Method, device and computer equipment for generating a simulated pneumoperitoneum image

Abstract

The application relates to a simulated pneumoperitoneum image generation method and device, computer equipment, a storage medium and a computer program product. The simulated pneumoperitoneum image generation method comprises the following steps: acquiring a peritoneal cavity image; determining a target region according to the peritoneal cavity image; performing simulated pneumoperitoneum fitting on the peritoneal cavity image according to the target region to determine a simulated pneumoperitoneum region; and determining a simulated pneumoperitoneum image according to the target region, the simulated pneumoperitoneum region and the peritoneal cavity image. The method can automatically divide the target region from the peritoneal cavity image, saves the time for manual labeling, enhances the automation degree in the simulated pneumoperitoneum image generation process, and adopts the simulated pneumoperitoneum fitting method to directly obtain the deformed peritoneal cavity state image after pneumoperitoneum on the image. Compared with the method for generating the peritoneal cavity state image by using a numerical simulation method, the image generation time is greatly saved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, computer device, storage medium, and computer program product for generating simulated pneumoperitoneum images. Background Technology

[0002] As the foundation of laparoscopic surgery, pneumoperitoneum has always been a focus of laparoscopic technology research. In liver laparoscopic surgery, medical gases such as carbon dioxide are first injected through the endoscope port to inflate the abdominal cavity and create the basic environment for laparoscopic surgery. Then, the laparoscope is inserted into the endoscope port to observe the abdominal cavity environment. Next, the surgical site is located through preoperative CT scans and external exploration. Combined with the surgeon's surgical experience, the location of the surgical port is determined. Finally, the surgeon uses a scalpel to open the operating port. In this way, the surgeon can observe the abdominal cavity environment and perform surgery while viewing the abdominal cavity image returned by the endoscope.

[0003] To facilitate preoperative exploration of the human body before laparoscopic surgery and to make adequate surgical preparations based on the exploration results, it is necessary to simulate the human abdominal cavity with pneumoperitoneum to create a simulated pneumoperitoneum thoracic cavity environment. However, the existing methods for generating images of simulated pneumoperitoneum require a lot of manpower for manual region division, which is not only time-consuming but also extremely inefficient. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, storage medium, and computer program product for generating simulated pneumoperitoneum images that can quickly generate simulated pneumoperitoneum images, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for generating simulated pneumoperitoneum images, including:

[0006] Obtain abdominal cavity images;

[0007] The target region is determined based on the abdominal cavity image;

[0008] Based on the target region, the abdominal cavity image is fitted with simulated pneumoperitoneum to determine the simulated pneumoperitoneum region;

[0009] Based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, a simulated pneumoperitoneum image is determined.

[0010] In one embodiment, determining the target region based on the abdominal cavity image includes:

[0011] At least one deep neural network is used to perform target recognition on the abdominal cavity image to determine the abdominal muscle region;

[0012] The abdominal cavity image is segmented to determine the abdominal wall region;

[0013] The target region is obtained by fusing the abdominal muscle region and the abdominal wall region.

[0014] In one embodiment, the step of using at least one deep neural network to perform target recognition on the abdominal cavity image to determine the abdominal muscle region includes:

[0015] The abdominal cavity image is input into at least one of the deep neural networks to perform target recognition and obtain the corresponding prediction region.

[0016] The abdominal muscle region is determined based on the predicted region and the preset weights corresponding to each deep neural network.

[0017] In one embodiment, determining the abdominal muscle region based on the predicted region and the preset weights corresponding to each deep neural network includes:

[0018] Based on the prediction region and the corresponding preset weight of each deep neural network, the probability value of the prediction region of each pixel in the abdominal cavity image is determined.

[0019] The abdominal muscle region is segmented from the abdominal cavity image based on the predicted region probability value of each pixel.

[0020] In one embodiment, the step of segmenting the abdominal cavity image to determine the abdominal wall region includes:

[0021] Threshold analysis is used to determine the outer contour of the abdominal wall region from the abdominal cavity image;

[0022] A dynamic contour algorithm is used to segment the inner contour of the abdominal wall region based on the outer contour.

[0023] The inner contour is optimized using a scattered contour algorithm;

[0024] The abdominal wall region is determined based on the outer contour and the optimized inner contour.

[0025] In one embodiment, before fusing the abdominal muscle region and the abdominal wall region to obtain the target region, the method further includes:

[0026] The abdominal muscle region was optimized.

[0027] The process of fusing the abdominal muscle region and the abdominal wall region to obtain the target region includes:

[0028] The optimized abdominal muscle region and the abdominal wall region are merged to obtain the target region.

[0029] In one embodiment, optimizing the abdominal muscle region includes:

[0030] Obtain the first sub-region in the abdominal cavity image;

[0031] Obtain the relative position of any pixel in the abdominal muscle region and any pixel in the first sub-region;

[0032] Remove pixels in the abdominal muscle region whose relative positions do not meet the preset conditions.

[0033] In one embodiment, prior to acquiring the first sub-region in the abdominal cavity image, the process includes:

[0034] The abdominal cavity image is identified to determine the characteristic areas;

[0035] The region where the feature is located is designated as the first sub-region.

[0036] In one embodiment, optimizing the abdominal muscle region includes:

[0037] Obtain pixels within a preset range in the abdominal cavity image;

[0038] Remove pixels in the abdominal muscle region that overlap with pixels within the preset range.

[0039] In one embodiment, the step of performing simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region includes:

[0040] The deformation range is determined based on the target region and the abdominal cavity image;

[0041] A deformation fitting algorithm is used to deform the inner contour of the target area within the deformation range;

[0042] Based on the correspondence between the pixels of the inner contour and the outer contour within the target area, the outer contour is determined based on the deformed inner contour, thus obtaining the simulated pneumoperitoneum region.

[0043] In one embodiment, determining the deformation range based on the target region and the abdominal cavity image includes:

[0044] Obtain the second sub-region in the abdominal cavity image;

[0045] The deformation range is determined based on the positions of pixels in the target region and the positions of pixels in the second sub-region.

