Brain grid model determination method and device, equipment and storage medium

Abstract

The invention discloses a brain grid model determination method and device, equipment and a storage medium. The method comprises the steps that an initial image stack corresponding to a brain medical image is determined, two adjacent binary images in the initial image stack serve as a detection group, the initial image stack comprises at least two two-dimensional images, and the two-dimensional images are binary images; for each detection group, determining a connected region detection result of the current detection group, and removing a connected region of which the area is greater than a set threshold value from the connected region detection result to update the current detection group; determining an isolated overlapping stack based on all the updated detection groups, and removing a connected region spanning at least two detection groups in the isolated overlapping stack to obtain a distortion array; and determining a target brain grid model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-established non-distortion lookup table and a distortion lookup table corresponding to the distortion array. According to the technical scheme, the accuracy of the brain grid model can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and particularly to a method, device, equipment and storage medium for determining a brain mesh model. Background Art

[0002] In the field of brain science, the three-dimensional imaging data volume of the primate brain with fine structures (mainly manifested as complex sulci and gyri) is relatively large. When displaying the whole-brain contour, the mesh model converted from the image data is usually used as the display data. However, this mesh model usually cannot accurately depict the shape characteristics of the brain sulci and gyri.

[0003] Therefore, the brain mesh model determined by the prior art has the problem of low accuracy. Summary of the Invention

[0004] The present invention provides a method, device, equipment and storage medium for determining a brain mesh model to solve the problem that the brain mesh model determined by the prior art has low accuracy.

[0005] According to one aspect of the present invention, a method for determining a brain mesh model is provided, including:

[0006] Determine an initial image stack corresponding to a brain medical image, and use two adjacent binary images in the initial image stack as a detection group. The initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images;

[0007] For each detection group, determine the connected region detection result of the current detection group, and remove the connected regions with an area larger than a set threshold from the connected region detection result to update the current detection group;

[0008] Based on all the updated detection groups, determine an isolated overlapping stack, and remove the connected regions in the isolated overlapping stack that span at least two detection groups to obtain a distortion array;

[0009] Based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, determine the target brain mesh model corresponding to the brain medical image.

[0010] According to another aspect of the present invention, a device for determining a brain mesh model is provided, including:

[0011] A detection group module, configured to determine an initial image stack corresponding to a brain medical image, and use two adjacent binary images in the initial image stack as a detection group. The initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images;

[0012] A detection group update module, configured to determine, for each of the detection groups, a connected region detection result of a current detection group, and update the current detection group by removing connected regions with an area greater than a set threshold from the connected region detection result;

[0013] A distortion array determination module, configured to determine an isolated overlapping stack based on all the updated detection groups, and obtain a distortion array by removing connected regions that span at least two of the detection groups from the isolated overlapping stack;

[0014] A brain mesh model determination module, configured to determine a target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array.

[0015] According to another aspect of the present invention, there is provided an electronic device, the electronic device includes:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the brain mesh model determination method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the brain mesh model determination method according to any embodiment of the present invention when executed.

[0020] The technical solution of the brain mesh model determination method provided by the embodiments of the present invention removes connected regions with an area greater than a set threshold from the connected region detection results of the detection groups, so as to achieve the purpose of removing the internal brain structure from the detection groups; by removing connected regions that span at least two detection groups from the isolated overlapping stack, the non-distortion regions in the isolated overlapping stack are removed, thereby obtaining the distortion regions, that is, the distortion array; through the coordinated use of the pre-created non-distortion lookup table for non-distortion regions and the distortion lookup table for the distortion array, the merging of the brain mesh model corresponding to the part of the distortion array and the part of the brain mesh model corresponding to the non-distortion data in the initial image stack is completed, and a target brain mesh model without distortion regions is obtained, improving the accuracy of determining the target brain mesh model.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Brief Description of the Drawings

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0023] Figure 1 is a flowchart of a method for determining a brain mesh model according to an embodiment of the present invention;

[0024] Figure 2A is a schematic diagram of an initial image stack according to an embodiment of the present invention;

