Brain image processing method and device based on 3D Slicer and program product

Abstract

The invention relates to the field of intelligent medical treatment, in particular to a brain image processing method and device based on 3D Slicer and a program product. Comprising the following steps: S1, acquiring brain image sequence data of a patient; s2, performing registration on the image sequence data through a 3D Slicer to obtain a registered image sequence; s3, performing three-dimensional reconstruction based on the registered sequence to obtain a brain three-dimensional model; s4, selecting a sequence serving as a center coordinate during registration or an image sequence after registration is completed and the coordinate is updated; s5, screening out a sequence containing a whole brain sequence in a scanning range from the sequence serving as the center coordinate during registration or the image sequence after registration and coordinate updating to obtain a screened image sequence; and S6, converting the brain three-dimensional model into a Labelmap file based on the screened image sequence. Three-dimensional model data reconstructed by modeling software such as Slicer and the like can be imported into an existing neural navigation system, so that the functions of the three-dimensional model data and the existing neural navigation system are combined, higher-quality existing intraoperative navigation is realized, and the method has very good clinical value.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical treatment, and specifically to a brain image processing method, device, program product and computer-readable storage medium based on 3D Slicer. Background Art

[0002] Neuronavigation systems (such as Brainlab neuronavigation) are currently the most important and most commonly used intraoperative auxiliary positioning systems for neurosurgery worldwide. They can effectively help neurosurgeons accurately locate and properly protect important anatomical structures (such as important brain functional area cortex, blood vessels and other structures) during surgery, significantly improving the quality of surgery. However, due to the obvious limitations of the segmentation and reconstruction functions of the current mainstream neuronavigation systems, they are unable to accurately and high-quality reconstruct some cranial nerves (such as the facial nerve), precentral gyrus and other anatomical structures, making their role in some neurosurgery operations extremely limited.

[0003] 3D Slicer software is a powerful and free open-source image post-processing software, which has been well known and used by neurosurgeons around the world. At present, there have been many studies and reports on neurosurgeons using 3D Slicer software for preoperative reconstruction and assisting neurosurgery. Studies have confirmed that 3D Slicer software can effectively assist in the implementation of a variety of neurosurgery operations, significantly improve the quality of surgery, and effectively reduce the occurrence of surgical side injuries and postoperative complications. Among them, the use of 3D Slicer software to reconstruct structures such as the facial nerve before surgery and assist in the surgical resection of intracranial tumors can significantly reduce surgical side injuries and improve the functional retention rate of the facial nerve, which has been recognized by many neurosurgeons. However, the auxiliary function of the three-dimensional images reconstructed by 3D Slicer software during surgery is limited to the presentation of the three-dimensional model image on the display screen. The reconstructed three-dimensional images cannot be directly imported into the neuronavigation system that only recognizes the Digital Imaging and Communications in Medicine (DICOM) format files, and it is impossible to use the existing neuronavigation system for intraoperative navigation and positioning of intracranial tumors. Summary of the invention

[0004] In view of the above problems, the present invention proposes a brain image processing method based on 3D Slicer, which imports the three-dimensional model image of the facial nerve and other structures reconstructed by 3DSlicer software into the existing neuronavigation system to combine the advantages of both, and finally realizes the navigation and positioning of the facial nerve in acoustic neuroma surgery, which specifically includes:

[0005] S1. Obtaining patient brain imaging sequence data;

[0006] S2, the image sequence data is registered by 3D Slicer to obtain a registered image sequence;

[0007] S3, performing three-dimensional reconstruction based on the registered sequence to obtain a three-dimensional brain model;

[0008] S4, selecting a sequence used as the center coordinates during registration or an image sequence after registration has been completed and the coordinates have been updated;

[0009] S5, selecting an image sequence whose scanning range includes a whole-brain sequence from the sequence used as the center coordinate during the registration or the image sequence after the registration is completed and the coordinates are updated;

[0010] S6, the three-dimensional brain model is converted into a Labelmap file based on the filtered image sequence. Further, S5 also includes performing thickness screening on the filtered sequence, screening the sequence with the thinnest scanning layer or the smallest voxel to obtain the image sequence after thickness screening, and the three-dimensional brain model and the image sequence after thickness screening are converted into a Labelmap file.

[0011] The registration is the registration of each image sequence with the image sequence as the center coordinate before modeling;

[0012] The registration is to unify the coordinates of all image sequences before establishing the brain model, so as to ensure that the coordinate system of the image sequence data is consistent with the coordinate system of the three-dimensional reconstruction model of the brain.