[0046] In one embodiment, before determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, the method further includes:

[0047] Receive adjustment instructions;

[0048] Adjust the target area and / or the simulated pneumoperitoneum area according to the adjustment instructions;

[0049] The step of determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image includes:

[0050] The simulated pneumoperitoneum image is determined based on the adjusted target area and / or the adjusted simulated pneumoperitoneum area.

[0051] In one embodiment, determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image includes:

[0052] The simulated pneumoperitoneum region is used to replace the target region in the abdominal cavity image to obtain the simulated pneumoperitoneum image.

[0053] Secondly, this application also provides a device for generating simulated pneumoperitoneum images, comprising:

[0054] The acquisition module is used to acquire images of the abdominal cavity;

[0055] The first determining module is used to determine the target region based on the abdominal cavity image;

[0056] The second determining module is used to perform simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region.

[0057] The third determining module is used to determine the simulated pneumoperitoneum image based on the target area, the simulated pneumoperitoneum area, and the abdominal cavity image.

[0058] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the simulated pneumoperitoneum image generation method described in any of the above embodiments.

[0059] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the simulated pneumoperitoneum image generation method described in any of the above embodiments.

[0060] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the simulated pneumoperitoneum image generation method described in any of the above embodiments.

[0061] The aforementioned method, apparatus, computer equipment, storage medium, and computer program for generating simulated pneumoperitoneum images automatically delineate the target region from the abdominal cavity image, eliminating the time required for manual annotation and enhancing the automation level in the simulated pneumoperitoneum image generation process. Furthermore, by employing a simulated pneumoperitoneum fitting method, the deformed abdominal cavity state image after pneumoperitoneum is directly obtained from the image. Compared with the commonly used numerical simulation method for generating abdominal cavity state images, this significantly saves image generation time. Attached Figure Description

[0062] Figure 1 This is a diagram illustrating the application environment of a method for simulating pneumoperitoneum image generation in one embodiment.

[0063] Figure 2 This is a diagram illustrating the application environment of a method for simulating pneumoperitoneum image generation in one embodiment.

[0064] Figure 3 This is a flowchart illustrating a method for generating simulated pneumoperitoneum images in one embodiment;

[0065] Figure 4 This is a flowchart illustrating the target region determination step in a method for generating simulated pneumoperitoneum images in one embodiment.

[0066] Figure 5 This is a schematic diagram of the target region determination step in a simulated pneumoperitoneum image generation method in one embodiment;

[0067] Figure 6 This is a schematic diagram of the process of determining the abdominal muscle region in a simulated pneumoperitoneum image generation method in one embodiment;

[0068] Figure 7 This is a schematic diagram of the process of determining the abdominal wall region in a simulated pneumoperitoneum image generation method in one embodiment;

[0069] Figure 8 This is a flowchart illustrating the target region determination step in a method for generating simulated pneumoperitoneum images in one embodiment.

[0070] Figure 9 This is a flowchart illustrating the step of determining the simulated pneumoperitoneum region in a simulated pneumoperitoneum image generation method in one embodiment.

[0071] Figure 10 This is a flowchart illustrating the step of obtaining the inner contour of the target region after deformation in a simulated pneumoperitoneum image generation method in one embodiment.

[0072] Figure 11 This is a flowchart illustrating the step of obtaining the outer contour of the target region after deformation in a method for generating simulated pneumoperitoneum images in one embodiment.

[0073] Figure 12This is a flowchart illustrating a method for generating simulated pneumoperitoneum images in one embodiment;

[0074] Figure 13 This is a schematic diagram of a device for generating simulated pneumoperitoneum images in one embodiment;

[0075] Figure 14 This is a schematic diagram of the structure of the first determining module in a simulated pneumoperitoneum image generation device in one embodiment;

[0076] Figure 15 This is a schematic diagram of the structure of the first determining module in a simulated pneumoperitoneum image generation device in one embodiment;

[0077] Figure 16 This is a schematic diagram of the structure of the second determining module in a simulated pneumoperitoneum image generation device in one embodiment;

[0078] Figure 17 This is a schematic diagram of a device for generating simulated pneumoperitoneum images in one embodiment;

[0079] Figure 18 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0081] The simulated pneumoperitoneum image generation method provided in this application embodiment can be applied to, for example... Figure 1-2 In the application environment shown, terminal 102 communicates with server 104 and medical imaging device 106 via a network.

[0082] For example, the simulated pneumoperitoneum image generation method is applied to terminal 102. Terminal 102 first acquires an abdominal cavity image through medical imaging device 106; determines the target region based on the abdominal cavity image; and performs simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region; finally, based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, determines the simulated pneumoperitoneum image. Terminal 102 sends the simulated pneumoperitoneum image to server 104, and server 104 stores the simulated pneumoperitoneum image in a data storage system. Terminal 102 can be, but is not limited to, a desktop computer. Medical imaging device 106 includes, but is not limited to, various imaging devices, such as CT imaging devices (CT: Computed Tomography, which uses precisely collimated X-ray beams and highly sensitive detectors to perform a series of cross-sectional scans around a part of the human body, and can reconstruct precise three-dimensional images of tumors, etc.); magnetic resonance imaging devices (which are a type of tomographic imaging that uses magnetic resonance to obtain electromagnetic signals from the human body and reconstruct human body information images), etc.

[0083] For example, the simulated pneumoperitoneum image generation method is applied to server 104. Server 104 first acquires an abdominal cavity image from medical imaging device 106 via terminal 102; it determines a target region based on the abdominal cavity image and, based on the target region, performs simulated pneumoperitoneum fitting on the abdominal cavity image to determine the simulated pneumoperitoneum region; finally, based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, it determines the simulated pneumoperitoneum image, and then server 104 stores the simulated pneumoperitoneum image in a data storage system. It is understood that the data storage system can be an independent storage device, or the data storage system can be located on the server, or the data storage system can be located on another terminal.

