Method for determining target decompression area of ​​bone structure, method for automatically planning decompression path of bone structure, electronic device and storage medium

Abstract

The present invention discloses a method for determining a target decompression area of ​​a bone structure, a method for automatically planning a decompression path of a bone structure, an electronic device, and a storage medium. The method accurately generates the soft tissue structure in the MRI image in the bony channel of the CT image through registration and fusion of CT images and MRI images, and then analyzes the position and morphology of the soft tissue structure in the bony channel based on the registered and fused images to determine whether there is a lesion, thereby realizing automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the prior art, the method reduces the problems of missed judgment and wrong judgment caused by manual positioning, lays a solid foundation for accurate planning of the decompression path, can improve the success rate of surgery, reduce the rate of postoperative revisions, and enhance the patient experience.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method for determining a target decompression area of ​​a bone structure, a method for automatically planning a bone structure decompression path, an electronic device, and a storage medium. Background Art

[0002] According to statistics, degenerative spinal diseases, such as lumbar disc herniation and spinal stenosis, have a global prevalence of 5.7%, making them the leading cause of global productivity loss and costing the economy hundreds of billions of dollars annually. Surgery is currently the primary treatment, with internal fixation and decompression techniques being the primary procedures. Surgeons typically use MRI (Magnetic Resonance Imaging) or X-rays to determine the type and location of the disc herniation and the factors causing spinal stenosis before surgery, and then plan their surgical plan based on this information.

[0003] The design of the surgical plan is not only related to the disease type but also to the surgeon's familiarity with the surgical procedure. While traditional open surgical approaches offer wide exposure and a good visual field, they can also cause significant damage to ligaments and muscles, potentially compromising postoperative outcomes. In recent years, minimally invasive techniques, particularly spinal endoscopy, have rapidly advanced. Compared with other minimally invasive spinal procedures or open surgery, these procedures, through the use of highly efficient optical equipment and precise surgical techniques, can reduce incision size, bone and soft tissue damage, and postoperative pain, enabling earlier return to normal activities and work. Furthermore, they preserve tissue structure as much as possible, effectively minimizing instability at the decompressed level and the risk of adjacent segment degeneration. Nevertheless, to ensure the accuracy and safety of minimally invasive spinal endoscopy, surgeons often rely on intraoperative fluoroscopy for channel positioning and minimally invasive manipulation, resulting in significant intraoperative radiation exposure, which poses significant risks to both patients and physicians. Furthermore, spinal endoscopy is not suitable for procedures involving long segments or spinal deformities.

[0004] Although computer-assisted navigation and robotics have been widely used in spinal surgery, their auxiliary value in decompression technology has not been fully demonstrated. Both are primarily used to guide the decompression channel, but neither addresses the scope of bony grinding and cutting, nor the decompression path. Without a good preoperative decompression plan, surgeons rely primarily on endoscopic images to manually determine whether decompression is adequate and whether internal fixation is necessary, leading to complications such as the need for revision surgery in a large proportion of spinal endoscopic surgeries.

[0005] Furthermore, existing techniques primarily determine the type and location of herniated discs, as well as the factors contributing to spinal stenosis, through manual observation of CT or MRI images. This heavily relies on the surgeon's experience, leading to significant variations in the location and number of lesions determined by different surgeons. This results in low accuracy in determining the target decompression zone. This can lead to inaccurate or incomplete coverage of lesions during subsequent decompression path planning, resulting in a high rate of postoperative revisions and an increased risk of secondary surgery or complications.

[0006] To facilitate understanding of the technical solution provided by this embodiment, several medical terms in the prior art are explained as follows:

[0007] 1. Herniated disc: The soft tissue structure between the vertebrae of the spine, the intervertebral disc, is responsible for spinal mobility. While this disc increases spinal flexibility, allowing us to perform large curvature movements like bending and arching our backs, it also creates a problem: a herniated disc.