[0013] The brain model is composed of models of surgery-related anatomical structures obtained by three-dimensional reconstruction of various anatomical structures related to brain and intracranial surgery, and each anatomical structure model is marked with a different color;

[0014] Optionally, the anatomical structures related to brain and intracranial surgery include one or more of the following: brain stem, facial nerve, trigeminal nerve, skull, vein;

[0015] Optionally, the brain imaging data includes one or more of the following: MRI, CT or DSA;

[0016] Optionally, based on the anatomical structure models related to the brain and intracranial surgery, steps S4-S5 are sequentially performed to obtain image sequences after screening of the anatomical structure models, and each anatomical structure model is converted into a Labelmap file based on the screened image sequences.

[0017] The object of the present invention is to provide a method for locating intracranial tumors based on intraoperative neuronavigation, comprising:

[0018] The method performs image processing by the brain image processing method based on 3D Slicer to obtain a Labelmap file of anatomical structures related to intracranial tumor surgery; wherein, in the method for locating intracranial tumors based on intraoperative neuronavigation, the brain model three-dimensionally reconstructed by the brain image processing method based on 3D Slicer also includes a tumor and a facial nerve;

[0019] Convert the Labelmap file into DICOM data;

[0020] The intraoperative neuronavigation system acquires the DICOM data and the original MRI and CT image data of the patient;

[0021] The DICOM data and the original image data of the patient are sequentially fused to obtain fused data;

[0022] Segmenting the target structure in the fused data to obtain segmentation data of surgery-related structures;

[0023] Tumors or nerves are localized based on the segmentation data.

[0024] Furthermore, the method also includes data labeling, labeling the DICOM data to obtain a labeled data set; intraoperative neuronavigation acquires the labeled data and the patient's brain imaging data.

[0025] The tumors in the method include one or more of the following: acoustic neuroma, trigeminal schwannoma, parasinus falx, convex or functional area meningioma and lymphoma, brain parenchymal astrocytoma, oligodendroglioma, brain metastasis, brain abscess.

[0026] The object of the present invention is to provide a computer program product having a computer program / instruction thereon, wherein the computer program / instruction is executed by a processor to implement the above-mentioned brain image processing method based on 3D Slicer.

[0027] The object of the present invention is to provide a computer program product having a computer program / instruction thereon, wherein the computer program / instruction is executed by a processor to implement the above-mentioned method for locating intracranial tumors based on intraoperative neuronavigation.

[0028] The object of the present invention is to provide a computer device having a memory or a processor and a computer program / instruction stored in the memory, wherein the computer program / instruction is executed by the processor to implement the above-mentioned brain image processing method based on 3D Slicer.

[0029] The object of the present invention is to provide a computer device having a memory or a processor and a computer program / instruction stored in the memory, wherein the computer program / instruction is executed by the processor to implement the above-mentioned method for locating intracranial tumors based on intraoperative neuronavigation.

[0030] The object of the present invention is to provide a computer-readable storage medium having a computer program / instruction stored thereon, wherein the computer program / instruction is executed by a processor to implement the above-mentioned brain image processing method based on 3D Slicer.

[0031] The object of the present invention is to provide a computer-readable storage medium having a computer program / instruction stored thereon, wherein the computer program / instruction is executed by a processor to implement the above-mentioned method for locating intracranial tumors based on intraoperative neuronavigation.

[0032] Advantages of the present invention:

[0033] 1. Combining the advantages of the 3D medical model reconstructed by 3Dslicer and other software with the intraoperative navigation and positioning function of the existing neuronavigation system, the navigation range of the existing neuronavigation system is expanded and the accuracy of intraoperative navigation and positioning is improved. In particular, for the facial nerve that was previously difficult to reconstruct and locate, it can be accurately identified and located, reducing the probability of intraoperative damage to the facial nerve and improving the functional retention rate of the facial nerve.

[0034] 2. In the process of using 3Dslicer software to generate the Labelmap file of the three-dimensional model of the surgery-related structure, a screening principle for the image sequence (or carrier sequence) based on the three-dimensional model was proposed. The sequence used as the central coordinate or the image sequence that has completed the alignment and updated the coordinates was screened to ensure that the target model is in the correct spatial position in the corresponding image sequence and is in the correct positional relationship with the surrounding anatomical structure; and the final generated DICOM data is easier to merge with the original image data or other image sequences after being imported into the navigation system.

[0035] 3. Select an image sequence whose scanning range covers the entire brain, so that the generated DICOM file can be imported into the navigation system and smoothly integrated with other image data.