[0025] Figure 2B is a schematic diagram of an isolated overlapping stack according to an embodiment of the present invention;

[0026] Figure 2C is a schematic diagram of a distortion array according to an embodiment of the present invention;

[0027] Figure 3A is a brain mesh model determined based on the initial image stack and the non-distortion lookup table according to an embodiment of the present invention;

[0028] Figure 3B is a target brain mesh model according to an embodiment of the present invention;

[0029] Figure 4 is a target brain mesh model displayed in a semi-transparent mode according to an embodiment of the present invention;

[0030] Figure 5 is another flowchart of a method for determining a brain mesh model according to an embodiment of the present invention;

[0031] Figure 6 is a schematic structural diagram of a device for determining a brain mesh model according to an embodiment of the present invention;

[0032] Figure 7 is another schematic structural diagram of a device for determining a brain mesh model according to an embodiment of the present invention;

[0033] Figure 8 is a schematic structural diagram of an electronic device for implementing the method for determining a brain mesh model according to an embodiment of the present invention. Detailed Embodiments

[0034] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0036] Figure 1 The flowchart of the brain mesh model determination method provided by the embodiment of the present invention is applicable to the case of determining the brain surface mesh model through the initial image stack corresponding to the brain medical image and the distortion array corresponding to the initial image stack. This method can be executed by a brain mesh model determination device, which can be implemented in the form of hardware and / or software, and the brain mesh model determination device can be configured in the processor of an electronic device. As Figure 1 shown, the method includes:

[0037] S110. Determine the initial image stack corresponding to the brain medical image, and use two adjacent binary images in the initial image stack as a detection group. The initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images.

[0038] Among them, the brain medical image is a three-dimensional brain CT (Computed Tomography) image or a three-dimensional brain MRI (Magnetic Resonance Imaging) image.

[0039] In one embodiment, downsample the brain medical image to obtain a downsampling result; perform binary segmentation on the downsampling result to obtain a binary result, and use the binary result as the initial image stack.

[0040] Specifically, downsampling is used to convert an image with a higher resolution into an image with a lower resolution, so as to reduce the size of image data by reducing the image resolution. Inevitably, image distortion will be introduced in this process. Binary segmentation is performed on the brain medical image to obtain a binary result. Data format verification is performed on each two-dimensional image in the binary result. After the format verification is successful, the binary result after successful verification is used as the initial image stack.

[0041] In one embodiment, the data is read as a numpy array through python and then stored as 8-bit data. Each two-dimensional image is verified to ensure that the pixel values in each two-dimensional image are either 0 or 255. Then, all two-dimensional images are stitched together to form the initial image stack.

[0042] Take two adjacent binary images in the initial image stack as a detection group (see Figure 2A shown). Exemplarily, it is assumed that the initial image stack includes two-dimensional images numbered from 1 to n. Take the two-dimensional image numbered 1 and the two-dimensional image numbered 2 as a detection group, take the two-dimensional image numbered 2 and the two-dimensional image numbered 3 as a detection group, and so on. Take the two-dimensional image numbered i and the two-dimensional image numbered i + 1 as a detection group, where i is less than or equal to n - 1.

[0043] S120. For each of the detection groups, determine the connected region detection result of the current detection group, and remove the connected regions with an area greater than a set threshold from the connected region detection result to update the current detection group.

[0044] Perform connected region detection on the current detection group to obtain the connected region detection result of the current detection group. It can be understood that the connected region detection result includes the connected regions that cause the distortion of the grid model and the real brain structure.

[0045] In one embodiment, remove the connected regions with an area greater than a set threshold from the connected region detection result to update the current detection group. This set threshold is determined based on the area of the brain conical structure in the cross-section. This embodiment realizes the purpose of removing the brain structure from the connected region detection result by removing the connected regions with an area greater than the set threshold from the connected region detection result, making the updated current detection group not include the brain structure, which is simple and fast.

[0046] S130. Determine an isolated overlapping stack based on all the updated detection groups, and remove the connected regions that span at least two of the detection groups from the isolated overlapping stack to obtain a distortion array.