[0008] The intervertebral disc is composed of an outer annulus fibrosus and a central nucleus pulposus. The nucleus pulposus is an elastic, milky white, translucent gelatinous body located between the two cartilage plates and the annulus fibrosus. It is an elastic jelly substance composed of a crisscrossing fibrous network structure, namely chondrocytes and a proteoglycan mucus-like matrix. Due to external forces or degenerative diseases, the intervertebral disc gradually protrudes. When the soft tissue structures in the spine (especially the nucleus pulposus) rupture the annulus fibrosus of the intervertebral disc under the influence of external forces or degenerative diseases, the nucleus pulposus tissue protrudes from the ruptured area into the spinal canal, causing the disc to protrude. The most serious problem of intervertebral disc protrusion is that the exposed nucleus pulposus compresses the spinal cord and spinal nerves.

[0009] 2. Spinal stenosis: Mainly caused by spinal degeneration:

[0010] (1) Herniated disc: When a herniated disc occurs, the protruding disc takes up space in the spinal canal, leading to spinal stenosis.

[0011] (2) Hypertrophy of the ligamentum flavum and posterior longitudinal ligament: When the spine degenerates and becomes unstable, the stress on the ligamentum flavum and posterior longitudinal ligament increases, often leading to their degeneration or rupture. Long-term damage and repair processes will inevitably cause the ligamentum flavum and posterior longitudinal ligament to thicken, and the space in the spinal canal will become smaller, leading to spinal stenosis.

[0012] (3) Osteophytes: When the spine degenerates, the cartilage on the joint surface wears away, causing friction between bones, which results in the formation of osteophytes. These osteophytes occupy space in the spinal canal or intervertebral foramen, leading to spinal stenosis. Summary of the Invention

[0013] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide a method for determining a target decompression area of ​​a bone structure, a method for automatically planning a bone structure decompression path, an electronic device and a storage medium, so as to solve the problem in the related art that manual determination of the target decompression area results in low accuracy of decompression path planning.

[0014] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0015] According to a first aspect of the present invention, a method for determining a target decompression area of ​​a bone structure is provided, comprising:

[0016] Obtaining a CT image of the vertebral structure within a preset marked area, and an MRI image of the vertebral structure corresponding to the preset marked area, wherein the CT image carries substructure information of the vertebral structure and information about bony channels formed between adjacent single vertebral structures;

[0017] registering the CT image with the MRI image, extracting a soft tissue structure from the MRI image based on a boundary position of the bony channel, and generating the soft tissue structure within the bony channel;

[0018] The morphological features of each soft tissue structure are extracted, and whether the soft tissue structure has a lesion is determined based on the morphological features, and the area where the lesion occurs is determined as the target decompression area.

[0019] Preferably, the bony channel information includes at least the spinal canal, intervertebral foramen, and intervertebral disc area, and the soft tissue structure is extracted from the MRI image and generated in the bony channel, including:

[0020] The MRI image within the spinal canal and nerve root range is cropped out, and the grayscale of the soft tissue structure in the cropped MRI image is threshold analyzed to extract the soft tissue structure, including: neural structure, yellow ligament, and possible herniated intervertebral disc;

[0021] The MRI image within the intervertebral disc area is cropped out, and the grayscale of the soft tissue structure in the cropped MRI image is threshold analyzed to extract the soft tissue structure, including the intervertebral disc and the possible protruding nucleus pulposus;

[0022] Generating the neural structure in the spinal canal, wherein the nerve roots branching from the spinal cord exit through the intervertebral foramina on both sides of the spinal canal;

[0023] Generating the yellow ligament in the spinal canal, the yellow ligament runs from the lower edge and inner surface of the upper lumbar vertebra to the upper edge and outer edge of the lower lumbar vertebra, participating in forming the posterior wall and posterolateral wall of the spinal canal, and gradually thickening from top to bottom;

[0024] The intervertebral disc and nucleus pulposus are generated in the intervertebral disc area.