[0036] 4. Screening the image sequence with the thinnest scanning layer thickness or the smallest voxel is used to improve the image quality of the target structure obtained after segmenting the target structure in the neuronavigation system, which can make the three-dimensional image generated by segmentation smoother, continuous, clearer and more beautiful from a morphological perspective. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A schematic flow chart of a brain image processing method based on 3D Slicer provided in an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a brain image processing system based on 3D Slicer provided in an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of a brain image processing device based on 3D Slicer provided in an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of marking anatomical structures with different colors provided in an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of converting the model provided in the embodiment of the present invention into a Labelmap file;

[0043] Figure 6 A schematic diagram of converting a Labelmap file provided in an embodiment of the present invention into a DICOM file;

[0044] Figure 7 Provide DICOM data corresponding to the brain and various anatomical structures fused in a neuronavigation system for embodiments of the present invention;

[0045] Figure 8 The neuronavigation system provided by the embodiment of the present invention displays the segmented images of various anatomical structures, wherein the facial nerve image is clear and coherent;

[0046] Fig. 9 The facial nerve image segmented by the carrier image sequence provided by the embodiment of the present invention without selecting the thinnest slice thickness or the smallest voxel is shown as an intermittent and discontinuous image;

[0047] Fig.10 The brain model provided in the embodiment of the present invention does not select the bearing sequence as the central coordinate, does not perform the whole brain range screening and does not perform the thickness screening, and the three-dimensional spatial coordinates of the facial nerve image segmented are inaccurate, and the spatial position relationship with the surrounding structures is inaccurate;

[0048] Fig.11 The three-dimensional model and loading sequence of other types of intracranial tumors provided in the embodiments of the present invention. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0050] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S101, S102, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0051] Figure 1 A schematic diagram of a brain image processing method based on 3D Slicer provided in an embodiment of the present invention specifically includes:

[0052] S1: Obtain the patient's brain imaging sequence data;

[0053] In one embodiment, the brain image sequence data includes one or more of the following: MRI, CT or DSA.

[0054] S2: The image sequence data is registered by 3D Slicer to obtain a registered image sequence;

[0055] In one embodiment, 3D Slicer is an open source software platform mainly used for medical image informatics, image processing and 3D visualization. The main functions include:

[0056] 1. Medical image viewing and analysis: 3D Slicer supports a variety of medical imaging data formats, such as CT, MRI, PET, etc. Users can load these data and browse, compare and analyze them.

[0057] 2. 3D reconstruction and visualization: The ability to convert 2D medical image data into 3D models is critical for understanding anatomical structures, pathological changes (such as tumors), and surgical planning.

[0058] 3. Image registration: Allows users to align images from different sources or time points, which is very useful for tracking disease progression or comparing changes before and after treatment.

[0059] 4. Image Segmentation and Annotation: Provides powerful tools to manually or automatically segment structures in images, which is critical for quantitative analysis (such as measuring organ volume). It also supports automatic segmentation using extensions such as Total Segmentator and MONAI Label.

[0060] In one embodiment, the registration is to register the image structure between the image coordinate system of the image sequence data and the coordinate system of the three-dimensional reconstructed model.

[0061] Before three-dimensional reconstruction of the medical model, various image sequences are unified in one coordinate system to ensure that the coordinate system of the finally selected image sequence (carrier sequence) and the coordinate system of the reconstructed three-dimensional model are in the same coordinate system.

[0062] In one embodiment, the brain model is composed of surgery-related anatomical structure models obtained by three-dimensionally reconstructing various anatomical structures related to brain and intracranial surgery using software such as Slicer, and each anatomical structure model is marked with a different color.

[0063] In one embodiment, the anatomical structures related to the brain and intracranial surgery include one or more of the following: brain stem, facial nerve, trigeminal nerve, skull, and vein.

[0064] Optionally, the surgery-related anatomical structures include tumors, brain stem, cranial nerves, skull, and veins.

[0065] S3: performing three-dimensional reconstruction based on the registered sequence to obtain a three-dimensional brain model;

[0066] In one embodiment, 3D reconstruction is a technique that aims to recover and create a 3D representation of a target object or environment from a series of 2D images, sensor data, or scanned data. This process is essential for understanding, simulating, analyzing, and visualizing real-world objects and scenes. 3D reconstruction technology is widely used in many fields such as medical imaging and engineering design.

[0067] 3D reconstruction usually involves the following key steps:

[0068] 1. Data collection: Scanning the internal structure of the human body through CT or MRI, etc.

[0069] 2. Feature extraction and matching: Identify and match key feature points in the collected 2D images. Features can be edges, corners, textures, etc., through which the relative positions between different views can be calculated.

[0070] 3. 3D space reconstruction: Use the matched feature points to calculate the position and attitude of the camera, and reconstruct the 3D point cloud through triangulation or multi-view geometry. The point cloud is a discrete representation of the object surface in 3D space.

[0071] 4. Surface generation and refinement: Converting point cloud data into a surface model can be done through a variety of algorithms, such as surface fitting, surface reconstruction algorithms (such as Poisson reconstruction, Marching Cubes, etc.), making the model smoother and more continuous.