[0084] In one embodiment, a method for generating simulated pneumoperitoneum images is provided. This embodiment uses the application of this simulated pneumoperitoneum image generation method to a terminal as an example for illustration. Figure 3 As shown, the method for generating simulated pneumoperitoneum images includes:

[0085] Step 202: Obtain abdominal cavity image.

[0086] Abdominal images refer to images used to show the actual abdominal cavity environment of the human body. Abdominal images can be X-ray computed tomography (CT) images or magnetic resonance imaging (MRI) images. Abdominal images are used to help medical staff view the actual condition of the human abdominal cavity.

[0087] As an example, in this embodiment, the terminal acquires CT images of the human abdominal cavity using a CT device, which are then used as abdominal cavity images.

[0088] Step 204: Determine the target area based on the abdominal cavity image.

[0089] Pneumoperitoneum refers to a treatment method in which medical gases such as carbon dioxide are injected through the endoscope port to inflate the abdominal cavity, creating the basic environment for laparoscopic surgery. The target area refers to the region of the abdominal cavity that changes accordingly when pneumoperitoneum occurs; the target area is typically composed of the abdominal muscles and abdominal wall. When the abdominal cavity is inflated, the abdominal muscles and abdominal wall expand with air.

[0090] In this embodiment, the terminal locates the area on the abdominal cavity image that can be deformed due to pneumoperitoneum and identifies this area as the target area.

[0091] Step 206: Based on the target region, perform simulated pneumoperitoneum fitting on the abdominal cavity image to determine the simulated pneumoperitoneum region.

[0092] Simulated pneumoperitoneum fitting refers to the action of simulating pneumoperitoneum in the human abdominal cavity. The simulated pneumoperitoneum region refers to the region in the human abdominal cavity that can change due to pneumoperitoneum, and the resulting region after the change in pneumoperitoneum.

[0093] In this embodiment, the terminal performs simulated pneumoperitoneum fitting on the target area of ​​the abdominal cavity image to obtain an image of the target area after pneumoperitoneum occurs, and uses this image as the simulated pneumoperitoneum area.

[0094] Step 208: Determine the simulated pneumoperitoneum image based on the target area, the simulated pneumoperitoneum area, and the abdominal cavity image.

[0095] Simulated pneumoperitoneum images refer to images in which areas that can change due to pneumoperitoneum are replaced with areas that have changed due to pneumoperitoneum.

[0096] Specifically, the terminal removes all pixels contained in the target region from the abdominal cavity image and replaces them with all pixels contained in the simulated pneumoperitoneum region to determine the simulated pneumoperitoneum image.

[0097] Simulated pneumoperitoneum images can be 2D images. For example, a simulated pneumoperitoneum image is a planar image obtained by fitting simulated pneumoperitoneum onto each abdominal cavity image.

[0098] The simulated pneumoperitoneum image can also be a 3D image. For example, after simulating the simulated pneumoperitoneum region in the abdominal cavity image in step 206, the target region in the abdominal cavity image is removed and replaced with the simulated pneumoperitoneum region, resulting in a simulated image corresponding to the current abdominal cavity image. Furthermore, the simulated images corresponding to all planar abdominal cavity images (i.e., each tomographic image) are stitched together, and the pixels of the virtual pneumoperitoneum region in each simulated image are processed using linear interpolation, connected component analysis, threshold analysis, etc., to obtain a 3D image of the human abdominal cavity after virtual pneumoperitoneum simulation. During the acquisition of abdominal cavity images, a timestamp can be added to each abdominal cavity image according to the acquisition order. During the stitching together of the simulated images corresponding to all planar abdominal cavity images, all abdominal cavity images are re-stitched according to the order of the timestamps.

[0099] It should be understood that, in this embodiment, removing the target region in the abdominal cavity image refers to adjusting the pixel values ​​of all pixels contained in the target region to a preset value, such as -1000.

[0100] In this embodiment, the terminal replaces the target region in the abdominal cavity image with a simulated pneumoperitoneum region to obtain a simulated pneumoperitoneum image.

[0101] In the aforementioned method for generating simulated pneumoperitoneum images, the terminal acquires an abdominal cavity image showing the actual human abdominal cavity environment through an image acquisition device. Within this image, a region that changes during pneumoperitoneum is identified as the target region. This target region is then fitted with simulated pneumoperitoneum data to obtain its shape after pneumoperitoneum, which is then used as the simulated pneumoperitoneum region. Finally, all pixels in the target region are removed from the abdominal cavity image and replaced with pixels from the simulated pneumoperitoneum region, resulting in the simulated pneumoperitoneum image. The terminal automatically delineates the target region from the abdominal cavity image, eliminating the time required for manual annotation and enhancing the automation of the simulated pneumoperitoneum image generation process. Furthermore, by employing simulated pneumoperitoneum fitting, the deformed abdominal cavity state image after pneumoperitoneum is directly obtained on a 2D planar image, significantly reducing image generation time compared to methods typically using numerical simulation to generate abdominal cavity state images. Moreover, a 3D simulated pneumoperitoneum image can be generated from the 2D simulated image after pneumoperitoneum deformation, achieving full automation of the process.

[0102] like Figure 4 As shown, in some optional embodiments, step 204, determining the target region based on the abdominal cavity image, includes: step 2042, using at least one deep neural network to perform target recognition on the abdominal cavity image to determine the abdominal muscle region; step 2044, performing image segmentation on the abdominal cavity image to determine the abdominal wall region; and step 2046, fusing the abdominal muscle region and the abdominal wall region to obtain the target region.

[0103] like Figure 5 The diagram shown illustrates the processing flow for determining the target region in this embodiment. The deep neural network can be any one of a feedforward neural network, a long short-term memory neural network, a generative adversarial network, a recurrent neural network, or a convolutional neural network. In this embodiment, at least one deep neural network can be of different types, or it can be multiple neural networks of the same type trained with different sample data. For example, at least one deep neural network can be a convolutional neural network, trained using abdominal cavity sample images of males, females, people from southern China, people from northern China, adults, and minors, respectively.

