Method, system, device and storage medium for segmenting a stomach lesion image

Abstract

The application relates to a stomach lesion image segmentation method, system, device, computer equipment and storage medium. The method comprises the following steps: acquiring a stomach medical image; segmenting the stomach medical image to obtain an outer stomach wall surface and an inner stomach wall surface; obtaining a volume relationship between a stomach cavity volume and a stomach volume according to the outer stomach wall surface and the inner stomach wall surface; judging a stomach filling condition according to the volume relationship, and segmenting a target lesion according to the filling condition. The method can improve the detection performance of the stomach lesion and improve the accuracy of the stomach lesion segmentation.

Description

Technical Field

[0001] This application relates to the field of image segmentation technology, and in particular to a method, system, apparatus, computer device, and storage medium for segmenting images of gastric lesions. Background Technology

[0002] The stomach is a hollow organ, and lesions there are reflected on the gastric lining. On CT (computed tomography), these lesions typically appear as thickening with enhancement, which differs significantly from the appearance of solid tumors such as those in the liver. There is no clear boundary between gastric lesions and normal tissue on CT images; therefore, methods used for segmenting solid lesions, such as those used for liver lesions, are not suitable for segmenting gastric lesions.

[0003] Traditional techniques involve segmenting the stomach wall to identify whether it has thickened and thus determine if a tumor exists. However, this method is not suitable for segmenting lesions in all stomach conditions, and there is an issue of inaccurate segmentation of stomach lesions. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, system, device, computer equipment, and storage medium for accurately segmenting gastric lesion images to address the aforementioned technical problems.

[0005] One embodiment of this specification provides a method for segmenting an image of a gastric lesion. The method includes: acquiring a medical image of the stomach; segmenting the medical image of the stomach to obtain the outer wall surface and the inner wall surface of the stomach; obtaining the volume relationship between the gastric cavity volume and the stomach volume based on the outer wall surface and the inner wall surface of the stomach; determining the filling status of the stomach based on the volume relationship; and segmenting the target lesion based on the filling status.

[0006] In one embodiment, the method for obtaining the volume relationship between the gastric cavity volume and the gastric volume based on the outer surface of the stomach wall and the inner surface of the stomach includes: calculating the gastric cavity volume based on the inner surface of the stomach wall; calculating the gastric volume based on the outer surface of the stomach wall; and obtaining the volume relationship between the gastric cavity volume and the gastric volume based on the gastric cavity volume and the gastric body.

[0007] In one embodiment, the volume relationship is the volume ratio of the gastric cavity volume to the stomach volume; the step of determining the stomach filling status based on the volume relationship and segmenting the target lesion based on the filling status includes: the volume relationship is the volume ratio of the gastric cavity volume to the stomach volume; when the volume ratio is greater than or equal to a first threshold, the stomach is considered to be full; when the volume ratio is less than the first threshold, the stomach is considered to be underfilled.

[0008] In one embodiment, the method for segmenting the target lesion based on the filling status includes: when the stomach is full, calculating the gastric arm thickness based on the gastric outer wall surface and the gastric inner wall surface; segmenting the region where the gastric wall thickness is greater than or equal to a second threshold to obtain at least one first gastric outer wall surface; and obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

[0009] In one embodiment, the method includes: the first gastric outer wall surface includes a plurality of outer wall points, and the gastric inner wall surface includes a plurality of inner wall points; for each outer wall point, a corresponding target point is determined from the plurality of inner wall points; wherein the distance between the corresponding outer wall point and the target point is the minimum distance among the distances between the outer wall point and each of the inner wall points; and the target lesion is obtained by determining the first gastric inner wall surface corresponding to the first gastric outer wall surface based on the target point corresponding to each of the outer wall points.

[0010] In one embodiment, the method for obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface includes: in response to a user selecting a seed point on the gastric outer wall surface, forming a growth region around the seed point according to a preset growth rule; and generating a second gastric outer wall surface based on the growth region and the first gastric outer wall surface. The second gastric outer wall surface is used as a new first gastric outer wall surface for obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

[0011] In one embodiment, the method for obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface includes: obtaining candidate lesions based on the first gastric outer wall surface and the gastric inner wall surface; determining the influence factors of the candidate lesions, the influence factors including one or more of the following: thickening weight, lesion enhancement weight, lymph node weight, and vascular tortuosity weight; calculating the confidence level of the candidate lesions based on the influence factors; when the confidence level is greater than or equal to a third threshold, retaining the candidate lesion as the target lesion; when the confidence level is less than the third threshold, removing the candidate lesion.

[0012] In one embodiment, the method includes: the thickening weight being positively correlated with at least one of the maximum thickness value of the candidate lesion and the area of ​​the candidate lesion; the lesion enhancement weight being correlated with at least one of the average gray value of the candidate lesion voxel and the uniformity of gray value distribution of the candidate lesion voxel, wherein the lesion enhancement weight is positively correlated with the average gray value of the candidate lesion voxel and negatively correlated with the uniformity of gray value distribution of the candidate lesion voxel; the lymph node weight being positively correlated with at least one of the specificity of lymph nodes surrounding the candidate lesion and the number of specific lymph nodes; and the vascular tortuosity weight being correlated with whether there is a tortuosity shadow on the nearest blood vessel to the candidate lesion, wherein if the blood vessel has a tortuosity shadow, the vascular tortuosity weight is retained; and if the blood vessel does not have a tortuosity shadow, the vascular tortuosity weight is removed.

[0013] In one embodiment, segmenting the target lesion based on the filling status may further include: when the stomach is underfilled, segmenting the target lesion in response to a user's operation on the gastric wall surface and / or within the gastric cavity.

[0014] In one embodiment, the method further includes: displaying the first gastric outer wall surface in the interface according to the second threshold; adjusting the second threshold in response to a user's adjustment operation; and updating the first gastric outer wall surface in the interface according to the adjusted second threshold.

[0015] In one embodiment, the method further includes: in response to a user's editing operation, performing layer-by-layer editing on the three-dimensional medical image of the target lesion to obtain an edited three-dimensional medical image of the target lesion.

[0016] One embodiment of this specification provides a segmentation system for gastric lesion images. The system includes a processor, which performs the following steps: acquiring a gastric medical image; segmenting the gastric medical image to obtain the outer wall surface and the inner wall surface of the stomach; obtaining the volume relationship between the gastric cavity volume and the stomach volume based on the outer wall surface and the inner wall surface of the stomach; determining the filling status of the stomach based on the volume relationship; and segmenting the target lesion based on the filling status.

[0017] One embodiment of this specification provides a segmentation device for gastric lesion images. The device includes: an acquisition module for acquiring gastric medical images; a first segmentation module for segmenting the gastric medical images to obtain the outer wall surface and the inner wall surface of the stomach; a volume relationship module for obtaining the volume relationship between the gastric cavity volume and the stomach volume based on the outer wall surface and the inner wall surface of the stomach; and a second segmentation module for determining the filling status of the stomach based on the volume relationship and segmenting the target lesion based on the filling status.