[0025] Preferably, the morphological characteristics include at least size and position. The determining whether the soft tissue structure has a lesion based on the morphological characteristics and determining the area with the lesion as the target decompression area includes:

[0026] If the position of the intervertebral disc exceeds the preset position and / or the size exceeds the preset size, it is determined that the intervertebral disc is herniated, and the herniated intervertebral disc is determined as a target decompression area;

[0027] If the position of the ligamentum flavum exceeds the preset position and / or the size exceeds the preset size, it is determined that the ligamentum flavum is hypertrophic, and the hypertrophic ligamentum flavum is determined as the target decompression area;

[0028] The dimensions include: length, width, height, and volume converted from the length, width, and height.

[0029] Preferably, the method further comprises:

[0030] According to the characteristic data of each sub-molecule structure, it is determined whether there is osteophyte on each sub-molecule structure;

[0031] If it is determined that osteophytes exist, extracting characteristic data of the osteophytes, the characteristic data of the osteophytes including at least: position and size information of the osteophytes;

[0032] Based on the position and size information of the osteophyte, it is determined whether the osteophyte intersects with the adjacent neural structure. If so, the area of ​​the intersection region between the osteophyte and the neural structure is calculated, and the intersection region is determined as the target decompression area.

[0033] According to a second aspect of the present invention, a method for automatically planning a bone structure decompression path is provided, comprising:

[0034] Determine the target decompression area according to the above method, and determine the target decompression area as the target area;

[0035] Determine the indication for bone structure based on the location of the target area;

[0036] Determine the entry area of ​​the decompression pathway based on the indication, the location of the target area, and the morphological characteristics of the substructures surrounding the target area;

[0037] The surgical procedure and decompression path are determined based on the target area and approach area.

[0038] Preferably, if the bone structure is a lumbar vertebra, determining the indication of the bone structure according to the location of the target area includes:

[0039] If the target area is located in the center of the posterior part of the spinal canal, the indication for determining the bone structure is central spinal stenosis; if the target area is located in the lateral recess, the indication for determining the bone structure is lateral recess spinal stenosis; if the target area is located in the intervertebral foramen, the indication for determining the bone structure is intervertebral foramen spinal stenosis;

[0040] If the target area is located in the center of the back of the intervertebral disc, the indication for determining the bone structure is central type intervertebral disc herniation; if the target area is located in the intervertebral foramen but does not compress the nerve, the indication for determining the bone structure is intervertebral foraminal type intervertebral disc herniation.

[0041] Preferably, if the bone structure is a lumbar vertebra, determining the approach area of ​​the decompression path according to the indication, the location of the target area, and the morphological characteristics of the sub-molecular structures around the target area includes:

[0042] If the indication is central disc herniation, the target area is the posterior aspect of the disc, and the approach area is defined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower vertebra;

[0043] If the indication is lateral recess type intervertebral disc herniation, the target area is the lateral recess area of ​​the intervertebral disc, and the approach area is determined as the ipsilateral intervertebral foramen area;

[0044] If the indication is foraminal disc herniation, the target area is the foraminal area, and the approach area is determined as the ipsilateral foraminal area;

[0045] If the indication is central spinal stenosis, the target area is the posterior part of the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine;

[0046] If the indication is lateral recess type spinal stenosis, the target area is the unilateral or bilateral lateral recess area in the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine;

[0047] If the indication is foraminal spinal stenosis, the target area is unilateral or bilateral foraminal area, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine.