[0104] The terminal first simultaneously inputs abdominal cavity images into at least one deep neural network to obtain region prediction results for at least one abdominal cavity image. Based on the region prediction results of at least one abdominal cavity image, the abdominal muscle region is determined. This setup enables this embodiment to determine the abdominal muscle region through collective judgment, thereby improving the recognition accuracy of the abdominal muscle region.

[0105] Furthermore, image segmentation processing is performed on the abdominal cavity image to determine the abdominal wall region.

[0106] Furthermore, since the abdominal wall region and the abdominal muscle region are adjacent, as an example, connected component analysis or threshold analysis can be used to merge the two to obtain the target region.

[0107] In this embodiment, the terminal employs at least one deep neural network to perform region recognition on the same abdominal cavity image, obtaining multiple region prediction results. Simultaneously, referencing these multiple prediction results, the abdominal muscle region is determined, resulting in more accurate localization. Furthermore, image segmentation is used to segment the abdominal wall region from the abdominal cavity image. Finally, the abdominal muscle region and the abdominal wall region are fused to obtain the target region. This embodiment uses a collective decision-making approach, combining the prediction results of multiple deep neural networks to determine the location of the abdominal muscle region, thus achieving more accurate region localization.

[0108] In some optional embodiments, step 2042, using at least one deep neural network to perform target recognition on the abdominal cavity image to determine the abdominal muscle region, includes: inputting the abdominal cavity image into at least one deep neural network to perform target recognition and obtain the corresponding predicted region; and determining the abdominal muscle region based on the predicted region and the preset weights corresponding to each deep neural network.

[0109] like Figure 6 As shown, the terminal simultaneously inputs the abdominal cavity image into multiple deep neural networks to obtain the predicted region corresponding to each pixel in the abdominal cavity image. This predicted region can be the abdominal muscle region or a non-abdominal muscle region.

[0110] Furthermore, for each pixel in the abdominal cavity image, multiple predicted regions output by deep neural networks can be obtained. Then, based on the preset weights corresponding to each deep neural network, the final predicted region corresponding to each pixel is determined. All pixels whose predicted region is the abdominal muscle region are divided as the abdominal muscle region, thus completing the determination of the abdominal muscle region.

[0111] In some optional embodiments, the abdominal muscle region is determined based on the predicted region and the preset weights corresponding to each deep neural network, including: determining the predicted region probability value of each pixel in the abdominal cavity image based on the predicted region and the corresponding preset weights of each deep neural network; and segmenting the abdominal muscle region from the abdominal cavity image based on the predicted region probability value of each pixel.

[0112] Specifically, the terminal pre-assigns corresponding weights to at least one deep neural network. When the abdominal image is input into all deep neural networks in step 204 and multiple corresponding region prediction results are obtained, each region prediction result includes the predicted region and corresponding prediction probability of each pixel in the abdominal image. The prediction region probability value of each pixel includes the predicted region and corresponding prediction probability of the current pixel. Then, according to the weights of the deep neural networks corresponding to each region prediction result, the final prediction region probability value of each pixel is calculated using the following formula:

[0113]

[0114] Where p represents the probability value of the predicted region for the current pixel in the abdominal cavity image; n represents the total number of deep neural networks, p i w represents the predicted region probability value of the current pixel output by the i-th deep neural network. i This represents the weights of the i-th deep neural network.

[0115] Furthermore, it is determined whether the predicted region probability value of the abdominal muscle area for each pixel reaches a pre-set probability threshold. If it does, the pixel is considered to belong to the abdominal muscle area. For example, the probability threshold can be 50%, which means that if the predicted region probability value of the abdominal muscle area is greater than 50%, the pixel is considered to belong to the abdominal muscle area.

[0116] As an example, when there are three deep neural networks, the weight of the first neural network is 0.1, the weight of the second neural network is 0.5, and the weight of the third neural network is 0.4. For a pixel in an abdominal image, the region prediction result output by the first neural network indicates that the pixel belongs to the abdominal muscle region with a probability of 50%, the region prediction result output by the second neural network indicates that the pixel belongs to the abdominal muscle region with a probability of 5%, and the region prediction result output by the third neural network indicates that the pixel belongs to the abdominal muscle region with a probability of 70%. Using the above formula, the probability that the current pixel belongs to the abdominal muscle region is p = 0.1 * 50% + 0.5 * 5% + 0.4 * 70% = 35.5%.

[0117] As an example, an abdominal image can be input into all deep neural networks to obtain the predicted region probability value P for a pixel, P = {P1, P2, P3, P4}, P1 = 0.3, P2 = 0.4, P3 = 0.1, P4 = 0.2, where P1 represents the probability that the current pixel belongs to the abdominal muscle region, P2 represents the probability that the current pixel belongs to the liver region, P3 represents the probability that the current pixel belongs to the bone region, and P4 represents the probability that the current pixel belongs to other regions. If the terminal pre-sets a probability threshold of 50%, then P1 does not reach the probability threshold, and the current pixel is considered not to belong to the abdominal muscle region. If the terminal pre-sets a probability threshold of 20%, then P1 reaches the probability threshold, and the current pixel is considered to belong to the abdominal muscle region.

[0118] Specifically, the region prediction results of the deep neural network can divide the abdominal cavity image into the abdominal muscle region and other regions. When the probability of a pixel in the abdominal cavity image obtained after combining the region prediction results of all deep neural networks is less than a preset probability threshold, the pixel included in the region prediction result is considered to belong to other regions.

[0119] In this embodiment, the terminal fuses the results output by multiple deep neural networks and makes a collective judgment based on the prediction results of multiple regions to determine the prediction region corresponding to each pixel in the abdominal cavity image. This improves the region segmentation accuracy of the terminal and makes the division of the abdominal muscle region more accurate.