[0018] One embodiment of this specification provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: acquiring a medical image of the stomach; segmenting the medical image of the stomach to obtain the outer wall surface and the inner wall surface of the stomach; obtaining the volume relationship between the gastric cavity volume and the stomach volume based on the outer wall surface and the inner wall surface of the stomach; determining the filling status of the stomach based on the volume relationship; and segmenting the target lesion based on the filling status.

[0019] One embodiment of this specification provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in any of the above method embodiments.

[0020] One embodiment of this specification provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it performs the following steps: acquiring a medical image of the stomach; segmenting the medical image of the stomach to obtain the outer wall surface and the inner wall surface of the stomach; obtaining the volume relationship between the gastric cavity volume and the stomach volume based on the outer wall surface and the inner wall surface of the stomach; determining the filling status of the stomach based on the volume relationship; and segmenting the target lesion based on the filling status.

[0021] One embodiment of this specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in any of the above method embodiments.

[0022] Compared with the prior art, the present invention provides a method, system, device, computer equipment, and storage medium for segmenting images of gastric lesions, which has the following beneficial effects:

[0023] 1. The stomach is a hollow organ, and its lesion information is reflected on the gastric arm. The lesion is attached to the inner wall and infiltrates and grows towards the outer wall. This invention fully considers the growth characteristics of gastric lesions, separates the outer wall surface and the inner wall surface of the stomach, avoids direct detection of lesions on the whole stomach, and improves the detection performance of gastric lesions.

[0024] 2. This invention utilizes the boundary between the outer and inner surfaces of the stomach wall to delineate the area of ​​the stomach cavity and the entire stomach, calculates the volume relationship between the stomach cavity volume and the stomach volume, determines the stomach's fullness based on this relationship, and segments the target lesion from the stomach according to the corresponding stomach fullness. Adopting different lesion segmentation strategies based on different stomach conditions will greatly improve the accuracy of stomach lesion segmentation. Attached Figure Description

[0025] Figure 1 This is an application environment diagram of a gastric lesion image segmentation method in one embodiment;

[0026] Figure 2 This is a flowchart illustrating a method for segmenting images of gastric lesions in one embodiment;

[0027] Figure 3 This is a flowchart illustrating the steps for obtaining the target lesion in one embodiment;

[0028] Figure 4 This is a schematic diagram visualizing the stomach wall thickness in one embodiment;

[0029] Figure 5 This is a schematic diagram of the visualization of the first gastric outer wall surface segmented according to a second threshold in one embodiment;

[0030] Figure 6 This is a flowchart illustrating the steps for generating the second gastric outer wall surface in one embodiment;

[0031] Figure 7 A flowchart illustrating the steps for determining the impact factors of candidate lesions and retaining target lesions in one embodiment;

[0032] Figure 8 This is a block diagram of the internal structure of a segmentation system for a gastric lesion image in one embodiment;

[0033] Figure 9 This is a structural block diagram of a device for segmenting images of gastric lesions in one embodiment;

[0034] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0035] 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.

[0036] Figure 1 This is a schematic diagram of a gastric lesion image segmentation system 100 according to an embodiment of this application. The gastric lesion image segmentation system 100 may include a data acquisition device 110, a processing device 120, a storage device 140, and an interaction device 150. The data acquisition device 110, processing device 120, storage device 140, and interaction device 150 can communicate with each other via a network 130.

[0037] The data acquisition device 110 can be a device that collects data. The acquired data may include image data, object feature data, etc. For example, the acquired image data may be a medical image of the stomach. In one embodiment, the data acquisition device 110 may include an imaging device. The imaging device can acquire image data. The imaging device may be one or more combinations of magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), b-scan ultrasonography, diasonography, thermal texture maps (TTM), and medical electronic endoscope (MEE). The image data may be images or data including blood vessels, tissues, or organs of an object. In one embodiment, an object feature acquisition device may be integrated into the imaging device to simultaneously acquire image data and object feature data. In one embodiment, the data acquisition device 110 may transmit the acquired data to a processing device 120, a storage device 140, and / or an interactive device 150 via a network 130.

[0038] The processing device 120 can process data. This data may be data acquired by the data acquisition device 110, data read from the storage device 140, feedback data obtained from the interaction device 150 (such as user input data), or data obtained from the cloud or external devices via the network 130. In one embodiment, the data may include image data, object feature data, user input data, etc. Data processing may include selecting a region of interest (ROI) in the image data. The ROI may be selected automatically by the processing device 120 or based on user input data. In one embodiment, the selected ROI may be a lesion, blood vessel, tissue, or organ. For example, the ROI may be the outer surface of the stomach, the inner surface of the stomach, or a target lesion. The processing device 120 may further segment the ROI in the image. Image segmentation methods may include edge-based image segmentation methods, such as the Perwitt operator, Sobel operator, gradient operator, Kirch operator, etc.; region-based image segmentation methods, such as region growing, thresholding, clustering, etc.; and other segmentation methods, such as those based on fuzzy sets or neural networks.

[0039] In one embodiment, the processing device 120 can calculate the volumetric relationship between tissues. For example, it can divide the gastric cavity and the entire stomach into regions based on the boundary between the outer and inner surfaces of the stomach wall, and calculate the volumetric relationship between the gastric cavity volume and the stomach volume. In one embodiment, the volumetric relationship between the gastric cavity volume and the stomach volume can be a proportional relationship between the gastric cavity volume and the stomach volume, or a relative difference between the gastric cavity volume and the stomach volume. In one embodiment, medical image processing software is used to perform three-dimensional reconstruction of the segmented contours of the outer and inner surfaces of the stomach to generate a three-dimensional model, and the gastric cavity volume and the stomach volume are calculated. In one embodiment, the volume calculation method includes voxel method or mesh method, etc.

[0040] In one embodiment, the processing device 120 can determine the fullness of the stomach based on the volume relationship between the gastric cavity volume and the stomach volume. In one embodiment, the fullness of the stomach is determined by the volume ratio or relative difference between the gastric cavity volume and the stomach volume. In one embodiment, the stomach is considered full if the volume ratio between the gastric cavity volume and the stomach volume is greater than or equal to a first threshold, and underfulfilled if the volume ratio is less than the first threshold. In another embodiment, the stomach is considered full if the relative difference between the gastric cavity volume and the stomach volume is less than or equal to a fourth threshold, and underfulfilled if the relative difference is less than the fourth threshold.

[0041] In one embodiment, the processing device 120 can segment the target lesion based on the gastric filling status. In one embodiment, when the stomach is full, the gastric arm thickness is calculated based on the outer and inner gastric wall surfaces; regions with a gastric wall thickness greater than or equal to a second threshold are segmented to obtain at least one first outer gastric wall surface, and the target lesion is obtained based on the first outer gastric wall surface and the inner gastric wall surface. In one embodiment, there are multiple first outer gastric wall surfaces, corresponding to multiple target lesions. In one embodiment, when the stomach is underfilled, the target lesion is segmented in response to user operations on the inner gastric wall surface and / or within the gastric cavity.