[0048] Preferably, determining the surgical procedure and decompression path according to the target area and approach area includes:

[0049] If the indication is central intervertebral disc herniation, or if the indication is central spinal stenosis, or if the indication is lateral recess spinal stenosis and it is unilateral lateral recess spinal stenosis; or if the indication is foraminal spinal stenosis and it is unilateral foraminal spinal stenosis, the surgical procedure is determined to be the interlaminar approach, and the decompression path is the outer circumferential tangent line between the target area and the approach area;

[0050] If the indication is lateral recess type disc herniation, or if the indication is foraminal type disc herniation, the surgical procedure is determined to be a transforaminal approach, and the decompression path is the outer circumferential tangent line between the target area and the approach area;

[0051] If the indication is lateral recess spinal stenosis and the lateral recess spinal stenosis is bilateral; or if the indication is intervertebral foraminal spinal stenosis and the intervertebral foraminal spinal stenosis is bilateral, the surgical procedure is determined to be ULBD, and the decompression path is the outer circumferential tangent line between the target area and the approach area.

[0052] According to a third aspect of the present invention, there is provided an electronic device, comprising:

[0053] A processor, and a memory connected to the processor;

[0054] The memory is used to store computer programs;

[0055] The processor is configured to call and execute the computer program in the memory to perform the above method.

[0056] According to a fourth aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the above method.

[0057] The technical solutions provided by the embodiments of the present invention may have the following beneficial effects:

[0058] By registering and fusing CT images with MRI images, the soft tissue structure in the MRI image is accurately generated in the bony channel of the CT image. Based on the registered and fused images, the position and morphology of the soft tissue structure in the bony channel are analyzed to see if there is any pathology, thereby achieving automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the existing technology, this method reduces the problems of missed and wrong judgments caused by manual positioning, lays a solid foundation for the accurate planning of the decompression path, can improve the success rate of surgery, reduce the rate of postoperative revisions, and enhance the patient experience.

[0059] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a flow chart showing a method for determining a target decompression area of ​​a bone structure according to an exemplary embodiment;

[0061] Figure 2 is a schematic diagram showing the generation of the spinal canal and the neural structures within the spinal canal according to an exemplary embodiment;

[0062] Figure 3is a schematic diagram showing the generation of an intervertebral foramen and a neural structure within the intervertebral foramen according to an exemplary embodiment;

[0063] Figure 4 is a schematic diagram showing an intervertebral disc region and generation of an intervertebral disc within the intervertebral disc region according to an exemplary embodiment;

[0064] Figure 5 is a schematic diagram showing the generation of the ligamentum flavum in the spinal canal according to an exemplary embodiment;

[0065] Figure 6 is a flow chart showing a method for automatically planning a bone structure decompression path according to an exemplary embodiment;

[0066] Figure 7 is a schematic diagram showing different types of intervertebral disc herniation according to an exemplary embodiment;

[0067] Figures 8A to 8C is a schematic diagram showing planning of an interlaminar approach according to an exemplary embodiment;

[0068] Figures 9A to 9C is a schematic diagram showing a transforaminal approach planning according to an exemplary embodiment;

[0069] Figures 10A to 10C is a schematic diagram showing ULBD approach planning according to an exemplary embodiment;

[0070] Figures 11A and 11B is a schematic diagram showing cutting using a curved surface component according to an exemplary embodiment;

[0071] Figures 12A to 12H is a schematic structural diagram of each single vertebral structure according to an exemplary embodiment;

[0072] Figures 13A to 13C is a subdivision graph obtained after secondary cutting according to an exemplary embodiment;

[0073] Figure 14 is a coarse segmented image obtained after primary cutting according to an exemplary embodiment;

[0074] Figures 15A to 15E is a schematic diagram showing the morphological characteristics of the lumbar vertebra structure according to an exemplary embodiment;

[0075] Figure 16 is a schematic structural diagram of a 3D U-net network model according to an exemplary embodiment;

[0076] Figure 17 is an output effect diagram of a 3D U-net network model according to an exemplary embodiment;

[0077] Figure 18 The figure is a flowchart of a method for automatic segmentation of the lumbar vertebrae according to an exemplary embodiment. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0079] As described in the background art above, there is a problem in the related art of manually determining the target decompression area, which results in the subsequent decompression path planning being unable to accurately or completely cover the lesion, resulting in low accuracy of the decompression path planning.