[0120] like Figure 7 As shown, in some optional embodiments, step 2044, performing image segmentation on the abdominal cavity image to determine the abdominal wall region, includes: using a threshold analysis method to determine the outer contour of the abdominal wall region from the abdominal cavity image; using a dynamic contour algorithm to segment the inner contour of the abdominal wall region based on the outer contour; using a scatter contour algorithm to optimize the inner contour; and determining the abdominal wall region based on the outer contour and the optimized inner contour.

[0121] Specifically, the terminal uses the difference in grayscale values ​​between the pixels of the outer contour of the abdominal wall region to be extracted in the abdominal cavity image and the pixels of the background, and sets a threshold to divide the pixels into several categories, thereby achieving the separation of the outer contour from the background.

[0122] Furthermore, starting from the continuous closed curve of the outer contour, the terminal defines an energy function based on the grayscale, gradient, and other information of the abdominal cavity image pixels, causing the outer contour to move along the direction of decreasing edge energy until the edge energy reaches its minimum, thereby obtaining the inner contour of the abdominal wall region.

[0123] Furthermore, based on the inner contour of the abdominal wall region, for all work points contained in the inner contour, a circle with a radius equal to the adjacent radius R is used to roll around the work points, generating an optimized inner contour in accordance with the principle of the scattered contour algorithm.

[0124] Furthermore, the terminal divides the abdominal cavity image into an outer contour and an optimized inner contour, and takes the set of all pixels between the outer contour and the optimized inner contour as the abdominal wall region.

[0125] In this embodiment, the terminal uses a threshold analysis method to separate the background and the entire human abdominal cavity in the abdominal cavity image, determines the outer contour of the human abdominal cavity, and further uses a dynamic contour algorithm to segment the inner contour curve based on the curve of the outer contour. The inner contour curve is then optimized using a scattered contour algorithm. The set of pixels between the outer contour curve and the optimized inner contour curve is taken as the set of pixels in the abdominal wall region, thereby determining the abdominal wall region.

[0126] like Figure 8 As shown, in some optional embodiments, step 2045, optimizing the abdominal muscle region, is included before step 2046.

[0127] Step 2046 includes: fusing the optimized abdominal muscle region and abdominal wall region to obtain the target region.

[0128] In this embodiment, the terminal further optimizes the abdominal muscle region obtained in step 2042 to prevent significant errors in the pixels contained in the abdominal muscle region.

[0129] In some optional embodiments, step 2045 includes: obtaining a first sub-region in the abdominal cavity image; obtaining the relative position of any pixel in the abdominal muscle region and any pixel in the first sub-region; and removing pixels in the abdominal muscle region whose relative positions do not meet preset conditions.

[0130] The first sub-region can be the human skeletal region. For example, when the abdominal cavity image is a CT image, the abdominal cavity image will contain the imaging of the human spine and ribs. In step 2045, for any pixel in the abdominal muscle region, the relative position of the pixel and any pixel in the skeletal region is obtained.

[0131] The preset condition can be that the pixels in the abdominal muscle region are located in the preset direction of the pixels in the first sub-region, and the preset direction can be the direction away from the center point of the abdominal cavity image.

[0132] If the relative position indicates that a pixel in the abdominal muscle region is not located in the preset direction of a pixel in the first sub-region, then the pixel in the abdominal muscle region that does not meet the preset condition will be removed.

[0133] In this embodiment, the terminal removes pixels located inside the human skeletal area in the abdominal muscle region, in accordance with common sense, to optimize the abdominal muscle region.

[0134] In some optional embodiments, before obtaining the first sub-region in the abdominal cavity image, the method includes: identifying the abdominal cavity image to determine feature areas; and taking the region where the feature areas are located as the first sub-region.

[0135] Specifically, the terminal uses a deep neural network, for example, to identify the abdominal cavity image and determine the feature location. The deep neural network can be any of the following: feedforward neural network, long short-term memory neural network, generative adversarial neural network, recurrent neural network, or convolutional neural network. The terminal inputs the abdominal cavity image into the deep neural network to obtain the recognition result of the feature region, and uses the set of all pixels contained in the feature region as the first sub-region.

[0136] For example, the terminal selects all pixels with gray values ​​less than a preset gray value threshold as feature regions based on the gray values ​​of all pixels in the abdominal cavity image, and uses the set of all pixels with gray values ​​less than the preset gray value threshold as the first sub-region.

[0137] In this embodiment, the terminal segments a feature region from the abdominal cavity image as a first sub-region, which is used as a standard for optimizing the abdominal muscle region, thereby making the localization of the abdominal muscle region more accurate.

[0138] In some optional embodiments, step 2045 includes: acquiring pixels within a preset range in the abdominal cavity image; and removing pixels in the abdominal muscle region that overlap with pixels within the preset range.

[0139] Specifically, the preset range is, for example, 50% of the range where the pixels in the skeletal region of the abdominal cavity image are concentrated, used to represent the side and back of the human body, and to remove errors caused by the muscles in the back of the human body.

[0140] In this embodiment, the terminal sets the side where the spine is located in the abdominal cavity image as a preset range and removes the pixels in the abdominal muscle area that are within the preset range to prevent errors caused by pixels in the muscle areas on the side and back of the human body.

[0141] like Figure 9 As shown, in some optional embodiments, step 206 includes: step 2062, determining the deformation range based on the target region and the abdominal cavity image; step 2064, using a deformation fitting algorithm to deform the inner contour of the target region within the deformation range; step 2066, based on the correspondence between the pixels of the inner contour and the pixels of the outer contour in the target region, determining the deformed outer contour on the basis of the deformed inner contour to obtain the simulated pneumoperitoneum region.