[0072] 5. Texture mapping: Map the texture information of the original image to the reconstructed 3D model.

[0073] 6. Post-processing and optimization: including model cleaning, hole filling, model simplification, lighting and material adjustment, etc., to improve the quality and practicality of the final model.

[0074] S4: Select the sequence used as the center coordinate during registration or the image sequence after registration has been completed and the coordinates have been updated;

[0075] In one embodiment, the corresponding image sequences (carrying sequences) are screened for the three-dimensional models of the brain, tumor, etc., and the sequences used as the center coordinates during registration or the image sequences that have completed registration and updated the coordinates are selected.

[0076] S5: Filtering out the image sequence whose scanning range includes the whole brain sequence from the sequence used as the center coordinate during the registration or the image sequence after the registration is completed and the coordinates are updated to obtain the filtered image sequence;

[0077] In one embodiment, the screened image sequence is a carrier sequence.

[0078] In one embodiment, S5 further includes performing thickness screening on the screened sequence, screening the sequence with the thinnest scanning layer or the smallest voxel to obtain a thickness screened image sequence, and converting the three-dimensional brain model into a Labelmap file based on the thickness screened image sequence.

[0079] To further illustrate, the S5 further includes further screening the screened sequences according to the scan layer thickness, selecting the sequence with the thinnest scan layer or the smallest voxel as the thickness bearing image sequence corresponding to the three-dimensional model, and generating a Labelmap file through format conversion.

[0080] In one embodiment, based on the model of the anatomical structure related to the surgery, such as the tumor and cranial nerve, steps S4-S5 are performed in sequence to obtain the image sequence (carrier sequence) based on each structure model, and the Labelmap file of each anatomical structure is obtained by conversion through 3DSlicer software, and then converted into a DICOM format file and exported.

[0081] In a specific embodiment, in 3D Slicer, "Model to LabelMap" is a conversion process that converts a 3D model (Model) into a label map (LabelMap). This function is very useful in certain specific application scenarios, especially when it is necessary to integrate the geometric shape information of the structure with the image data.

[0082] In one embodiment, in 3D Slicer, the user can perform the “Model to LabelMap” conversion by following the steps below:

[0083] First, load a 3D model (such as brain, cranial nerves, etc.) into 3D Slicer.

[0084] Select the model, find and start the "Model to LabelMap" module in the "Modules" menu.

[0085] In the module interface, select the corresponding image sequence (carrier sequence) for the 3D model in the "Input volume" option according to steps S4 and S5, select the target model to be loaded and converted in the "Model" option, name the generated Labelmap file in the "Output Volume", and click "Apply" or the corresponding button to start the conversion process.

[0086] After the conversion is completed, the newly generated label map will appear as a new data node in the data browsing library of 3D Slicer and can be used for subsequent analysis or visualization.

[0087] In a specific embodiment, the facial nerve and brainstem model reconstructed by 3DSlicer software is converted using the "Model To Label Map" module in the 3DSlicer software, wherein an image sequence should be selected in the "Input Volume" option in the "Model To Label Map" module according to the following principles: (1) selecting a sequence that is used as the center coordinate when aligning image data during model making or a sequence that has completed alignment to ensure that the model coordinates are consistent with the coordinates of the sequence; (2) selecting a sequence whose scanning range includes the entire brain; (3) selecting a sequence with the thinnest scanning layer thickness or the smallest voxel.

[0088] The above selection criteria are summarized after testing all sequences one by one and comparing them repeatedly. Failure to follow the above principles may affect the quality of the generated Labelmap file.

[0089] In a specific embodiment, the present invention can also convert the Model into a Labelmap file through a "Mesh ToLabel Map" module. By comparison, it is found that the operation of this module is relatively complicated and the image sequence carrying the Labelmap cannot be selected, which brings many difficulties to the subsequent DICOM data processing. Therefore, the present invention is the best method to convert a three-dimensional Model into a Labelmap file through a "Model To Label Map" module.

[0090] S6: The three-dimensional brain model is converted into a Labelmap file based on the screened image sequence.

[0091] In one embodiment, before the three-dimensional model is converted into a format, an image sequence is first selected for each three-dimensional model, and different models can eventually select the same sequence as a carrier sequence.