[0080] In order to effectively solve the problems in the related art, the present invention provides a method for determining a target decompression area of ​​a bone structure, a method for automatically planning a bone structure decompression path, an electronic device and a storage medium, which are described in detail below.

[0081] Example 1

[0082] Figure 1 is a flow chart showing a method for determining a target decompression area of ​​a bone structure according to an exemplary embodiment, wherein the bone structure includes a plurality of single vertebral bone structures, and each single vertebral bone structure includes a plurality of sub-structures, such as Figure 1 As shown, the method includes:

[0083] Step S11: Acquire a CT image of the vertebral structure within a preset marked area, and an MRI image of the vertebral structure corresponding to the preset marked area, wherein the CT image carries substructure information of the vertebral structure and information about bony channels formed between adjacent single vertebral structures;

[0084] Step S12: registering the CT image with the MRI image, extracting the soft tissue structure from the MRI image according to the boundary position of the bony channel, and generating the soft tissue structure within the bony channel;

[0085] Step S13: extracting morphological features of each soft tissue structure, and judging whether the soft tissue structure has a lesion based on the morphological features, and determining the area where the lesion occurs as the target decompression area.

[0086] It should be noted that the technical solution provided in this embodiment, in specific practice, runs in the controller of the medical device, or is loaded into an electronic device connected to the controller and runs. The controller of the medical device executes the corresponding method by calling the program stored in the electronic device.

[0087] It can be understood that the technical solution provided in this embodiment accurately generates the soft tissue structure in the MRI image in the bony channel of the CT image through the registration and fusion of the CT image and the MRI image, and then analyzes the position and morphology of the soft tissue structure in the bony channel based on the registered and fused image to see whether there is any pathology, thereby realizing automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the prior art, since the problems of missed judgment and wrong judgment caused by manual positioning are reduced, a solid foundation is laid for the accurate planning of the decompression path, which can improve the success rate of the operation, reduce the postoperative revision rate, and enhance the patient experience.

[0088] In specific practice, if the single vertebral structure is a lumbar vertebra, the bony channel information includes at least: the spinal canal, the intervertebral foramen, and the intervertebral disc area. In step S12, the soft tissue structure is extracted from the MRI image and generated in the bony channel, including:

[0089] The MRI image within the range of the spinal canal and nerve roots is cropped, and the grayscale of the soft tissue structure in the MRI image is threshold analyzed to extract the soft tissue structure, including: neural structure, yellow ligament, and possible protruding intervertebral disc;

[0090] The MRI image within the intervertebral disc area is cropped, and the grayscale of the soft tissue structure in the MRI image is threshold analyzed to extract the soft tissue structure, including the intervertebral disc and the possible protruding nucleus pulposus;

[0091] Generating the neural structure in the spinal canal, wherein the nerve roots branching from the spinal cord exit through the intervertebral foramina on both sides of the spinal canal;

[0092] Generating the yellow ligament in the spinal canal, the yellow ligament runs from the lower edge and inner surface of the upper lumbar vertebra to the upper edge and outer edge of the lower lumbar vertebra, participating in forming the posterior wall and posterolateral wall of the spinal canal, and gradually thickening from top to bottom;

[0093] The intervertebral disc and nucleus pulposus are generated in the intervertebral disc area.

[0094] It should be noted that the above extraction process and generation process are not in particular order and can be executed in parallel or sequentially.

[0095] See also Figure 2 The spinal canal A10 is formed based on the medial surface of the left and right pedicles, the inner surface of the vertebral body, and the ventral surface of the lamina. The neural structure A11 is generated in the spinal canal A10. Among them, the nerve roots that branch out from the spinal cord pass through the intervertebral foramina B10 on both sides of the spinal canal A10 (see intervertebral foramina B10). Figure 3 shown).