[0047] Stitch the updated detection groups together to form an isolated overlapping stack (see Figure 2Bas shown). The isolated overlapping stack is a three-dimensional image. Removing the connected regions that span at least two detection groups from the isolated overlapping stack gives a distortion array (see Figure 2C as shown), achieving the purpose of removing the brain surface structures that are continuously distributed longitudinally from the isolated overlapping stack, where the longitudinal direction is from the top of the head to the neck. It can be understood that the distortion array includes surface structures that are not continuously distributed longitudinally.

[0048] S140. Based on the initial image stack, the distortion array, a pre-created non-distortion look-up table, and a distortion look-up table corresponding to the distortion array, determine the target brain mesh model corresponding to the brain medical image.

[0049] Among them, the role of the distortion look-up table is that the partial brain mesh model determined in cooperation with the distortion array is continuous and smooth, without including jagged or hole-like structures. The distortion look-up table addresses the situation where there is a lack of cube representation in the non-distortion look-up table. Therefore, the distortion look-up tables corresponding to non-distortion look-up tables determined based on different algorithms are different. For the specific mesh surfaces corresponding to each cube missing in the distortion look-up table, it can be determined based on experience or based on an approximation method. For example, for the case where each vertex of the cube corresponds to 1 (i.e., the pixel value is 255), its corresponding mesh surface is approximated as a plane, and this plane is parallel to one plane of the cube. Other approximation methods for expression missing situations are not elaborated in this embodiment.

[0050] In one embodiment, the non-distortion look-up table is determined based on the Marching_Cubes (MC) algorithm. The basic idea of this algorithm is to divide the three-dimensional space where the initial image stack is located into a series of small cubes, and generate corresponding meshes according to the presence or absence of the vertex signal values of the cubes. Specifically, for each cube, each vertex with a signal is regarded as the internal region of the generated surface, and interpolation is performed on the common edge where the vertices with signals and without signals are located as the mesh vertices.

[0051] In one embodiment, the target brain mesh model is determined through the following steps:

[0052] Step a1. Determine all the smallest cubes determined by the initial image stack.

[0053] The smallest cube refers to the cube determined by the 8 closest pixels.

[0054] Exemplarily, the initial image stack includes binary images identified from 1 to n. Each two-dimensional image is 256×256. Taking the binary image identified as 1 and the binary image identified as 2 as an example, the pixel combinations in the binary image identified as 1: {(000, 000), (001, 000), (000, 001), (001, 001)} and the pixel combinations in the binary image identified as 2: {(000, 000), (001, 000), (000, 001), (001, 001)} correspond to a minimum cube.

[0055] Step a2: For each minimum cube among all the minimum cubes, if the pixel value of any vertex in the current minimum cube in the initial image stack and the pixel value in the distortion array are both 255, then determine the grid face corresponding to the current minimum cube based on the distortion lookup table and the distortion array.

[0056] If the pixel value of any vertex in the current minimum cube in the initial image stack is 255 and the pixel value in the distortion array is also 255, it means that the vertex is a distorted pixel. Therefore, the current minimum cube belongs to the distorted cube, and it is necessary to determine the grid face corresponding to the current minimum cube based on the distortion lookup table and the distortion array.

[0057] Step a3: If the pixel values of all vertices of the current minimum cube in the distortion array are 0, but the pixel value of at least one vertex in the initial image stack is 255, then determine the grid face corresponding to the current minimum cube based on the non-distortion lookup table and the initial image stack.

[0058] If the pixel values of all vertices of the current minimum cube in the distortion array are 0, it means that the current cube does not include distorted pixels. Therefore, determine the grid face corresponding to the current minimum cube based on the non-distortion lookup table and the initial image stack.

[0059] Step a4: Determine the target brain grid model corresponding to the brain medical image based on the grid faces corresponding to all the minimum cubes.

[0060] After the grid faces corresponding to each minimum cube are determined, determine the target brain grid model corresponding to the brain medical image according to the grid faces corresponding to each minimum cube.

[0061] In one embodiment, display the target brain grid model on the visualization display interface, which is convenient for the user to view the target brain grid model.