[0042] In one embodiment, the processing device 120 may perform noise reduction or smoothing on the acquired data or processing results. In another embodiment, the processing device 120 may send the acquired data or processing results to the storage device 140 for storage or to the interactive device 150 for display. The processing result may be an intermediate result generated during the processing, such as the surface of the outer and inner walls of the stomach, or it may be the final result of the processing, such as the final target lesion. In one embodiment, the processing device 120 may be one or more processing elements or devices, such as a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), system on a chip (SoC), microcontroller unit (MCU), etc. In another embodiment, the processing device 120 may also be a specially designed processing element or device with special functions. The processing device 120 may be local or remote relative to the data acquisition device 110.

[0043] Storage device 140 can store data or information. The data or information may include data acquired by data acquisition device 110, processing results or control commands generated by processing device 120, and user input data received by interactive device 140. Storage device 140 can be one or more read- or write-able storage media, including static random access memory (SRAM), random-access memory (RAM), read-only memory (ROM), hard disk, flash memory, etc. In one embodiment, storage device 140 can also be a remote storage device, such as a cloud drive.

[0044] The interactive device 150 can receive, send, and / or display data or information. The received data or information may include data acquired by the data acquisition device 110, processing results generated by the processing device 120, and data stored by the storage device 140. For example, the data or information displayed by the interactive device 150 may include gastric medical images acquired by the data acquisition device 110, gastric outer and inner wall surfaces segmented by the processing device 120, the volume relationship between the gastric cavity volume and the gastric volume calculated by the processing device 120, and target lesions segmented by the processing device 120. The display format may include one or more combinations of two-dimensional color medical images, three-dimensional color medical images, color geometric models and their mesh analysis, vector graphics (such as velocity vector lines), contour maps, filled contour maps (cloud maps), XY scatter plots, particle trajectory maps, and simulated flow effects. As another example, the data or information sent by the interactive device 150 may include user input information. The interactive device 150 can receive one or more operating parameters input by the user and send them to the processing device 120. In one embodiment, the interactive device 150 may include a user interface. Users can input user data into interactive device 150 through specific interactive devices such as mouse, keyboard, touchpad, microphone, etc.

[0045] In one embodiment, the interactive device 150 may be a display screen or other device with display functionality. In one embodiment, the interactive device 150 may have some or all of the functions of the processing device 120. For example, the interactive device 150 may perform operations such as smoothing, noise reduction, and color changing on the results generated by the processing device 120. For example, a color changing operation can convert a grayscale image into a color image, or a color image into a grayscale image. In one embodiment, the interactive device 150 and the processing device 120 may be an integrated device. The integrated device can simultaneously implement the functions of the processing device 120 and the interactive device 150. In one embodiment, the interactive device 150 may include a desktop computer 150-2, a server, a mobile device, etc. Mobile devices may include mobile phones 150-1, laptops, tablets 150-3, built-in devices in vehicles (e.g., motor vehicles, ships, airplanes, etc.), wearable devices, etc. In one embodiment, the interactive device 150 may include or be connected to a display device, printer, fax machine, etc.

[0046] Network 130 can be used for communication within the gastric lesion image segmentation system 100, receiving information from outside the system, and sending information to outside the system. In one embodiment, the data acquisition device 110, processing device 120, and interaction device 150 can access network 130 via wired connection, wireless connection, or a combination thereof. Network 130 can be a single network or a combination of multiple networks. In one embodiment, network 130 can be one or more of, but not limited to, local area networks (LANs), wide area networks (WANs), public networks, private networks, wireless LANs, virtual networks, metropolitan area networks (MANs), and public switch telephone networks. In one embodiment, network 130 can include multiple network access points, such as wired or wireless access points, base stations, or network switching points, through which data sources connect to network 130 and send information via the network.

[0047] In one embodiment, such as Figure 2 The flowchart 200 provided a method for segmenting images of gastric lesions. This embodiment applies this method to... Figure 1 The following example is provided for illustration within the system. The steps in this embodiment are executed by the processing device 120. Process 200 includes the following steps.

[0048] Step 201: Obtain medical images of the stomach.

[0049] In one embodiment, gastric medical imaging may include XR images, CT images, MR images, ultrasound images, or any combination thereof. In one embodiment, gastric medical imaging may be a three-dimensional image, a two-dimensional image, or a three-dimensional image composed of multiple two-dimensional layers.

[0050] Step 202: Segment the gastric medical image to obtain the outer and inner surfaces of the stomach.

[0051] In one embodiment, a medical image of the stomach is segmented according to a medical image segmentation method to obtain the outer and inner surfaces of the stomach wall. The medical image segmentation method may include edge-based image segmentation methods, such as the Perwitt operator, Sobel operator, gradient operator, and Kirch operator; region-based image segmentation methods, such as region growing, thresholding, and clustering; and other segmentation methods, such as those based on fuzzy sets and neural networks.

[0052] Step 203: Obtain the volume relationship between the gastric cavity volume and the gastric volume based on the outer and inner surfaces of the gastric wall.

[0053] The volume relationship can be either the ratio of the gastric cavity volume to the gastric volume, or the relative difference between the two volumes. The gastric cavity volume refers to the hollow space within the stomach, the area where food, liquids, and gases are temporarily stored. The gastric volume refers to the total volume of the entire stomach organ, including the stomach wall (mucosa, muscle layer, etc.) and the gastric cavity. In one embodiment, the fullness of the stomach is determined based on the ratio or relative difference between the gastric cavity volume and the gastric volume.

[0054] Step 204: Determine the fullness of the stomach based on the volume relationship, and segment the target lesion according to the fullness.

[0055] The target lesion refers to proliferative lesions, epithelial tumors, mesenchymal tumors, lymphomas, metastatic tumors, etc., on the gastric arm. In one embodiment, the volume ratio or relative difference between the gastric lumen volume and the gastric volume is used to determine the gastric fullness. Different processing strategies are selected to segment the target lesion based on the gastric fullness. In one embodiment, when the stomach is full, the target lesion is automatically segmented by the processing device. In another embodiment, when the stomach is full, the outer and inner surfaces of the stomach are segmented by the processing device, and the outer surface of the stomach is edited by the user on the interactive device, resulting in semi-automatic segmentation of the target lesion. In a third embodiment, when the stomach is not full, the user actively outlines the target lesion on the interactive device.

[0056] The above method fully considers the growth characteristics of gastric lesions, segmenting the outer and inner surfaces of the stomach to avoid direct detection of lesions on the entire stomach, thus improving the detection performance of gastric lesions. The boundary between the outer and inner surfaces of the stomach is used to delineate the area of ​​the gastric cavity and the entire stomach. The volume relationship between the gastric cavity volume and the stomach volume is calculated, and the stomach's fullness is determined based on this relationship. The target lesion is then segmented from the stomach according to the corresponding fullness condition. Adopting different gastric lesion segmentation strategies based on different stomach conditions will greatly improve the accuracy of gastric lesion segmentation.