[0096] See also Figure 3The intervertebral foramen B10 is formed based on the gap formed by the upper and lower cones. The upper and lower boundaries of the intervertebral foramen B10 are the pedicles, the anterior boundary is the posterolateral surface of the vertebral body and intervertebral disc, and the posterior boundary is the joint capsule of the intervertebral joint.

[0097] See also Figure 4 The intervertebral disc region C10 is formed by the lower endplate of the upper lumbar vertebra and the upper endplate of the lower lumbar vertebra. The posterior boundary of the intervertebral disc region C10 is the ventral surface of the spinal canal. The intervertebral disc C11 is generated within the intervertebral disc region C10.

[0098] See also Figure 5 The yellow ligament A12 is generated in the spinal canal A10, extending from the lower edge and inner surface of the upper lumbar vertebra to the upper edge and outer edge of the lower lumbar vertebra, and participates in forming the posterior wall and posterolateral wall of the spinal canal, thickening from top to bottom.

[0099] It is understandable that each soft tissue structure, such as the neural structure, the ligamentum flavum, and the possibly protruding intervertebral disc, has a different grayscale value. By performing a threshold analysis based on the grayscale, it is possible to classify the soft tissue structures into neural structures, ligamentum flavum, and intervertebral discs.

[0100] Specifically, the morphological features include at least size and position. In step S13, judging whether the soft tissue structure has a lesion based on the morphological features and determining the area with the lesion as the target decompression area includes:

[0101] If the position of the intervertebral disc exceeds the preset position and / or the size exceeds the preset size, it is determined that the intervertebral disc is herniated, and the herniated intervertebral disc is determined as a target decompression area;

[0102] If the position of the ligamentum flavum exceeds the preset position and / or the size exceeds the preset size, it is determined that the ligamentum flavum is hypertrophic, and the hypertrophic ligamentum flavum is determined as the target decompression area;

[0103] The dimensions include: length, width, height, and volume converted from the length, width, and height.

[0104] Preferably, the subdivision structure information includes osteophyte information, and the method further includes:

[0105] Calculating the position and size information of the osteophyte;

[0106] Based on the position and size information of the osteophyte, it is determined whether the osteophyte intersects with the adjacent neural structure. If so, the area of ​​the intersection region between the osteophyte and the neural structure is calculated, and the intersection region is determined as the target decompression area.

[0107] Preferably, the method further comprises:

[0108] Generate a 3D model of the target decompression area to facilitate subsequent target decompression path planning.

[0109] It is understandable that there are three main factors that lead to intervertebral disc herniation and spinal stenosis: 1. Intervertebral disc herniation compresses the nerves, 2. Flavum hypertrophy compresses the nerves, and 3. Osteophyte hyperplasia compresses the nerves.

[0110] The technical solution provided in this embodiment takes into account the main factors that lead to intervertebral disc herniation and spinal canal stenosis, and can comprehensively locate the lesions that lead to intervertebral disc herniation and spinal canal stenosis, laying a solid foundation for accurate planning and comprehensive coverage of the subsequent decompression path, which can improve the success rate of surgery and reduce the rate of postoperative revisions.

[0111] Example 2

[0112] Figure 6 The flowchart of a method for automatically planning a bone structure decompression path according to an exemplary embodiment includes:

[0113] Step S21: determining a target decompression area according to the above method, and determining the target decompression area as a target area;

[0114] Step S22: determining the indication of the bone structure according to the location of the target area;

[0115] Step S23, determining the entry area of ​​the decompression path according to the indication, the location of the target area, and the morphological characteristics of the subdivided molecular structures around the target area;

[0116] Step S24: Determine the surgical procedure and decompression path according to the target area and approach area.

[0117] See also Figure 7 Taking the lumbar vertebra as an example, if the bone hyperplasia is located at Figure 7 The area enclosed by the dotted line 1 in the figure will result in a central disc herniation. If the area where the osteophyte grows is in Figure 7 The area enclosed by the dotted line 2 in the figure will result in lateral recess type disc herniation; if the area where the osteophyte is located is Figure 7 The area enclosed by the dotted line 3 in the figure will result in intervertebral foramen-type disc herniation; if the area where the osteophyte is located is Figure 7 If the area enclosed by the dotted line 4 in the figure is located outside the intervertebral foramen, then a herniated disc will occur. As can be seen, different locations of the target decompression area will lead to different types of symptoms.