[0092] In one embodiment, after acquiring the patient's brain image sequence data, the image coordinates of the image sequence are registered with the model coordinates of the three-dimensional model through 3D Slicer software to obtain a registered image sequence, so that the image coordinates and the left side of the model have a unified coordinate transformation relationship, wherein the image sequence with the largest scanning range or any image sequence in the image sequence is selected as the central coordinate sequence, the central coordinates are registered with the model coordinates to obtain a coordinate transformation relationship, and the remaining image sequences are registered based on the coordinate transformation relationship to obtain the registered image sequence;

[0093] Further, a three-dimensional brain model is obtained by performing three-dimensional reconstruction in a model coordinate system based on the registered image sequence, wherein the three-dimensional brain model includes various anatomical structures related to cranial surgery;

[0094] At the same time, the sequence used as the center coordinate during registration or the image sequence after registration and coordinate update is selected from the image sequence, and the sequence whose scanning range includes the whole brain is further selected to obtain the selected sequence image;

[0095] Finally, the filtered sequence images are used as the carrier sequence of the three-dimensional model and exported with the three-dimensional brain model to be converted into a labelmap file;

[0096] Among them, the sequence of scanning the whole brain also includes further screening, and the sequence with the thinnest layer or the smallest voxel is selected as the filtered sequence image;

[0097] Among them, each anatomical structure in the three-dimensional brain model is screened as above to obtain the bearing file of each anatomical structure.

[0098] In one embodiment, the three-dimensional brain model is converted into a Labelmap file based on the load sequence output, and the load sequence (the image sequence after the image coordinates and the model coordinates are registered) is screened by any of the following methods:

[0099] First, select the sequence used as the center coordinate during registration or the image sequence after registration and coordinate update, further select the sequence whose scanning range includes the whole brain, and then further select the sequence with the thinnest layer or the smallest voxel as the selected sequence image;

[0100] First, select the sequence used as the center coordinate during registration or the image sequence after registration and coordinate update, and further select the sequence with the thinnest layer or the smallest voxel as the selected sequence image; then further select the sequence whose scanning range includes the whole brain;

[0101] First, the sequences whose scanning range includes the whole brain are selected, and then the sequences used as the center coordinates during registration or the image sequences after registration and coordinate update are selected, and then the sequences with the thinnest layers or the smallest voxels are further selected as the selected sequence images;

[0102] First, the sequences whose scanning range includes the whole brain are screened, and then the sequences with the thinnest layers or the smallest voxels are further screened as the screened sequence images, and then the sequences used as the central coordinates during the registration or the image sequences after the registration is completed and the coordinates are updated are further selected;

[0103] First, the sequence with the thinnest layer or the smallest voxel is selected as the selected sequence image, and then the sequence used as the center coordinate during registration or the image sequence after the registration is completed and the coordinates are updated is selected, and then the sequence whose scanning range includes the whole brain is further selected;

[0104] First, the sequence with the thinnest layer or the smallest voxel is selected as the selected sequence image, and then the sequence whose scanning range includes the whole brain is further selected, and then the sequence used as the central coordinate during the alignment or the image sequence after the alignment is completed and the coordinates are updated is further selected.

[0105] In a specific embodiment, the three-dimensional models of the surgically relevant anatomical structures are converted into DICOM files and exported, such as Figure 4 As shown, each model is displayed in a different color. In the 3D Slicer interface, select "models" in the "Modules" module, and the structural models represented by different colors will be displayed in the drop-down list.

[0106] In a specific embodiment, the facial nerve and brainstem models reconstructed by 3DSlicer software are converted into corresponding Labelmap files by using the "Model To Label Map" module in the 3DSlicer software. Figure 5shown.

[0107] In a specific embodiment, select "Data" in the "Modules" module in the 3D Slicer interface, select the newly generated Labelmap file in the drop-down list of the "Subject hierarchy" option, right-click to select the Labelmap file, right-click to select the "Export to DICOM" module to obtain the DICOM file, such as Figure 6 shown.

[0108] In one embodiment, clearer and more specific image data is obtained by screening sequences and three-dimensional modeling using the method of the present invention, such as Figure 7 , Figure 8 As shown, it is helpful for subsequent intraoperative navigation surgery, provides basic conditions for preoperative surgical planning, reduces surgical risks, and has good clinical application value.

[0109] The embodiments of the present disclosure further provide a computer program product or system, including a computer program, which implements the steps of the above-mentioned brain image processing method based on 3D Slicer when executed by a processor.

[0110] Figure 2 A schematic diagram of a brain image processing system based on 3D Slicer provided in an embodiment of the present invention specifically includes:

[0111] Acquisition module: acquires the patient's brain imaging sequence data;

[0112] Registration module: the image sequence data is registered through 3D Slicer to obtain a registered image sequence;

[0113] Reconstruction module: performing three-dimensional reconstruction based on the registered sequence to obtain a three-dimensional brain model;

[0114] The first screening module: selects the sequence used as the center coordinate during registration or the image sequence after registration has been completed and the coordinates have been updated;

[0115] A second screening module: screening out a sequence whose scanning range includes the whole brain from the sequence used as the center coordinate during the registration or the image sequence after the registration is completed and the coordinates are updated to obtain a screened image sequence;

[0116] Conversion module: The three-dimensional brain model is converted into a Labelmap file based on the screened image sequence.