[0057] In addition, the target lesion is only for reference for doctors to make judgments, and the specific disease diagnosis of the target lesion is determined by the doctor's experience.

[0058] Step 203, obtaining the volume relationship between the gastric cavity volume and the gastric volume based on the outer and inner surfaces of the gastric wall, includes the following steps. In one embodiment, step 203 can be performed by the processing device 120.

[0059] Step 2031: Calculate the gastric cavity volume based on the surface of the gastric inner wall.

[0060] Step 2032: Calculate the stomach volume based on the surface of the stomach wall.

[0061] Step 2033: Obtain the volume relationship between the gastric cavity volume and the gastric body volume based on the gastric cavity volume and the gastric body volume.

[0062] In one embodiment, the processing device 120 calculates the gastric cavity volume and gastric volume using methods such as voxel analysis or mesh analysis. The gastric cavity volume and gastric volume can be calculated intuitively and effectively using the outer and inner surfaces of the stomach wall, thereby obtaining the volume relationship between the gastric cavity volume and gastric volume, providing a theoretical basis for subsequently determining the gastric fullness based on this volume relationship.

[0063] Step 204: Determine the fullness of the stomach based on the volume relationship, and segment the target lesion according to the fullness relationship; the volume relationship is the volume ratio of the gastric cavity volume to the gastric volume. In one embodiment, step 204 may include the following steps performed by the processing device 120.

[0064] Step 2041: When the volume ratio is greater than or equal to the first threshold, the stomach is considered to be full.

[0065] Step 2042: When the volume ratio is less than the first threshold, the stomach is considered underfilled.

[0066] In one embodiment, the first threshold can be any value in [0.8, 0.9], and the first threshold can also be adjusted according to user needs. In one embodiment, the processing device 120 determines that the stomach is full when the relative difference between the gastric cavity volume and the stomach volume is less than or equal to a fourth threshold; and that the stomach is underfilled when the relative difference is less than the fourth threshold. The fourth threshold can be any value in [0.1, 0.2], and the fourth threshold can also be adjusted according to user needs.

[0067] In one embodiment, the first or fourth threshold is related to at least one of the patient's diet, weight, age, sex, and height. Gastric volume and gastric lumen volume are closely related to the patient's individual physical indicators and vary significantly with food intake and emptying. The physician can adjust the first or fourth threshold based on experience and the patient's specific situation.

[0068] like Figure 3 The flowchart 300 shown is a schematic diagram of the steps for obtaining the target lesion in one embodiment. Step 204 determines the fullness of the stomach based on volume relationship and segments the target lesion according to the fullness, including:

[0069] Step 301: When the stomach is full, calculate the thickness of the gastric arm based on the outer and inner surfaces of the stomach.

[0070] In one embodiment, the gastric wall thickness is characterized by the distance between each point on the outer surface of the stomach and the nearest point on the inner surface of the stomach. In one embodiment, the processing device 120 searches for the nearest point on the inner surface of the stomach to each point on the outer surface of the stomach, calculates the Euclidean distance between these two points as the thickness value of that point on the outer surface of the stomach, and iterates through the outer surface of the stomach to obtain the stomach wall thickness corresponding to each point on the outer surface of the stomach. In one embodiment, as... Figure 4 This diagram visualizes gastric wall thickness, with different thickness ranges represented by different shades of gray or colors. The thickness range in the diagram is [0.01mm, 10mm]. The method described above fully considers the growth characteristics of gastric lesions and accurately calculates the gastric wall thickness for subsequent lesion segmentation. The diagram also visually displays the gastric wall thickness range, allowing users to easily identify the location of gastric wall thickening for lesion segmentation.

[0071] Step 302: Segment the region where the gastric wall thickness is greater than or equal to the second threshold to obtain at least one first gastric outer wall surface, and obtain the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

[0072] In one embodiment, when the stomach wall thickness is greater than or equal to a second threshold, it can be considered that there may be a lesion in the stomach wall at this location. Therefore, the area where the stomach wall thickness is greater than or equal to the second threshold is segmented to obtain at least one first external stomach wall surface. In one embodiment, there may be one or more lesions on the stomach wall, and the corresponding first external stomach wall surface may be one or more. A lesion refers to abnormal tissue on the stomach wall, so it is necessary to combine the first external stomach wall surface and the internal stomach wall surface to obtain the target lesion. In one embodiment, the second threshold is any value in [9.0 mm, 9.8 mm]. In one embodiment, the processing device 120 segments the external stomach wall surface corresponding to the area where the stomach wall thickness is greater than or equal to the second threshold according to a preset second threshold to form one or more first external stomach wall surfaces. In one embodiment, the user actively adjusts the second threshold according to experience or surgical requirements, and the processing device segments the external stomach wall surface corresponding to the area where the stomach wall thickness is greater than or equal to the second threshold according to the adjusted second threshold to form one or more first external stomach wall surfaces. Figure 4 If the second threshold is selected as 0.677 mm, the first segmented gastric lateral wall surface is almost equal to the gastric lateral wall surface. Therefore, a second threshold that is too low cannot be used to segment the target lesion. Figure 5 This is a schematic diagram of the visualization of the first gastric outer wall surface segmented according to a second threshold. The preset second threshold or the second threshold actively selected by the user is 9.77 mm. The first gastric outer wall surface segmented according to the second threshold is a dark area. In the above method, the area of ​​the first gastric outer wall surface corresponding to the lesion can be obtained more accurately, thereby accurately obtaining the target lesion.

[0073] Step 302 involves obtaining the target lesion based on the outer surface and inner surface of the first gastric wall, including:

[0074] Step 401: Obtain a unique point on the surface of the stomach corresponding to a point on the outer surface of the first stomach wall, where the distance between the two points on the outer surface of the first stomach wall and the inner surface of the stomach wall is the shortest.

[0075] To obtain the target lesion enclosed by the first gastric lateral wall surface and the corresponding gastric endowall surface, in one embodiment, the first gastric lateral wall surface and the gastric endowall surface are surfaces composed of multiple grids, each grid having vertices, and points on both the first gastric lateral wall surface and the gastric endowall surface can be grid vertices. In one embodiment, the first gastric lateral wall surface and the gastric endowall surface are composed of points. The processing device 120 needs to find a unique point on the gastric endowall surface corresponding to a point on the first gastric lateral wall surface using a search algorithm, where the distance between the two points on the first gastric lateral wall surface and the gastric endowall surface is the shortest. In one embodiment, the search algorithm includes at least one of brute-force search, KD-tree (K-Dimensional Tree), octree, bounding box hierarchy, projection method, pre-computation of signed distance function, and variants of the iterative nearest-point algorithm.

