surgical instruments and human organs.

[0102] Optionally, the device further includes:

[0103] The second model training module 407 is used to acquire a second sample X-ray image and a third sample X-ray image, wherein the second sample X-ray image and the third sample X-ray image are images of the same region; input the second sample X-ray image and the third sample X-ray image into the change analysis model to obtain the image change amount between the second sample X-ray image and the third sample X-ray image, which is used as the image change amount to be optimized; transform the second sample X-ray image according to the image change amount to be optimized to obtain a transformed X-ray image; and train the change analysis model with the optimization objective of minimizing the deviation between the transformed X-ray image and the third sample X-ray image.

[0104] Optionally, the control module 405 is further configured to, if the detected image change for the entire X-ray image is greater than a set value, expand the irradiation range of the medical X-ray emitted by the medical X-ray device by controlling the medical dynamic X-ray collimator.

[0105] This specification also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 A control method for a medical dynamic X-ray collimator is provided.

[0106] This instruction manual also provides Figure 5 One of the corresponding Figure 1 A schematic diagram of the structure of an electronic device. (e.g.) Figure 5 At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for the business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to achieve the above-mentioned functions. Figure 1 The control method for the medical dynamic X-ray collimator is described above. Of course, besides software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0107] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0108] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0109] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0110] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0111] Those skilled in the art will understand that embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0113] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0114] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0115] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0116] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0117] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0118] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0119] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0121] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

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

Claims

1. A control method for a medical dynamic X-ray collimator, characterized in that, include: Acquire continuous multi-frame X-ray images from medical dynamic X-ray equipment; For each frame of X-ray image, identify and label the regions of interest from that frame of X-ray image; The labeled X-ray images are input into a pre-trained change analysis model to determine the amount of image change corresponding to each region of interest contained in each X-ray image frame, based on the images of that region of interest contained in different X-ray images. Based on the image change amount corresponding to each region of interest, regions of interest with image change amounts less than a set threshold are determined and used as masking regions. The image change amount is used to represent the amount of change in the difference between different frames of X-ray images. The medical dynamic X-ray collimator is controlled according to the covered area so as to block the medical rays emitted by the medical dynamic X-ray device that irradiate the covered area. The training of the change analysis model specifically includes: acquiring a second sample X-ray image and a third sample X-ray image, wherein the second sample X-ray image and the third sample X-ray image are images of the same region; inputting the second sample X-ray image and the third sample X-ray image into the change analysis model to obtain the image change amount between the second sample X-ray image and the third sample X-ray image, which is used as the image change amount to be optimized; transforming the second sample X-ray image according to the image change amount to be optimized to obtain a transformed X-ray image; and training the change analysis model with minimizing the deviation between the transformed X-ray image and the third sample X-ray image as the optimization objective.

2. The method as described in claim 1, characterized in that, For each frame of X-ray image, the regions of interest are identified and labeled, specifically including: For each frame of X-ray image, the frame of X-ray image is input into a pre-trained region recognition model to determine and label the regions of interest contained in the frame of X-ray image.

3. The method as described in claim 2, characterized in that, The training region recognition model specifically includes: Acquire the first sample X-ray image; The region of interest involving the target object in the first sample X-ray image is marked, and the marked first sample X-ray image is input into the region recognition model to be trained, so as to obtain the region of interest contained in the first sample X-ray image recognized by the region recognition model; The region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest identified by the region recognition model in the first sample X-ray image and the region of interest marked in the first sample X-ray image.

4. The method as described in claim 3, characterized in that, The region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest identified by the region recognition model in the first sample X-ray image and the region of interest marked in the first sample X-ray image. The training process specifically includes: If the number of regions of interest in the first sample X-ray image is less than the preset number of regions of interest, then at least one region in the first sample X-ray image that does not involve the target object is identified as an interference region. The region recognition model is trained with the optimization objective of minimizing the deviation between the region of interest (ROI) identified by the region recognition model in the first sample X-ray image and the ROI marked in the first sample X-ray image, and maximizing the deviation between the ROI identified by the region recognition model in the first sample X-ray image and the interference region.

5. The method as described in claim 3 or 4, characterized in that, The targets include surgical instruments and human organs.

6. The method as described in claim 1, characterized in that, The method further includes: If the detected change in the X-ray image exceeds a set threshold, the irradiation range of the medical rays emitted by the medical dynamic X-ray device is expanded by controlling the medical dynamic X-ray collimator.

7. A control device for a medical dynamic X-ray collimator, characterized in that, include: The acquisition module is used to acquire multiple consecutive X-ray images collected by the medical dynamic X-ray equipment; The recognition module is used to identify and label the regions of interest in each frame of X-ray image. The input module is used to input the labeled X-ray images of each frame into a pre-trained change analysis model, so as to determine the image change corresponding to each region of interest contained in each X-ray image frame based on the images of that region of interest contained in different X-ray images. The determination module is used to determine regions of interest whose image change is less than a set threshold based on the image change amount corresponding to each region of interest, and to use these regions as occlusion areas. The image change amount is used to represent the amount of change in the difference between different frames of X-ray images. The control module is used to control the medical dynamic X-ray collimator according to the covered area, so as to block the medical rays emitted by the medical dynamic X-ray device that irradiate the covered area. The second model training module is used to acquire a second sample X-ray image and a third sample X-ray image, which are images of the same region; input the second sample X-ray image and the third sample X-ray image into the change analysis model to obtain the image change amount between the second sample X-ray image and the third sample X-ray image, which is used as the image change amount to be optimized; transform the second sample X-ray image according to the image change amount to be optimized to obtain a transformed X-ray image; and train the change analysis model with the optimization objective of minimizing the deviation between the transformed X-ray image and the third sample X-ray image.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method described in any one of claims 1 to 6.

Images