[0117] Figure 3 A schematic diagram of a brain image processing device based on 3D Slicer provided in an embodiment of the present invention specifically includes:

[0118] A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned brain image processing methods based on 3D Slicer is executed.

[0119] The disclosed embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it is any one of the above-mentioned brain image processing methods based on 3D Slicer.

[0120] The embodiment of the present invention further provides a method for locating an intracranial tumor based on intraoperative neuronavigation, comprising:

[0121] The method for locating intracranial tumors based on intraoperative neuronavigation performs image processing by the brain image processing method based on 3D Slicer to obtain a Labelmap file of the intracranial tumor sequence; wherein,

[0122] Optionally, the brain model reconstructed in the three-dimensional reconstruction based on the brain image processing method of 3D Slicer also includes a tumor and a facial nerve;

[0123] Convert the Labelmap file into DICOM data;

[0124] The intraoperative neuronavigation system acquires the DICOM data and the intraoperative brain imaging data of the patient,

[0125] fusing the DICOM data with the patient's intraoperative brain data according to the navigation system operation steps to obtain fused data;

[0126] Segmenting the target structure in the fused data to obtain segmentation data of surgery-related structures and saving the segmentation data;

[0127] The tumor or nerve is localized by a navigation system based on the segmentation data.

[0128] In one embodiment, the method further includes data labeling, labeling the converted data to obtain a labeled data set; and intraoperative neuronavigation to obtain the labeled data and the patient's brain imaging data.

[0129] In one embodiment, the tumor in the method includes one or more of the following: acoustic neuroma, trigeminal schwannoma, astrocytoma, oligodendroglioma, ependymoma, meningioma, metastasis, lymphoma, brain abscess, pituitary adenoma, suprasellar meningioma, craniopharyngioma, astrocytoma, epidermoid cyst, arachnoid cyst, and aneurysm.

[0130] In one embodiment, acoustic neuroma is the most common benign tumor in the posterior cranial fossa, accounting for about 80%-90% of tumors in the cerebellopontine angle area and about 8%-10% of intracranial tumors. It often causes symptoms such as tinnitus, hearing loss, and dizziness on the affected side. As the tumor grows in size, it compresses surrounding structures such as the brainstem and trigeminal nerve, which can cause symptoms such as headache, dizziness, and facial numbness. Surgical resection is the main treatment for acoustic neuroma. Although great progress has been made in neurosurgery techniques and surgical auxiliary means, the postoperative facial nerve function retention rate of acoustic neuroma patients is still not very ideal.

[0131] In a specific embodiment, the software is used to convert the three-dimensional image format (such as STereoLithography, stl format) of the reconstructed facial nerve and other structures, and finally the file of the three-dimensional model image format is converted into a file of DICOM format that can be imported into the neuronavigation system, and the relevant image information is edited (such as name, image number, etc.) so that the image information of the above file is consistent with the image information of the patient's MRI and CT. The converted file is imported into the neuronavigation system together with the MRI and CT image files, and then the steps used in neuronavigation are followed, such as fusion, segmentation, etc. to complete the surgical plan, and finally neuronavigation is used to accurately locate the position and course of the facial nerve during surgery to improve the functional retention rate of the facial nerve after acoustic neuroma surgery. Specifically, the following steps are included:

[0132] Step S1: The facial nerve and brainstem model reconstructed by 3DSlicer software is converted into a corresponding Labelmap file using the "Model To Label Map" module in the 3DSlicer software.

[0133] Step S21: Use the "Export to DICOM" function option in the "Data" module to create a corresponding DICOM file using the Labelmap file, and input the image information such as "name" and "image number" to make it consistent with the original MRI and CT image information of the patient, and finally export and label the DICOM file name according to the anatomical structure. Among them, the information of name (including uppercase and lowercase), gender, date of birth, and image ID number needs to be consistent.

[0134] Step S22: Import the DICOM file with the marked name and the patient's original CT and MRI data in DICOM format into the same folder.

[0135] Step S3: Import all the DICOM format files into the neuronavigation system, perform fusion, segmentation and other processing in sequence, display the anatomical structures such as the tumor, facial nerve, brainstem, etc. in the neuronavigation system, and finally complete and save the surgical plan.

[0136] Step S4: The surgical plan is imported into the intraoperative display screen of the neuronavigation to display structures such as the tumor and the facial nerve. After the fusion is completed before the operation, the facial nerve is positioned and protected during the operation.

[0137] In a specific embodiment, the present invention screens the sequences with central coordinates during registration or the sequences that have completed registration, and then screens the sequences whose scanning range includes the whole brain, and then based on the above screening, screens the sequences with the thinnest scanning layer or the smallest voxels as the bearing sequences. It can be found from the figure that the intracranial tumors and brain structure tissues are displayed clearly and continuously, with a good intraoperative navigation and positioning field of view.