[0076] Step 402: Traverse all points on the surface of the first outer wall of the stomach, obtain the corresponding points on the surface of the inner wall of the stomach, and obtain the surface of the first inner wall of the stomach corresponding to the surface of the first outer wall of the stomach based on the corresponding points.

[0077] Step 403: Obtain the target lesion based on the first gastric inner wall surface corresponding to the first gastric outer wall surface.

[0078] In one embodiment, the processing device 120 traverses all points on the outer surface of the first gastric wall according to the search algorithm mentioned above, and obtains the corresponding points on the inner surface of the gastric wall. The area formed by the corresponding points on the inner surface of the gastric wall is the first inner surface of the gastric wall. The area enclosed by the first inner surface of the gastric wall and the inner surface of the gastric wall is the lesion area. In the above method, the inner surface of the gastric wall corresponding to the outer surface of the first gastric wall can be obtained quickly and accurately, thereby more accurately segmenting the target lesion.

[0079] Step 204, segmenting the target lesion according to the filling status, may further include, when the stomach is underfilled, segmenting the target lesion in response to the user's operation on the gastric wall surface and / or within the gastric cavity.

[0080] In one embodiment, a user interacts with the interactive device 150, and the processing device 120 responds to the user's operation, segmenting the target lesion on the surface of the gastric wall and / or within the gastric cavity. In another embodiment, the user manually selects seed points on the surface of the gastric wall and / or within the gastric cavity, and the growth area obtained based on the seed points is taken as the target lesion. In yet another embodiment, the user manually outlines the target lesion on the surface of the gastric wall and within the gastric cavity.

[0081] like Figure 6 The flowchart 600 shown is a schematic diagram of the steps for generating the second gastric outer wall surface in one embodiment. Step 302, which obtains the target lesion based on the first gastric outer wall surface and the gastric inner wall surface, includes:

[0082] Step 601: In response to the user's operation of selecting a seed point on the surface of the stomach wall, a growth area is formed around the seed point according to a preset growth rule.

[0083] In one embodiment, a user interacts with the interactive device 150 to select a seed point on the surface of the stomach wall. In response to the user's selection of the seed point, the processing device 120 forms a growth region around the seed point according to a preset growth rule. In one embodiment, the interactive action may be that the user selects at least one point on the stomach wall surface as a seed point using a mouse, keyboard, or touch screen. In one embodiment, the preset growth rule may be that growth occurs radiating outwards from the seed point. In one embodiment, in response to the user interrupting the selection of the seed point on the stomach wall surface, the growth region stops growing. A growth region is formed from the seed point to the location where growth stops.

[0084] In one embodiment, a user performs an interactive action through the interactive device 150 to select multiple seed points on the surface of the stomach wall. In response to the user's selection of seed points, the processing device 120 can use a preset growth rule to form a growth region by surrounding the multiple seed points.

[0085] Step 602: Generate a second gastric outer wall surface based on the growth region and the first gastric outer wall surface. Use the second gastric outer wall surface as a new first gastric outer wall surface to perform the acquisition of the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

[0086] In one embodiment, the processing device 120 iteratively generates a second gastric outer wall surface based on the growth region of the first gastric outer wall surface. Using the second gastric outer wall surface as the new first gastric outer wall surface, the processing device 120 obtains the target lesion based on the new first gastric outer wall surface and the gastric inner wall surface. In another embodiment, the processing device 120 obtains a corresponding new first gastric inner wall surface based on the new first gastric outer wall surface, and then obtains the target lesion. Specific steps are described in steps 402-403, and will not be elaborated here.

[0087] The above method adjusts the surface of the first gastric outer wall according to needs, which is more in line with the user's operating habits and makes up for the problem of inaccuracy in directly segmenting the surface of the first gastric outer wall, making the segmentation of the first gastric outer wall surface more accurate, thereby improving the accuracy of the segmentation results of the target lesion.

[0088] In one embodiment, step 602 may further include generating the second gastric outer wall surface in three ways.

[0089] Method 1: In response to the user's first operation of selecting a seed point on the surface of the stomach wall, a first growth region is formed around the seed point according to a preset growth rule; in response to the user interrupting the operation of selecting a seed point on the surface of the stomach wall, the first growth region stops growing; the union of the first growth region and the first surface of the stomach wall generates a second surface of the stomach wall. The first growth region is the area that needs to be added to the first surface of the stomach wall.

[0090] Method 2: In response to the user's second operation of selecting a seed point on the gastric outer wall surface, a second growth region is formed around the seed point according to a preset growth rule; in response to the user interrupting the selection of the seed point on the gastric outer wall surface, the second growth region stops growing; the difference between the first gastric outer wall surface and the second growth region generates a second gastric outer wall surface. The first growth region is the area on the first gastric outer wall surface that needs to be removed.

[0091] Method 3: The second gastric outer wall surface is generated by taking the union of the first gastric outer wall surface and the first growth region, and then subtracting the union from the second growth region. The first and second growth regions can be obtained using the methods described in Methods 1 and 2, which will not be repeated here.

[0092] Using any of the three methods described above, a second gastric outer wall surface is generated. This second gastric outer wall surface is then used as the new first gastric outer wall surface to obtain the target lesion based on the first gastric outer wall surface and the gastric inner wall surface. Based on the user's operation on the interactive device 150, the processing device 120 adds a first growth region and subtracts a second growth region from the first gastric outer wall surface to generate a second gastric outer wall surface. This second gastric outer wall surface is then used as the new first gastric outer wall surface to obtain the target lesion. The specific steps for determining the corresponding first gastric inner wall surface based on the first gastric outer wall surface and then obtaining the target lesion are described in steps 402-403, and will not be elaborated here.

[0093] like Figure 7 The flowchart 700 shown is a schematic diagram of the steps for determining the influence factors of candidate lesions and retaining target lesions in one embodiment. Step 302, which obtains the target lesion based on the first gastric outer wall surface and the gastric inner wall surface, includes the following steps:

[0094] Step 701: Determine the influencing factors of candidate lesions. The influencing factors include one or more of the following: thickening weight, lesion enhancement weight, lymph node weight, and vascular tortuosity weight.

[0095] The processing device 120 directly segments the target lesion based on the outer and inner surfaces of the first gastric wall. However, missegmentation is possible. Therefore, lesions directly segmented based on these surfaces are defined as candidate lesions. Candidate lesions can also be understood as target lesions that have not undergone influencing factor analysis. Thickness weight w1 characterizes the degree of thickening in candidate lesions; a higher thickness weight indicates a thicker candidate lesion and a higher probability of it being a target lesion. Enhancement weight w2 characterizes the degree of image enhancement of the candidate lesion under contrast imaging; a higher enhancement weight indicates a higher degree of image enhancement and a higher probability of it being a target lesion. Lymph node weight w3 characterizes the specificity of lymph nodes surrounding the candidate lesion; a higher lymph node weight indicates a greater likelihood of lymph node specificity and a higher probability of it being a target lesion. Vascular tortuosity weight w4 characterizes the presence of tortuosity in the blood vessels surrounding the candidate lesion; the presence of tortuosity increases the probability of the candidate lesion becoming a target lesion compared to a candidate lesion without surrounding tortuosity. In one embodiment, the lymph nodes surrounding the candidate lesion refer to lymph nodes radiating outwards within a predetermined distance from the lesion. In another embodiment, the blood vessels surrounding the candidate lesion refer to blood vessels radiating outwards within a predetermined distance from the lesion.