[0062] In one embodiment, in response to a rotation operation, a corresponding rotation operation is performed on the target brain mesh model, and the rotation process and rotation result of the brain network model are displayed. This facilitates the user to view the target brain mesh model from various angles.

[0063] Figure 3A is a macaque brain mesh model determined only based on a non-distortion lookup table and an initial image stack, Figure 3B is a macaque brain mesh model determined by using a distortion lookup table, a non-distortion lookup table, an initial image stack, and a distortion array in combination. Figure 3A There are obvious holes. From the position of the holes on the brain mesh model, the holes correspond to the sulcus and gyrus (the main fine structures inside the brain) region. In fact, the sulcus and gyrus should be continuous, as Figure 3B shown. It can be seen that the brain mesh model determination method provided by the present application can determine an accurate target brain mesh model.

[0064] Set Figure 3B to semi-transparent mode to obtain Figure 4 . Figure 4 The folded sulcus and gyrus in the target brain mesh model in

[0065] The technical solution of the brain mesh model determination method provided by the embodiments of the present invention removes the connected regions with an area greater than a set threshold from the connected region detection results of the detection group, so as to achieve the purpose of removing the internal brain structures from the detection group; by removing the connected regions in the isolated overlapping stack that span at least two detection groups, the non-distortion regions in the isolated overlapping stack are removed, thereby obtaining the distortion region, that is, the distortion array; by using the pre-created non-distortion lookup table for the non-distortion region and the distortion lookup table for the distortion array in combination, the merging of the brain mesh model corresponding to the corresponding part of the distortion array and the brain mesh model corresponding to the non-distortion data in the initial image stack is completed, and the target brain mesh model without distortion region is obtained, improving the accuracy of the determination of the target brain mesh model.

[0066] Figure 5 This is another flowchart of the brain mesh model determination method provided by the embodiments of the present invention. This embodiment is used to refine the step of removing the area greater than the set threshold in the above embodiment. As Figure 5 shown, the method includes:

[0067] S210. Determine an initial image stack corresponding to the brain medical image, and use two adjacent binary images in the initial image stack as a detection group. The initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images.

[0068] S2201. For each of the said detection groups, determine the detection result of the connected regions of the current detection group, and remove the largest connected region from the detection result of the connected regions to obtain an isolated overlapping group.

[0069] This step aims to remove the main brain structure by removing the largest connected region from the detection result of the connected regions.

[0070] S2202. Remove the connected regions with an area greater than a set threshold from the isolated overlapping group to update the current detection group.

[0071] This step aims to remove the brain conical structure from the isolated overlapping group by removing the connected regions with an area greater than a set threshold from the isolated overlapping group.

[0072] S2300. Based on all the updated detection groups, determine an isolated overlapping stack, and remove the connected regions in the isolated overlapping stack that span at least two of the said detection groups to obtain a distortion array.

[0073] In one embodiment, after obtaining the distortion array, determine the connected region parameters corresponding to the distortion array, and based on the connected region parameters, determine the review result of the distortion array. The connected region parameters include at least one of the number and volume of the connected regions. If the review result is a review failure, then take two adjacent binary images in the distortion array as a detection group, and return to the step of removing the largest connected region from the detection result of the connected regions for each of the said detection groups to obtain an isolated overlapping group. This embodiment can delete obvious and hidden connected regions, so that the latest distortion array does not include outlier connected regions and passes the review smoothly. It can be understood that if the set threshold remains unchanged, then during the iteration process, the area of the largest connected region removed from the detection result of the connected regions of each detection group is less than the set threshold. This results in S2202 being a no-op, that is, this step will not delete any connected regions because after the first execution of S2202, the size of each connected region in the isolated overlapping group is less than the set threshold.