[0142] Based on the abdominal cavity image and the positional relationship between the abdominal muscle region and the abdominal wall region, the terminal determines the range of the abdominal cavity that can deform with pneumoperitoneum. Then, deformation fitting is used to deform the inner contour of the target area within the deformation range. In the deformation fitting process, linear interpolation and spline curve difference methods can be used.

[0143] like Figure 10 As shown, during the deformation of the inner contour of the target area within the deformation range, a pre-set deformation mode can be used. This deformation mode can be applied to the inner contour of the target area. For example, a coordinate system can be established with the center point of the abdominal cavity image as the origin, the vertical direction of the abdominal cavity image as the y-axis, and the horizontal direction of the abdominal cavity image as the x-axis. The pre-set deformation mode in the terminal may include the correspondence between the coordinate position of any point on the inner contour curve of the target area before deformation and the coordinate position after deformation. This correspondence can be reflected by a mapping formula, and the specific content of the mapping formula can be changed according to the user settings.

[0144] like Figure 11 As shown, further, after the human abdominal cavity is deformed, the total area of ​​the target region should remain fixed, because after the inner contour of the target region is deformed, the outer contour of the target region should also change accordingly, so that the total number of pixels between the inner contour and the outer contour of the target region remains unchanged, that is, the area of ​​the target region in the abdominal cavity image remains unchanged.

[0145] For example, a coordinate system can be established with the center point of the abdominal cavity image as the origin, the vertical direction of the abdominal cavity image as the y-axis, and the horizontal direction of the abdominal cavity image as the x-axis. The coordinate positions of all pixels on the inner contour curve of the target region before deformation, the outer contour curve of the target region before deformation, and the inner contour curve of the target region after deformation can be obtained. First, based on the obtained coordinate positions of all points on the inner contour curve and the outer contour curve of the target region before deformation, the initial area A0 of the target region between the inner and outer contours before deformation is obtained. Then, for pixel i on the inner contour curve before deformation, the coordinate position of the pixel with the same horizontal coordinate on the inner contour curve after deformation is obtained as the coordinate position of pixel i after deformation. The lifting distance Δ of pixel i in the y-axis direction before and after deformation is further calculated. i Furthermore, obtain pixel j on the outer contour curve before deformation, which has the same horizontal coordinate as pixel i. Assume that pixel j is raised by Δ in the vertical direction. i The y-distance is used to obtain the raised outer contour curve. Based on the coordinate positions of all pixels on the deformed inner contour curve and the raised outer contour curve, the area A1 of the target region between the raised inner and outer contours can be calculated. The initial area A0 is divided by the area A1 to obtain a ratio r. Finally, pixel j is actually raised by rΔ in the vertical axis direction. i The y-distance is used to obtain the coordinates of all pixels on the outer contour curve of the final deformed target area. This method ensures that the area of ​​the target region remains unchanged between the inner and outer contours before and after deformation.

[0146] In this embodiment, the terminal first determines the deformation range that can occur from the abdominal cavity image, and then uses a deformation fitting algorithm to determine the inner contour of the target area after deformation within the deformation range. Subsequently, adhering to the principle that the total number of pixels in the target area remains unchanged, the outer contour of the target area after deformation is determined based on the inner contour. Through this setup, the curves of the inner and outer contours of the deformed target area can be obtained, thereby obtaining the simulated pneumoperitoneum region after simulation fitting.

[0147] In some optional embodiments, step 2062, determining the deformation range based on the target region and the abdominal cavity image, includes: obtaining a second sub-region in the abdominal cavity image; and determining the deformation range based on the positions of pixels in the target region and the positions of pixels in the second sub-region.

[0148] As an example, since the human skeleton cannot deform, the abdominal muscles connected to the skeleton cannot deform either. Therefore, the area in the human abdominal cavity that can change with pneumoperitoneum should be the abdominal part not surrounded by ribs. The second sub-region can be, for example, the human skeleton region. That is, the first sub-region and the second sub-region can be the same region. The identification and positioning process of the second sub-region refers to the acquisition process of the first sub-region, which will not be described in detail here.

[0149] Furthermore, the terminal determines the deformation range that is not surrounded by the human skeleton based on the current abdominal cavity image.

[0150] In this embodiment, the terminal uses the abdominal cavity area not surrounded by the human skeleton as the deformation range based on the position of the human skeleton.

[0151] like Figure 12 As shown, in some optional embodiments, before step 208, the method further includes: step 207, receiving an adjustment instruction; adjusting the target area and / or the simulated pneumoperitoneum area according to the adjustment instruction.

[0152] Step 208 includes: determining a simulated pneumoperitoneum image based on the adjusted target region and / or the adjusted simulated pneumoperitoneum region.

[0153] In this embodiment, medical personnel can view the target area segmented in step 204 and the simulated pneumoperitoneum area determined in step 206 through an interactive device, and re-segment and / or delineate the abdominal cavity image through interactive or semi-automatic drawing. Alternatively, they can modify the target area and / or simulated pneumoperitoneum area obtained by the terminal's automatic algorithm through interactive drawing. The terminal will then prioritize the modifications made by the medical personnel through the interactive device and use the modified target area and / or simulated pneumoperitoneum area as the standard for generating the simulated pneumoperitoneum image.

[0154] In some optional embodiments, step 208, determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, includes: replacing the target region in the abdominal cavity image with the simulated pneumoperitoneum region to obtain the simulated pneumoperitoneum image.

[0155] Specifically, the terminal removes all pixels from the target region in the abdominal cavity image and adds all pixels from the simulated pneumoperitoneum region to the abdominal cavity image after removing the target region, thus obtaining the simulated pneumoperitoneum image.

[0156] Furthermore, the terminal can also display the final simulated pneumoperitoneum image through an interactive device, for example, it can also highlight the simulated pneumoperitoneum area in the displayed simulated pneumoperitoneum image.