[0096] Step 702: Calculate the confidence level of candidate lesions based on the impact factor. If the confidence level is greater than or equal to the third threshold, the candidate lesion is retained as the target lesion; if the confidence level is less than the third threshold, the candidate lesion is removed.

[0097] In one embodiment, the confidence level of a candidate lesion determines whether it can be retained as a target lesion. The processing device 120 calculates the confidence level of the candidate lesion based on the influence factor and the confidence level calculation rule. When the confidence level is greater than or equal to a third threshold, the candidate lesion is retained as a target lesion; when the confidence level is less than the third threshold, the candidate lesion is removed and not used as a target lesion. In one embodiment, the third threshold can be any value in [0.7, 0.9]. The third threshold can be set according to the user's needs. In one embodiment, the processing device 120 directly segments three candidate lesions A, B, and C based on the first gastric outer wall surface and the gastric inner wall surface. The confidence levels of the candidate lesions are calculated according to the influence factor and the confidence level calculation rule: A_DC = f(w1_a, w2_a, w3_a, w4_a), B_DC = f(w1_b, w2_b, w3_b, w4_b), and C_DC = f(w1_c, w2_c, w3_c, w4_c). Where A_DC represents the confidence level of candidate lesion A, w1_a represents the thickening weight of candidate lesion A, w2_a represents the enhancement weight of candidate lesion A, w3_a represents the lymph node weight of candidate lesion A, and w4_a represents the vascular tortuosity weight of candidate lesion A. B_DC represents the confidence level of candidate lesion B, w1_b represents the thickening weight of candidate lesion B, w2_b represents the enhancement weight of candidate lesion B, w3_b represents the lymph node weight of candidate lesion B, and w4_b represents the vascular tortuosity weight of candidate lesion B. C_DC represents the confidence level of candidate lesion C, w1_c represents the thickening weight of candidate lesion C, w2_c represents the enhancement weight of candidate lesion C, w3_c represents the lymph node weight of candidate lesion C, and w4_c represents the vascular tortuosity weight of candidate lesion C. If the confidence level of candidate lesion A is less than the third threshold, candidate lesion A is removed from the target lesion pool. If the confidence levels of candidate lesion B and candidate lesion C are greater than or equal to the third threshold, candidate lesions B and C are retained as target lesions. In one embodiment, the confidence level calculation rule can be at least one of linear weighted summation, nonlinear function, combination of logical rules, and statistical or probabilistic models.

[0098] In one embodiment, the thickening weight is positively correlated with at least one of the maximum thickness value and the area of ​​the candidate lesion. A larger maximum thickness value results in a greater thickening weight; conversely, a smaller maximum thickness value results in a smaller thickening weight. Similarly, a larger area results in a greater thickening weight; conversely, a smaller area results in a smaller thickening weight.

[0099] In one embodiment, the lesion enhancement weight is related to at least one of the average gray value of the candidate lesion voxel and the uniformity of the gray distribution of the candidate lesion voxel.

[0100] In one embodiment, the enhancement weight of a lesion is positively correlated with the average gray value of the candidate lesion voxels; the larger the average gray value of the candidate lesion voxels, the greater the enhancement weight; conversely, the smaller the average gray value of the candidate lesion voxels, the smaller the enhancement weight. In another embodiment, the enhancement weight of a lesion is negatively correlated with the uniformity of the gray-level distribution of the candidate lesion voxels; the worse the uniformity of the gray-level distribution of the candidate lesion voxels, the greater the enhancement weight. A poorer uniformity of the gray-level distribution of the candidate lesion voxels can also be understood as a concentrated distribution of voxel gray values, meaning the gray values ​​are concentrated in one or a few small ranges, rather than widely distributed across the entire gray-level range. Conversely, a uniform gray-level distribution means the gray values ​​cover a wider gray-level space; the better the uniformity of the gray-level distribution of the candidate lesion voxels, the smaller the enhancement weight.

[0101] In one embodiment, lymph node weight is positively correlated with at least one of the specificity of lymph nodes surrounding the candidate lesion and the number of specific lymph nodes. The stronger the specificity of the lymph nodes surrounding the candidate lesion, the greater the lymph node weight; conversely, the weaker the specificity of the lymph nodes surrounding the candidate lesion, the smaller the lymph node weight. The more specific lymph nodes there are, the greater the lymph node weight; conversely, the fewer specific lymph nodes there are, the smaller the lymph node weight. In one embodiment, an increased short diameter of a lymph node indicates lymph node specificity; a longer short diameter compared to a normal lymph node indicates stronger lymph node specificity.

[0102] In one embodiment, the vascular tortuosity weight is correlated with the presence of a tortuosity shadow on the nearest vessel to the candidate lesion. In one embodiment, the vascular tortuosity weight is retained when a tortuosity shadow is present, and removed when a tortuosity shadow is absent.

[0103] like Figure 5 As shown, in one embodiment, the surface of the first gastric outer wall is displayed on the interface according to a second threshold. The interactive device 150 has an interface, and the processing device 120 calculates the surface of the first gastric outer wall according to the second threshold and displays it on the interface of the interactive device 150. The calculation of the surface of the first gastric outer wall by the processing device 120 according to the second threshold is described in step 302, and will not be repeated here. The processing device 120 also adjusts the second threshold in response to a user's adjustment operation. The user can adjust the second threshold as needed. The processing device 120 updates the surface of the first gastric outer wall on the interface according to the adjusted second threshold. According to the above method, the user can independently adjust the second threshold as needed, segment and display the surface of the first gastric outer wall according to the second threshold, providing a more intuitive way for the user to determine the surface of the first gastric outer wall. Furthermore, the user can also intuitively judge whether the surface of the first gastric outer wall meets the requirements. If it does not meet the requirements, the user can further adjust the surface of the first gastric outer wall according to steps 601-602 to more accurately segment the target lesion.