[0074] In one embodiment, in response to a failed review operation, two adjacent binary images in the distortion array are taken as a detection group, and the step of removing the largest connected region from the connected region detection results for each of the detection groups to obtain an isolated overlapping group is returned. This embodiment allows users to perform manual review on the distortion array, improving the flexibility of use of the system implementing the brain mesh model determination method. If it is determined during the manual review process that the distortion array includes outlier connected regions, an instruction for a failed review is sent to the processor by clicking or touching the corresponding option. In response to this instruction, the processor submits an operation to take two adjacent binary images in the distortion array as a detection group, and returns the step of removing the largest connected region from the connected region detection results for each of the detection groups to obtain an isolated overlapping group, thereby deleting the outlier connected regions in the distortion array by repeating the above-described deletion operation of the connected regions, so that the distortion array only includes the pixel points of the distortion region.

[0075] Since the outlier connected regions are different from other connected regions, the outlier connected regions in the distortion array are considered to correspond to non-distortion regions. Therefore, in one embodiment, the method for the processor to determine the review result includes: determining whether there are outlier connected regions in the distortion array based on the connected region parameters; if so, determining that the review fails; if not, determining that the review succeeds. Exemplarily, it is determined whether there are outlier connected regions in the distortion array based on the number of connected regions included in the distortion array and the area of each connected region. If there are outlier connected regions, it means that there are non-distortion regions in the distortion array, so it is determined that the review fails; if there are no outlier connected regions, it means that there are no non-distortion regions in the distortion array, so it is determined that the review succeeds.

[0076] In one embodiment, if the review result is a failed review, two adjacent binary images in the distortion array are taken as a detection group, and the step of removing the largest connected region from the connected region detection results for each of the detection groups to obtain an isolated overlapping group is returned, that is, the connected region deletion operation is re-executed. Optionally, the processor re-executes the operation of deleting the connected region with the largest area from the detection group to obtain an isolated overlapping array, deletes the connected regions with an area greater than a set threshold from the isolated overlapping array to update the detection group, determines an isolated overlapping stack based on all the updated detection groups, and deletes the connected regions spanning at least two detection groups from the isolated overlapping stack to obtain a distortion array. First, deleting the connected region with the largest area and then deleting the connected regions with an area greater than the set threshold can achieve the purpose of removing the outlier connected regions in the isolated overlapping stack. This embodiment is applicable to the processor to automatically complete the review of the distortion array, and in the case where the review result is a failed review, the non-distortion data in the distortion array is deleted by means of iterative looping to update the distortion array.

[0077] In one embodiment, the outlier connected regions are directly deleted from the distortion array to update the distortion array. This embodiment can simply and directly complete the deletion operation of the outlier connected regions.

[0078] S240. Based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, determine the target brain mesh model corresponding to the brain medical image.

[0079] In one embodiment, in the case of detecting a signal generated by the processor indicating successful review, based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table, determine the target brain mesh model. This embodiment is applicable to the scenario where the processor automatically performs the distortion array review operation.

[0080] In one embodiment, in response to a successful review operation, based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table, determine the target brain mesh model. This embodiment is applicable to the scenario where the distortion array is manually reviewed and the review result is successful.

[0081] In the technical solution provided by the embodiments of the present invention, since the connected region with the largest area can be greater than the set threshold, or less than or equal to the set area threshold, the outlier overlapping array is obtained by deleting the connected region with the largest area in the detection group, and then the connected regions in the outlier overlapping array with an area greater than the set threshold are deleted, which can improve the flexibility of connected region deletion, especially improve the flexibility of connected region deletion in the iterative process.

[0082] Figure 6 It is a schematic structural diagram of the brain mesh model determination device provided by the embodiments of the present invention. As Figure 6 shown, the device includes:

[0083] A detection group module 31, configured to determine an initial image stack corresponding to a brain medical image, and use two adjacent binary images in the initial image stack as a detection group, where the initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images;

[0084] A detection group update module 32, configured to, for each of the detection groups, determine the connected region detection result of the current detection group, and remove the connected regions with an area greater than the set threshold from the connected region detection result to update the current detection group;

[0085] A distortion array determination module 33, configured to determine an outlier overlapping stack based on all the updated detection groups, and remove the connected regions in the outlier overlapping stack that span at least two of the detection groups to obtain a distortion array;

[0086] The brain mesh model determination module 34 is configured to determine a target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array.