[0138] In a specific embodiment, the present invention establishes a target model by screening sequences with central coordinates during registration or sequences that have completed registration and then screening sequences whose scanning range includes the whole brain, and does not select the sequence with the thinnest scanning layer or the smallest voxel as the bearing sequence, such as Fig. 9 As shown, the target model (right facial nerve) is imported into the navigation system after format conversion. It can be seen that the image of the right facial nerve after segmentation is intermittent, not smooth, and of low quality.

[0139] In a specific embodiment, a target model that does not perform sequence screening is Fig.10 As shown, after the target model was converted into a new format and imported into the navigation system, it was very difficult to fuse it with the original image data, resulting in fusion errors and ultimately incorrect spatial position of the right facial nerve.

[0140] In a specific embodiment, the present invention realizes the real-time positioning of the facial nerve during surgery by a neuronavigation system, which helps neurosurgeons to protect the facial nerve during surgery, and is beneficial to improving the functional retention rate of the facial nerve after acoustic neuroma surgery.

[0141] In a specific embodiment, the image processing method of the present invention perfectly presents other intracranial tumor (functional area convex meningioma) surgery-related structures in the navigation system and realizes accurate intraoperative navigation positioning, such as Fig.11 shown.

[0142] The present invention can also be used for intraoperative navigation of functional area tumors, helping neurosurgeons to accurately identify important structures such as the precentral gyrus, central sulcus, and central sulcus vein of the brain, which is beneficial to the protection of the above-mentioned important structures during surgery and reduces the occurrence of postoperative neurological dysfunction.

[0143] In a specific embodiment, the present invention realizes the combination of medical modeling software such as 3DSlicer with the existing neuronavigation system, giving full play to the functional advantages of each, jointly assisting the surgical resection of intracranial tumors and the protection of important functional structures, and improving the quality of neurosurgery.

[0144] In one embodiment, the present invention indirectly expands the types of image data that can be imported into the neuronavigation system, improves the functionality of the existing navigation system without adding additional economic costs, and expands the application scope of the neuronavigation system in the field of neurosurgery, making it easier to promote in a large number of hospitals.

[0145] In a specific embodiment, the clinical case:

[0146] Patient: Ms. XX, female, years old, with left acoustic neuroma, from the Second Hospital of Hebei Medical University.

[0147] (1) Data acquisition and model reconstruction: The original data of thin-slice temporal bone CT, thin-slice MRI T1-weighted imaging, diffusion tensor imaging and other sequences were collected and imported into the 3DSlicer software. The “Segment Editor” module was used to create three-dimensional model images of structures such as tumors, veins, skull, brainstem, and trigeminal nerves. The function options in the “Diffusion” module were used to create a model image of the facial nerve.

[0148] (2) Use the "Model To Label Map" module in the 3DSlicer software to create label maps of the corresponding structures of the tumor, vein, skull, facial nerve and other structures.

[0149] (3) Use the "Export to DICOM" function option in the "Data" module of the 3DSlicer software to create a DICOM file of the corresponding structure, and enter the image information such as "name" and "image number" in the file to make it consistent with the original MRI and CT image information of the patient. Finally, export and label the DICOM file name according to the anatomical structure. Put the newly created DICOM file and the original MRI and CT data into the same folder.

[0150] (4) All the data in the folder were imported into the neuronavigation system (Brainlab neuronavigation), and all the DICOM data were fused one by one according to the steps. After fusion, the Segmentation function option in the neuronavigation system was used to reconstruct and mark the tumor, skull, facial nerve and other structures with different colors. After the surgical plan was completed, it was saved.

[0151] (5) The saved plan is imported into the display of the neuronavigation system to present images of the tumor, skull, facial nerve, and brainstem. The facial nerve is located during surgery after infrared navigation fusion registration and electrophysiological monitoring is used for confirmation under a microscope to further enhance the protection of the facial nerve during surgery and achieve intraoperative navigation and localization of the facial nerve for acoustic neuroma.

[0152] An embodiment of the present invention further provides a computer program product having a computer program / instruction thereon, wherein the computer program / instruction is executed by a processor to implement the method for locating intracranial tumors based on intraoperative neuronavigation.

[0153] The disclosed embodiment of the present invention also provides a computer device having a memory or a processor and a computer program / instructions stored in the memory, wherein the computer program / instructions are executed by the processor to implement the above-mentioned method for locating intracranial tumors based on intraoperative neuronavigation.

[0154] The disclosed embodiments of the present invention further provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, is any one of the above-mentioned methods for locating intracranial tumors based on intraoperative neuronavigation.