[0157] In the aforementioned method for generating simulated pneumoperitoneum images, multiple deep neural networks are used to identify abdominal cavity images, obtaining prediction results for multiple regions. These prediction results are then referenced simultaneously to determine the abdominal muscle region. A collective judgment approach is employed to improve region localization accuracy. Furthermore, pixels in the abdominal muscle region are optimized one-by-one based on the location of human bones or the position of pixels in the abdominal muscle region, eliminating errors caused by muscle pixels in other areas of the abdominal cavity image, thus improving the localization of the abdominal muscle region. Further still, the range within which the human abdominal cavity can deform with pneumoperitoneum is determined based on the location of human bones, and a user-preset deformation mode is used to target the region. The inner contour of the target area is deformed. Then, while ensuring the area of ​​the target area remains unchanged, the curve of the outer contour of the deformed target area is determined based on the curve of the inner contour, thus defining the simulated pneumoperitoneum region. Before determining the simulated pneumoperitoneum image, user-defined region adjustment commands via the interactive interface are prioritized to ensure that the simulated pneumoperitoneum image is generated according to the user's region division results. Finally, pixels contained in the final target area are removed from the abdominal cavity image and replaced with pixels from the simulated pneumoperitoneum region, resulting in a planar simulated pneumoperitoneum image. A 3D simulated pneumoperitoneum image can also be generated from multiple planar simulated pneumoperitoneum images. This method of generating simulated pneumoperitoneum images simplifies the process of generating virtual pneumoperitoneum, reduces human error, and improves reliability.

[0158] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0159] Based on the same inventive concept, this application also provides a simulated pneumoperitoneum image generation apparatus for implementing the simulated pneumoperitoneum image generation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the simulated pneumoperitoneum image generation apparatus provided below can be found in the limitations of the simulated pneumoperitoneum image generation method described above, and will not be repeated here.

[0160] In one embodiment, such as Figure 13As shown, a simulated pneumoperitoneum image generation device 1300 is provided, including: an acquisition module 1302, a first determination module 1304, a second determination module 1306, and a third determination module 1308, wherein: the acquisition module 1302 is used to acquire an abdominal cavity image; the first determination module 1304 is used to determine a target region based on the abdominal cavity image; the second determination module 1306 is used to perform simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region; and the third determination module 1308 is used to determine the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image.

[0161] like Figure 14 As shown, in some optional embodiments, the first determining module 1304 includes: a recognition unit 13042, used to perform target recognition on the abdominal cavity image using at least one deep neural network to determine the abdominal muscle region; a segmentation unit 13044, used to perform image segmentation on the abdominal cavity image to determine the abdominal wall region; and a fusion unit 13046, used to fuse the abdominal muscle region and the abdominal wall region to obtain the target region.

[0162] In some optional embodiments, the recognition unit 13042 is configured to: input an abdominal cavity image into at least one deep neural network to perform target recognition and obtain a corresponding prediction region; and determine the abdominal muscle region based on the prediction region and the preset weights corresponding to each deep neural network.

[0163] In some optional embodiments, the recognition unit 13042 is further configured to: determine the probability value of the predicted region of each pixel in the abdominal cavity image based on the predicted region and the corresponding preset weight of each deep neural network; and segment the abdominal muscle region from the abdominal cavity image based on the probability value of the predicted region of each pixel.

[0164] In some optional embodiments, the segmentation unit 13044 is configured to: use a threshold analysis method to determine the outer contour of the abdominal wall region from the abdominal cavity image; use a dynamic contour algorithm to segment the inner contour of the abdominal wall region based on the outer contour; use a scatter contour algorithm to optimize the inner contour; and determine the abdominal wall region based on the outer contour and the optimized inner contour.

[0165] like Figure 15 As shown, in some optional embodiments, the first determining module 1304 further includes an optimization unit 13045 for optimizing the abdominal muscle region; the third determining module 1308 is further configured to fuse the optimized abdominal muscle region and the abdominal wall region to obtain the target region.

[0166] In some optional embodiments, the optimization unit 13045 is configured to: acquire a first sub-region in the abdominal cavity image; acquire the relative position of any pixel in the abdominal muscle region and any pixel in the first sub-region; and remove pixels in the abdominal muscle region whose relative positions do not meet preset conditions.

[0167] In some optional embodiments, the optimization unit 13045 is further configured to: identify the abdominal cavity image and determine the feature region; and designate the region where the feature region is located as the first sub-region.

[0168] In some optional embodiments, the optimization unit 13045 is configured to: acquire pixels within a preset range in the abdominal cavity image; and remove pixels in the abdominal muscle region that overlap with pixels within the preset range.

[0169] like Figure 16 As shown, in some optional embodiments, the second determining module 1306 includes: a first determining unit 13062, used to determine the deformation range based on the target region and the abdominal cavity image; a deformation unit 13064, used to deform the inner contour of the target region within the deformation range using a deformation fitting algorithm; and a second determining unit 13066, used to determine the deformed outer contour based on the deformed inner contour according to the correspondence between the pixels of the inner contour and the pixels of the outer contour within the target region, thereby obtaining a simulated pneumoperitoneum region.

[0170] In some optional embodiments, the first determining unit 13062 is configured to: acquire a second sub-region in the abdominal cavity image; and determine the deformation range based on the position of the pixel in the target region and the position of the pixel in the second sub-region.

[0171] like Figure 17 As shown, in some optional embodiments, the simulated pneumoperitoneum image generation device 1300 further includes: a receiving module 1307, configured to receive an adjustment instruction and adjust the target area and / or the simulated pneumoperitoneum area according to the adjustment instruction; and a third determining module 1308 further configured to determine a simulated pneumoperitoneum image based on the adjusted target area and / or the adjusted simulated pneumoperitoneum area.

[0172] In some optional embodiments, the third determining module 1308 is further configured to: replace the target region in the abdominal cavity image with a simulated pneumoperitoneum region to obtain a simulated pneumoperitoneum image.

[0173] Each module in the aforementioned simulated pneumoperitoneum image generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0174] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 18 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for generating simulated pneumoperitoneum images. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0175] Those skilled in the art will understand that Figure 18 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0176] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the various steps of the above-described method for generating simulated pneumoperitoneum images.