[0087] In one embodiment, the detection group update module 32 is configured to:

[0088] Remove the largest connected region from the connected region detection result to obtain an isolated overlap group;

[0089] Remove connected regions with an area greater than a set threshold from the isolated overlap group to update the current detection group.

[0090] In one embodiment, as Figure 7 shown, the apparatus further includes an audit module 35, and the audit module includes:

[0091] An audit unit configured to determine connected region parameters corresponding to the distortion array, and determine an audit result of the distortion array based on the connected region parameters, where the connected region parameters include at least one of the number and volume of connected regions;

[0092] An iteration unit configured to, if the audit result is an audit failure, use two adjacent binary images in the distortion array as a detection group, and return the step of removing the largest connected region from the connected region detection result for each detection group to obtain an isolated overlap group.

[0093] In one embodiment, the audit unit is specifically configured to:

[0094] Determine whether there are outlier connected regions in the distortion array based on the connected region parameters;

[0095] If there are, determine that the audit fails;

[0096] If there are not, determine that the audit is successful.

[0097] In one embodiment, the apparatus further includes an audit module, and the audit module is configured to:

[0098] In response to an audit failure operation, use two adjacent binary images in the isolated overlap stack as a detection group, and return the step of removing the largest connected region from the connected region detection result for each detection group to obtain an isolated overlap group;

[0099] The brain mesh model determination module is configured to:

[0100] In response to the operation of passing the review, based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, determine the target brain mesh model corresponding to the brain medical image.

[0101] In one embodiment, the brain mesh model determination module is specifically configured to:

[0102] Determine all the minimum cubes determined by the initial image stack;

[0103] For each minimum cube among all the minimum cubes, if the pixel value of any vertex in the current minimum cube in the initial image stack and the pixel value in the distortion array are both 255, then based on the distortion lookup table and the distortion array, determine the mesh surface corresponding to the current minimum cube;

[0104] If the pixel values of all vertices of the current minimum cube in the distortion array are 0, but the pixel value of at least one vertex in the initial image stack is 255, then based on the non-distortion lookup table and the initial image stack, determine the mesh surface corresponding to the current minimum cube;

[0105] Based on the mesh surfaces corresponding to all the minimum cubes, determine the target brain mesh model corresponding to the brain medical image.

[0106] In one embodiment, the detection group module is further configured to:

[0107] Downsample the brain medical image to obtain a downsampling result;

[0108] Perform binary segmentation on the downsampling result to obtain a binarization result, and use the binarization result as the initial image stack.

[0109] The technical solution of the brain mesh model determination method provided by the embodiments of the present invention removes the connected regions with an area larger than a set threshold from the connected region detection results of the detection group, so as to achieve the purpose of removing the internal brain structures from the detection group; by removing the connected regions that span at least two detection groups in the isolated overlapping stack, the non-distortion regions in the isolated overlapping stack are removed, thereby obtaining the distortion region, that is, the distortion array; through the combined use of the pre-created non-distortion lookup table for the non-distortion region and the distortion lookup table for the distortion array, the merging of the brain mesh model corresponding to the corresponding part of the distortion array and the brain mesh model corresponding to the non-distortion data in the initial image stack is completed, and the target brain mesh model without distortion regions is obtained, improving the accuracy of determining the target brain mesh model.

[0110] The brain mesh model determination device provided by the embodiments of the present invention can execute the brain mesh model determination method provided by any embodiment of the present invention, and has function modules and beneficial effects corresponding to the execution of the method.

[0111] Figure 8 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, personal digital processing and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0112] As Figure 8 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0113] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0114] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the brain mesh model determination method.

[0115] In some embodiments, the method for determining a brain mesh model can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining a brain mesh model described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the method for determining a brain mesh model by any other suitable means (e.g., by means of firmware).