[0118] Specifically, if the bone structure is a lumbar vertebra, the determination of the indication for the bone structure based on the location of the target area includes:

[0119] If the target area is located in the center of the posterior part of the spinal canal, the indication for determining the bone structure is central spinal stenosis; if the target area is located in the lateral recess, the indication for determining the bone structure is lateral recess spinal stenosis; if the target area is located in the intervertebral foramen, the indication for determining the bone structure is intervertebral foramen spinal stenosis;

[0120] If the target area is located in the center of the back of the intervertebral disc, the indication for determining the bone structure is central type intervertebral disc herniation; if the target area is located in the intervertebral foramen but does not compress the nerve, the indication for determining the bone structure is intervertebral foraminal type intervertebral disc herniation.

[0121] It should be noted that the above-mentioned indications for determining bone structure are calculated based on the morphology (geometric dimensions, area and volume, etc.) of the corresponding segmental spinal canal, dural sac and lateral recess after detailed segmentation. According to existing clinical quantitative standards, spinal stenosis and / or intervertebral disc herniation are automatically diagnosed and classified according to the location of the target area.

[0122] Specifically, if the bone structure is a lumbar vertebra, determining the approach area of ​​the decompression path according to the indication, the location of the target area, and the morphological characteristics of the sub-structures around the target area includes:

[0123] If the indication is central disc herniation, the target area is the posterior part of the intervertebral disc, and the approach area is determined as the lower 1 / 3 lamina area (from the junction of the inferior articular process and the lamina to the junction of the lamina and the spinous process, about 5 mm). 3 and the upper 1 / 3 lamina area of ​​the lower spine (from the junction of the superior articular process and the lamina to the inner 1 / 3 between the lamina and the spinous process, about 5 mm). 3 area);

[0124] If the indication is lateral recess type intervertebral disc herniation, the target area is the lateral recess area of ​​the intervertebral disc, and the approach area is determined as the ipsilateral intervertebral foramen area (Kambin's triangle area surrounded by the superior articular process, nerve root and superior endplate);

[0125] If the indication is foraminal disc herniation, the target area is the foraminal area, and the approach area is determined as the ipsilateral foraminal area (the Kambin triangle area surrounded by the superior articular process, nerve root, and superior endplate);

[0126] If the indication is central spinal stenosis, the target area is the posterior part of the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area (from the junction of the inferior articular process and the lamina to the junction of the lamina and the spinous process, about 5 mm). 3 and the upper 1 / 3 lamina area of ​​the lower spine (from the junction of the superior articular process and the lamina to the inner 1 / 3 of the junction between the lamina and the spinous process, about 5 mm). 3 area);

[0127] If the indication is lateral recess type spinal stenosis, the target area is the unilateral or bilateral lateral recess area in the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area (from the junction of the inferior articular process and lamina to the junction of the lamina and spinous process, about 5 mm). 3 and the upper 1 / 3 lamina area of ​​the lower spine (from the junction of the superior articular process and the lamina to the inner 1 / 3 between the lamina and the spinous process, about 5 mm). 3 area);

[0128] If the indication is intervertebral foraminal spinal stenosis, the target area is unilateral or bilateral intervertebral foramina, and the approach area is determined as the lower 1 / 3 lamina area (from the junction of the inferior articular process and lamina to about 5mm between the junction of the lamina and spinous process). 3 and the upper 1 / 3 lamina area of ​​the lower spine (from the junction of the superior articular process and the lamina to the inner 1 / 3 between the lamina and the spinous process, about 5 mm). 3 area).