[0177] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the various steps of the simulated pneumoperitoneum image generation method described above.

[0178] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0179] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0180] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for generating simulated pneumoperitoneum images, characterized in that, include: Obtain abdominal cavity images; The target region is determined based on the abdominal cavity image; Based on the target region, the abdominal cavity image is fitted with simulated pneumoperitoneum to determine the simulated pneumoperitoneum region; Based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, a simulated pneumoperitoneum image is determined; The target area includes the abdominal wall region; The step of determining the target region based on the abdominal cavity image includes: performing image segmentation on the abdominal cavity image to determine the abdominal wall region, including: using a threshold analysis method to determine the outer contour of the abdominal wall region from the abdominal cavity image; using a dynamic contour algorithm to segment the inner contour of the abdominal wall region based on the outer contour; using a scatter contour algorithm to optimize the inner contour; and determining the abdominal wall region based on the outer contour and the optimized inner contour. The step of performing simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region includes: The deformation range is determined based on the target region and the abdominal cavity image; A deformation fitting algorithm is used to deform the inner contour of the target area within the deformation range; Based on the correspondence between the pixels of the inner contour and the pixels of the outer contour within the target area, the outer contour after deformation is determined on the basis of the deformed inner contour, thus obtaining the simulated pneumoperitoneum region, wherein the total number of pixels between the inner contour and the outer contour of the target area remains unchanged before and after deformation.

2. The method according to claim 1, characterized in that, The step of determining the target region based on the abdominal cavity image further includes: At least one deep neural network is used to perform target recognition on the abdominal cavity image to determine the abdominal muscle region; The target region is obtained by fusing the abdominal muscle region and the abdominal wall region.

3. The method according to claim 2, characterized in that, The step of using at least one deep neural network to perform target recognition on the abdominal cavity image to determine the abdominal muscle region includes: The abdominal cavity image is input into at least one of the deep neural networks to perform target recognition and obtain the corresponding prediction region. The abdominal muscle region is determined based on the predicted region and the preset weights corresponding to each deep neural network.

4. The method according to claim 3, characterized in that, The step of determining the abdominal muscle region based on the predicted region and the preset weights corresponding to each deep neural network includes: Based on the prediction region and the corresponding preset weight of each deep neural network, the probability value of the prediction region of each pixel in the abdominal cavity image is determined. The abdominal muscle region is segmented from the abdominal cavity image based on the predicted region probability value of each pixel.

5. The method according to any one of claims 2-4, characterized in that, Before fusing the abdominal muscle region and the abdominal wall region to obtain the target region, the method further includes: The abdominal muscle region was optimized. The process of fusing the abdominal muscle region and the abdominal wall region to obtain the target region includes: The optimized abdominal muscle region and the abdominal wall region are merged to obtain the target region.

6. The method according to claim 5, characterized in that, The optimization of the abdominal muscle region includes: Obtain the first sub-region in the abdominal cavity image; Obtain the relative position of any pixel in the abdominal muscle region and any pixel in the first sub-region; Remove pixels in the abdominal muscle region whose relative positions do not meet the preset conditions.

7. The method according to claim 6, characterized in that, Before acquiring the first sub-region in the abdominal cavity image, the process includes: The abdominal cavity image is identified to determine the characteristic areas; The region where the feature is located is designated as the first sub-region.

8. The method according to claim 5, characterized in that, The optimization of the abdominal muscle region includes: Obtain pixels within a preset range in the abdominal cavity image; Remove pixels in the abdominal muscle region that overlap with pixels within the preset range.

9. The method according to claim 1, characterized in that, Determining the deformation range based on the target region and the abdominal cavity image includes: Obtain the second sub-region in the abdominal cavity image; The deformation range is determined based on the positions of pixels in the target region and the positions of pixels in the second sub-region.

10. The method according to claim 1, characterized in that, Before determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image, the method further includes: Receive adjustment instructions; Adjust the target area and / or the simulated pneumoperitoneum area according to the adjustment instructions; The step of determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image includes: The simulated pneumoperitoneum image is determined based on the adjusted target area and / or the adjusted simulated pneumoperitoneum area.

11. The method according to claim 1, characterized in that, The step of determining the simulated pneumoperitoneum image based on the target region, the simulated pneumoperitoneum region, and the abdominal cavity image includes: The simulated pneumoperitoneum region is used to replace the target region in the abdominal cavity image to obtain the simulated pneumoperitoneum image.

12. A device for generating simulated pneumoperitoneum images, characterized in that, include: The acquisition module is used to acquire images of the abdominal cavity; The first determining module is used to determine the target region based on the abdominal cavity image; The second determining module is used to perform simulated pneumoperitoneum fitting on the abdominal cavity image based on the target region to determine the simulated pneumoperitoneum region. The third determining module is used to determine the simulated pneumoperitoneum image based on the target area, the simulated pneumoperitoneum area, and the abdominal cavity image; The target region includes the abdominal wall region; the first determining module includes: a segmentation unit configured to: determine the outer contour of the abdominal wall region from the abdominal cavity image using a threshold analysis method; segment the inner contour of the abdominal wall region based on the outer contour using a dynamic contour algorithm; optimize the inner contour using a scatter contour algorithm; and determine the abdominal wall region based on the outer contour and the optimized inner contour. The second determining module includes: The first determining unit is used to determine the deformation range based on the target region and the abdominal cavity image; A deformation unit is used to deform the inner contour of the target area within the deformation range using a deformation fitting algorithm. The second determining unit is used to determine the deformed outer contour based on the deformed inner contour according to the correspondence between the pixels of the inner contour and the pixels of the outer contour within the target area, thereby obtaining the simulated pneumoperitoneum region, wherein the total number of pixels between the inner contour and the outer contour of the target area remains unchanged before and after deformation.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 11.

15. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 11.

Images