[0116] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0117] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0118] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0119] In order to provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0120] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0121] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0122] It should be understood that the various forms of processes shown above can be used, steps can be reordered, added or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0123] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining a brain mesh model, characterized in that, it includes: Determine an initial image stack corresponding to a brain medical image, and use two adjacent binary images in the initial image stack as a detection group. The initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images; For each of the detection groups, determine the connected region detection result of the current detection group, and remove the connected regions with an area greater than a set threshold from the connected region detection result to update the current detection group; Determine an isolated overlapping stack based on all the updated detection groups, and remove the connected regions that span at least two of the detection groups from the isolated overlapping stack to obtain a distortion array; Based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, determine the target brain mesh model corresponding to the brain medical image.

2. The method according to claim 1, characterized in that, The step of removing the connected regions with an area greater than a set threshold from the connected region detection result to update the current detection group includes: Removing the largest connected region from the connected region detection result to obtain an isolated overlapping group; Removing the connected regions with an area greater than a set threshold from the isolated overlapping group to update the current detection group.

3. The method according to claim 2, characterized in that, Before determining the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, it further includes: Determine the connected region parameters corresponding to the distortion array, and determine the review result of the distortion array based on the connected region parameters. The connected region parameters include at least one of the number and volume of connected regions; If the review result is a review failure, use two adjacent binary images in the distortion array as a detection group, and return to the step of removing the largest connected region from the connected region detection result for each of the detection groups to obtain an isolated overlapping group.

4. The method according to claim 3, characterized in that, The step of determining the review result of the distortion array based on the connected region parameters includes: Determine whether there are outlier connected regions in the distortion array based on the connected region parameters; If so, determine that the review fails; If not, determine that the review is successful.

5. The method according to claim 2, characterized in that, Before determining the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created non-distortion lookup table, and a distortion lookup table corresponding to the distortion array, it further includes: In response to a review failure operation, use two adjacent binary images in the isolated overlapping stack as a detection group, and return to the step of removing the largest connected region from the connected region detection result for each of the detection groups to obtain an isolated overlapping group; Determining the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created undistorted lookup table, and a distortion lookup table corresponding to the distortion array includes: In response to an approval operation, determining the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created undistorted lookup table, and a distortion lookup table corresponding to the distortion array.

6. The method according to claim 1, wherein, determining the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created undistorted lookup table, and a distortion lookup table corresponding to the distortion array includes: Determining all the minimum cubes determined by the initial image stack; For each minimum cube among all the minimum cubes, if the pixel values of any vertex in the current minimum cube in the initial image stack and in the distortion array are both 255, then determining the mesh surface corresponding to the current minimum cube based on the distortion lookup table and the distortion array; If the pixel values of all vertices of the current minimum cube in the distortion array are 0, but the pixel value of at least one vertex in the initial image stack is 255, then determining the mesh surface corresponding to the current minimum cube based on the undistorted lookup table and the initial image stack; Determining the target brain mesh model corresponding to the brain medical image based on the mesh surfaces corresponding to all the minimum cubes.

7. The method according to claim 1, wherein, determining the initial image stack corresponding to the brain medical image includes: Performing downsampling on the brain medical image to obtain a downsampling result; Performing binary segmentation on the downsampling result to obtain a binarization result, and using the binarization result as the initial image stack.

8. A brain mesh model determination device, wherein, it includes: A detection group module, configured to determine an initial image stack corresponding to a brain medical image, and use two adjacent binary images in the initial image stack as a detection group, where the initial image stack includes at least two two-dimensional images, and the two-dimensional images are binary images; A detection group update module, configured to, for each of the detection groups, determine a connected region detection result of the current detection group, and remove connected regions with an area greater than a set threshold from the connected region detection result to update the current detection group; A distortion array determination module, configured to determine an isolated overlapping stack based on all the updated detection groups, and remove connected regions that span at least two of the detection groups from the isolated overlapping stack to obtain a distortion array; A brain mesh model determination module, configured to determine the target brain mesh model corresponding to the brain medical image based on the initial image stack, the distortion array, a pre-created undistorted lookup table, and a distortion lookup table corresponding to the distortion array.

9. An electronic device, wherein, the electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and when executed by the at least one processor, enables the at least one processor to execute the brain grid model determination method according to claims 1-7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions for implementing the brain grid model determination method according to claims 1-7 when executed by a processor.

Images