[0155] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of the method relative to the default setting. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units. A person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above-mentioned embodiments may be completed by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0156] A person skilled in the art will appreciate that all or part of the steps in the above-mentioned embodiment can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer.

[0157] In the readable storage medium, the above-mentioned medium storage can be a read-only memory, a magnetic disk or an optical disk, etc. The above is a detailed introduction to a computer device provided by the present invention. For those skilled in the art, according to the idea of ​​the embodiment of the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A brain image processing method based on 3D Slicer, characterized in that: include: S1. Obtaining patient brain imaging sequence data; S2, the image sequence data is registered by 3D Slicer to obtain a registered image sequence; S3, performing three-dimensional reconstruction based on the registered sequence to obtain a three-dimensional brain model; S4, selecting a sequence used as the center coordinates during registration or an image sequence after registration has been completed and the coordinates have been updated; S5, selecting an image sequence whose scanning range includes a whole-brain sequence from the sequence used as the center coordinate during the registration or the image sequence after the registration is completed and the coordinates are updated to obtain the selected image sequence; S6. The three-dimensional brain model is converted into a Labelmap file based on the screened image sequence.

2. The brain image processing method based on 3D Slicer according to claim 1, characterized in that: The S5 further includes performing thickness screening on the screened sequence, screening the sequence with the thinnest scanning layer or the smallest voxel to obtain an image sequence after thickness screening, and converting the three-dimensional brain model into a Labelmap file based on the image sequence after thickness screening.

3. The brain image processing method based on 3D Slicer according to claim 1, characterized in that: The registration is the registration of each image sequence with the image sequence as the center coordinate before modeling; Optionally, the registration is to unify the coordinates of all image sequences before establishing the brain model, so as to ensure that the coordinate system of the image sequence data is consistent with the coordinate system of the three-dimensional reconstruction model of the brain.

4. The brain image processing method based on 3D Slicer according to claim 1, characterized in that: The brain model is composed of surgery-related anatomical structure models obtained by three-dimensionally reconstructing various anatomical structures related to brain and intracranial surgery, and each anatomical structure model is marked with a different color; Optionally, the anatomical structures related to brain and intracranial surgery include one or more of the following: brain stem, facial nerve, trigeminal nerve, skull, vein; Optionally, the brain imaging data includes one or more of the following: MRI, CT or DSA; Optionally, based on the anatomical structure models related to the brain and intracranial surgery, steps S4-S5 are sequentially performed to obtain image sequences after screening of the anatomical structure models, and each anatomical structure model is converted into a Labelmap file based on the image sequences after screening of the anatomical structures.

5. A method for locating intracranial tumors based on intraoperative neuronavigation, characterized in that: include: The method performs image processing based on the brain image processing method based on 3D Slicer described in claims 1-4 to obtain a Labelmap file of anatomical structures related to intracranial tumor surgery; wherein the brain model three-dimensionally reconstructed by the method based on the brain image processing method based on 3D Slicer also includes a tumor, an artery, and a facial nerve; Convert the Labelmap file into DICOM data; The intraoperative neuronavigation system acquires the DICOM data and the original MRI and CT image data of the patient. The DICOM data and the original image data of the patient are sequentially fused to obtain fused data; Segmenting the target structure in the fused data to obtain segmentation data of surgery-related structures; Tumors or nerves are localized based on the segmentation data.

6. The method for locating intracranial tumors based on intraoperative neuronavigation according to claim 5, characterized in that: The method further comprises data labeling, labeling the DICOM data to obtain a labeled data set; and obtaining the labeled data and the patient's brain image data by intraoperative neuronavigation.

7. The method for locating intracranial tumors based on intraoperative neuronavigation according to claim 5, characterized in that: The tumors in the method include one or more of the following: acoustic neuroma, trigeminal schwannoma, parasinus falx, convex or functional area meningioma and lymphoma, brain parenchymal astrocytoma, oligodendroglioma, brain metastasis, brain abscess.

8. A computer program product having a computer program / instructions thereon, characterized in that: The computer program / instructions are executed by the processor to implement any one of the brain image processing methods based on 3DSlicer described in claims 1-4, or to implement the method for locating intracranial tumors based on intraoperative neuronavigation described in claims 5-7.

9. A computer device having a memory or a processor and a computer program / instruction stored in the memory, characterized in that: The computer program / instructions are executed by the processor to implement any one of the brain image processing methods based on 3DSlicer described in claims 1-4, or to implement the method for locating intracranial tumors based on intraoperative neuronavigation described in claims 5-7.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: The computer program / instructions are executed by a processor to implement any one of the brain image processing methods based on 3D Slicer described in claims 1-4, or to implement the method for locating intracranial tumors based on intraoperative neuronavigation described in claims 5-7.

Images