[0104] In one embodiment, in response to a user's editing operation, the three-dimensional medical image of the target lesion is edited layer by layer to obtain an edited three-dimensional medical image of the target lesion. The segmentation result of the lesion is displayed in 3D on the interactive device 130, and a pop-up window prompts the user to confirm the target lesion. If the user clicks the confirmation option, the target lesion is confirmed. If the displayed target lesion result does not meet the user's needs, the user can edit the three-dimensional medical image of the target lesion through the display device 150. For example, the user can edit the three-dimensional medical image of the target lesion layer by layer or multiple layers. The processing device 120, in response to the user's editing operation on the interactive device 150, performs layer-by-layer editing on the three-dimensional medical image of the target lesion to obtain an edited three-dimensional medical image of the target lesion. According to the above method, the user can edit the target lesion himself, so that the shape of the target lesion is closer to the actual shape of the current patient, which facilitates subsequent surgical planning based on the target lesion and improves the accuracy of subsequent surgery.

[0105] It should be understood that, although Figures 2-3 , Figures 6-7 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figures 2-3 , Figures 6-7 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0106] In one embodiment, a segmentation system for gastric lesion images is provided, the internal structure of which can be shown as follows: Figure 8As shown, the gastric lesion image segmentation system includes a processor, memory, input / output (I / O) interface, and communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data related to the aforementioned methods. The I / O interface allows the processor to exchange information with external devices. The communication interface allows communication with external interactive devices via a network connection. When executed by the processor, the computer program implements an image registration method.

[0107] Those skilled in the art will understand that Figure 8 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 segmentation system of the gastric lesion image to which the present application is applied. A specific segmentation system of the gastric lesion image may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0108] In one embodiment, a segmentation system for images of gastric lesions is provided, the system including a processor for performing the following steps:

[0109] Obtain medical images of the stomach;

[0110] Segmenting gastric medical images to obtain the outer and inner surfaces of the stomach;

[0111] The volume relationship between the gastric cavity volume and the gastric volume is obtained based on the outer and inner surfaces of the gastric wall.

[0112] Based on the volume relationship, determine the fullness of the stomach, and segment the target lesion according to the fullness.

[0113] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor that, when performing the step of segmenting the gastric medical image to obtain the outer and inner surfaces of the stomach, includes performing the following steps:

[0114] Calculate the gastric cavity volume based on the surface area of ​​the gastric inner wall;

[0115] Calculate stomach volume based on the surface area of ​​the stomach's outer wall;

[0116] The volume relationship between the gastric cavity volume and the gastric body volume is obtained based on the gastric cavity volume and the gastric body volume.

[0117] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor, and the volume relationship is the ratio of the gastric cavity volume to the gastric volume. The processor, when performing operations to determine the gastric fullness based on the volume relationship and segment the target lesion based on the fullness, includes the following steps:

[0118] When the volume ratio is greater than or equal to the first threshold, the stomach is considered to be full.

[0119] When the volume ratio is less than the first threshold, the stomach is considered underfilled.

[0120] In one embodiment, a segmentation system for images of gastric lesions is provided. The system includes a processor, which, in performing segmentation of target lesions based on filling conditions, further includes performing the following steps:

[0121] When the stomach is full, the thickness of the gastric arm is calculated based on the surface of the outer wall and the surface of the inner wall of the stomach.

[0122] The region with a gastric wall thickness greater than or equal to a second threshold is segmented to obtain at least one first gastric outer wall surface, and the target lesion is obtained based on the first gastric outer wall surface and the gastric inner wall surface.

[0123] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor, a first gastric outer wall surface including a plurality of outer wall points, and a gastric inner wall surface including a plurality of inner wall points. The processor, in acquiring a target lesion based on the first gastric outer wall surface and the gastric inner wall surface, further includes performing the following steps:

[0124] For each of the outer wall points, a corresponding target point is determined from the plurality of inner wall points; wherein the distance between the corresponding outer wall point and the target point is the minimum distance among the distances between the outer wall point and each of the inner wall points.

[0125] Based on the target points corresponding to each of the aforementioned outer wall points, determine the first gastric inner wall surface corresponding to the first gastric outer wall surface;

[0126] The target lesion is obtained from the surface of the first gastric inner wall corresponding to the surface of the first gastric outer wall.

[0127] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor, which, in addition to performing the steps of acquiring target lesions based on a first gastric outer wall surface and a gastric inner wall surface, further includes performing the following steps:

[0128] In response to the user's selection of a seed point on the surface of the stomach wall, a growth area is formed around the seed point according to a preset growth rule;

[0129] A second gastric outer wall surface is generated based on the growth region and the first gastric outer wall surface. This second gastric outer wall surface is used as a new first gastric outer wall surface to perform the acquisition of the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

[0130] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor, which, in addition to performing the steps of acquiring target lesions based on a first gastric outer wall surface and a gastric inner wall surface, further includes performing the following steps:

[0131] Candidate lesions were obtained based on the outer and inner surfaces of the first gastric wall.

[0132] Determine the influencing factors of candidate lesions, including one or more of the following: thickening weight, lesion enhancement weight, lymph node weight, and vascular tortuosity weight;

[0133] The confidence level of the candidate lesion is calculated based on the impact factor. When the confidence level is greater than or equal to the third threshold, the candidate lesion is retained as the target lesion; when the confidence level is less than the third threshold, the candidate lesion is removed.

[0134] In one embodiment, a segmentation system for images of gastric lesions is provided, the system including a processor, the processor being configured to perform the following steps:

[0135] The thickening weight is positively correlated with at least one of the maximum thickness value of the candidate lesion and the area of ​​the candidate lesion;

[0136] The enhancement weight of a lesion is related to at least one of the average gray value of the candidate lesion voxels and the uniformity of the gray value distribution of the candidate lesion voxels.

[0137] Lymph node weight is positively correlated with at least one of the specificity of lymph nodes surrounding candidate lesions and the number of specific lymph nodes.

[0138] The weight of vascular tortuosity is related to whether there is a tortuous shadow in the nearest blood vessel to the candidate lesion.

[0139] In one embodiment, a segmentation system for gastric lesion images is provided. The system includes a processor, which, in performing segmentation of target lesions based on the filling condition, further includes performing the following steps:

[0140] When the stomach is underfilled, the target lesion is segmented in response to the user's operation on the surface of the stomach wall and / or within the stomach cavity.

[0141] In one embodiment, a segmentation system for images of gastric lesions is provided, the system including a processor, the processor being configured to perform the following steps:

[0142] The first gastric outer wall surface is displayed in the interface according to the second threshold.

[0143] In response to the user's adjustment action, adjust the second threshold;

[0144] The first gastric outer wall surface in the interface is updated according to the adjusted second threshold.

[0145] In one embodiment, a segmentation system for images of gastric lesions is provided, the system including a processor, the processor being configured to perform the following steps:

[0146] In response to the user's editing operation, the three-dimensional medical image of the target lesion is sliced ​​and edited to obtain the edited three-dimensional medical image of the target lesion.

[0147] In one embodiment, such as Figure 9 As shown, a segmentation device 900 for gastric lesion images is provided, comprising: an acquisition module 901, a first segmentation module 902, a volume relationship module 903, and a second segmentation module 904, wherein:

[0148] Acquisition module 901 is used to acquire medical images of the stomach;

[0149] The first segmentation module 902 is used to segment the gastric medical image to obtain the outer surface of the stomach and the inner surface of the stomach.

[0150] The volume relationship module 903 is used to obtain the volume relationship between the gastric cavity volume and the gastric volume based on the outer surface of the gastric wall and the inner surface of the gastric wall;

[0151] The second segmentation module 904 determines the fullness of the stomach based on volume relationships and segments the target lesion accordingly.

[0152] For specific limitations regarding the segmentation device for gastric lesion images, please refer to the limitations of the segmentation method for gastric lesion images described above. The steps for implementing the above method embodiments will not be repeated here. Each module in the above-described segmentation device for gastric lesion images can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in a computer device in hardware form, or it can be stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0153] In one embodiment, a computer device is provided, which may be an interactive device, and its internal structure diagram may be as follows: Figure 10As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing 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 communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for segmenting images of gastric lesions. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0154] Those skilled in the art will understand that Figure 10 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.

[0155] In one embodiment, a computer device is provided, including a memory and a processor that, when executing a computer program, implements the steps in the above-described method embodiments.

[0156] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0157] 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 methods described above. Any references to memory, storage, 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, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0158] 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.

[0159] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. 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 patent application should be determined by the appended claims.

Claims

1. A method for segmenting images of gastric lesions, characterized in that, Includes the following steps: Obtain medical images of the stomach; The gastric medical image was segmented to obtain the outer and inner surfaces of the stomach. The volume relationship between the gastric cavity volume and the gastric volume is obtained based on the outer surface of the gastric wall and the inner surface of the gastric wall. Based on the volume relationship, the fullness of the stomach is determined, and the target lesion is segmented according to the fullness.

2. The method according to claim 1, characterized in that, The step of obtaining the volume relationship between the gastric cavity volume and the gastric volume based on the outer surface of the gastric wall and the inner surface of the gastric wall includes: The gastric cavity volume is calculated based on the surface of the gastric inner wall. The stomach volume is calculated based on the surface area of ​​the outer wall of the stomach; The volume relationship between the gastric cavity volume and the gastric body is obtained based on the gastric cavity volume and the gastric body.

3. The method according to claim 1, characterized in that, The volume relationship is the volume ratio of the gastric cavity volume to the stomach volume; the step of determining the gastric fullness based on the volume relationship and segmenting the target lesion based on the fullness includes: When the volume ratio is greater than or equal to the first threshold, the stomach is considered to be full. When the volume ratio is less than the first threshold, the stomach is considered underfilled.

4. The method according to claim 3, characterized in that, The segmentation of the target lesion based on the filling condition includes: When the stomach is full, the thickness of the gastric arm is calculated based on the surface of the outer wall of the stomach and the surface of the inner wall of the stomach; The region where the gastric wall thickness is greater than or equal to a second threshold is segmented to obtain at least one first gastric outer wall surface, and the target lesion is obtained based on the first gastric outer wall surface and the gastric inner wall surface.

5. The method according to claim 4, characterized in that, The first outer wall surface of the stomach includes multiple outer wall points, and the inner wall surface of the stomach includes multiple inner wall points; The step of obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface includes: For each of the outer wall points, a corresponding target point is determined from the plurality of inner wall points; wherein the distance between the corresponding outer wall point and the target point is the minimum distance among the distances between the outer wall point and each of the inner wall points. Based on the target points corresponding to each of the aforementioned outer wall points, determine the first gastric inner wall surface corresponding to the first gastric outer wall surface; The target lesion is obtained from the first gastric inner wall surface corresponding to the first gastric outer wall surface.

6. The method according to claim 4, characterized in that, The step of obtaining the target lesion based on the first gastric outer wall surface and the gastric inner wall surface includes: In response to the user's operation of selecting a seed point on the surface of the stomach wall, a growth area is formed around the seed point according to a preset growth rule; A second gastric outer wall surface is generated based on the growth region and the first gastric outer wall surface; the second gastric outer wall surface is used as a new first gastric outer wall surface to perform the acquisition of the target lesion based on the first gastric outer wall surface and the gastric inner wall surface.

7. The method according to claim 4, characterized in that, The step of obtaining the target lesion based on the first outer gastric wall surface and the inner gastric wall surface includes: Candidate lesions were obtained based on the first outer gastric wall surface and the inner gastric wall surface; Determine the influencing factors of candidate lesions, including one or more of the following: thickening weight, lesion enhancement weight, lymph node weight, and vascular tortuosity weight; The confidence level of the candidate lesion is calculated based on the impact factor. When the confidence level is greater than or equal to the third threshold, the candidate lesion is retained as the target lesion; when the confidence level is less than the third threshold, the candidate lesion is removed.

8. The method according to claim 7, characterized in that, The thickening weight is positively correlated with at least one of the maximum thickness value of the candidate lesion and the area of ​​the candidate lesion; The lesion enhancement weight is related to at least one of the average gray value of the candidate lesion voxel and the uniformity of the gray distribution of the candidate lesion voxel; The lymph node weight is positively correlated with at least one of the specificity of the lymph nodes surrounding the candidate lesion and the number of specific lymph nodes; The vascular tortuosity weight is related to whether there is a tortuosity shadow in the nearest blood vessel to the candidate lesion.

9. The method according to claim 3, characterized in that, The segmentation of the target lesion based on the filling condition may further include: When the stomach is underfilled, the target lesion is segmented in response to the user's operation on the surface of the stomach wall and / or within the stomach cavity.

10. The method according to any one of claims 4-8, characterized in that, Also includes: Display the first gastric outer wall surface in the interface according to the second threshold; In response to the user's adjustment operation, adjust the second threshold; The first gastric outer wall surface in the interface is updated according to the adjusted second threshold.

11. The method according to any one of claims 1-9, characterized in that, Also includes: In response to the user's editing operation, the three-dimensional medical image of the target lesion is sliced ​​and edited to obtain the edited three-dimensional medical image of the target lesion.

12. A segmentation system for images of gastric lesions, characterized in that, Includes a processor, the processor being configured to perform the following steps: Obtain medical images of the stomach; The gastric medical image was segmented to obtain the outer and inner surfaces of the stomach. The volume relationship between the gastric cavity volume and the gastric volume is obtained based on the outer surface of the gastric wall and the inner surface of the gastric wall. Based on the volume relationship, the fullness of the stomach is determined, and the target lesion is segmented according to the fullness.

13. A segmentation device for images of gastric lesions, characterized in that, The device includes: The acquisition module is used to acquire medical images of the stomach. The first segmentation module is used to segment the gastric medical image to obtain the outer surface of the stomach and the inner surface of the stomach. The volume relationship module is used to obtain the volume relationship between the gastric cavity volume and the gastric volume based on the outer surface of the gastric wall and the inner surface of the gastric wall; The second segmentation module determines the fullness of the stomach based on the volume relationship and segments the target lesion based on the fullness.

14. 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.

15. 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.

Images