[0129] Specifically, the surgical procedure and decompression path are determined based on the target area and approach area, including:

[0130] If the indication is central disc herniation, or if the indication is central spinal stenosis, or if the indication is lateral recess spinal stenosis and it is unilateral lateral recess spinal stenosis; or if the indication is foraminal spinal stenosis and it is unilateral foraminal spinal stenosis, the surgical procedure is determined to be the interlaminar approach (such as Figures 8A to 8C The decompression path is the outer circumferential tangent of the target area and the approach area (as shown in Figure 8A and Figure 8B in the direction of the dotted arrow);

[0131] If the indication is lateral recess type disc herniation, or if the indication is foraminal type disc herniation, the surgical procedure is determined to be a transforaminal approach (such as Figures 9A to 9C The decompression path is the outer circumferential tangent of the target area and the approach area (as shown in Figure 9A and Figure 9B in the direction of the dotted arrow);

[0132] If the indication is lateral recess type spinal stenosis and the lateral recess type spinal stenosis is bilateral; or if the indication is intervertebral foraminal type spinal stenosis and the intervertebral foraminal type spinal stenosis is bilateral, the surgical procedure is determined to be ULBD (such as Figures 10A to 10C The decompression path is the outer circumferential tangent of the target area and the approach area (as shown in Figure 10A and Figure 10B in the direction of the dotted arrow).

[0133] It should be noted that ULBD (Unilateral laminotomy for Bilateral Decompression), a unilateral laminectomy for bilateral spinal canal decompression surgery, can effectively decompress the bilateral lateral recesses and central spinal canal while fully preserving the stable structure of the posterior lumbar spine.

[0134] It can be understood that the technical solution provided in this embodiment accurately generates the soft tissue structure in the MRI image in the bony channel of the CT image through the registration and fusion of the CT image and the MRI image, and then analyzes the position and morphology of the soft tissue structure in the bony channel based on the registered and fused image to see whether there is any pathology, thereby realizing automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the prior art, since the problems of missed judgment and wrong judgment caused by manual positioning are reduced, a solid foundation is laid for the accurate planning of the decompression path, which can improve the success rate of the operation, reduce the postoperative revision rate, and enhance the patient experience.

[0135] Furthermore, based on the automatic determination of the target decompression area, the technical solution provided in this embodiment provides automatic planning of the bone structure decompression path, generates a complete decompression plan including channel positioning and bone cutting methods, and realizes the complete technical implementation from disease positioning to path planning in clinical practice, which can help surgeons improve the success rate of surgery and reduce the postoperative revision rate.

[0136] It should be noted that the above embodiments are all implemented based on step S11 "the CT image carries the subdivided substructure information of the spinal structure", and step S11 can be implemented in multiple ways in practice, one of which may include:

[0137] Step S110: obtaining an original image containing a complete vertebral structure, and continuously cutting the vertebral structure in the original image to obtain a plurality of original images of single vertebral structures;

[0138] Step S120: Subdividing the single vertebral structure in the original image of each single vertebral structure into substructures, and adding a mask to the subdivided substructures to obtain a substructure mask map;

[0139] Step S130: For any single vertebra structure, the original image and substructure mask of the single vertebra structure are input into a pre-trained single-vertebra multi-level multi-task model group to obtain a subdivided substructure of the single vertebra structure.

[0140] In specific practice, in step S110, the vertebral structure in the original image is continuously cut to obtain multiple original images of single vertebral structures, specifically:

[0141] Obtain 3D original images containing the complete spinal structure;

[0142] The complete vertebral structure in the 3D original image is continuously cut using a surface component, and the area between two adjacent target surfaces is determined as a single vertebral structure.

[0143] See also Figure 11A and Figure 11B When the surface component is initially placed in a certain position in the complete spinal structure, it defaults to a plane. However, when the bone surface at that position is irregular in shape, the adaptability will be adjusted manually. Figure 11B Curved surfaces are more accurate but more time-consuming to cut. Both flat and curved surfaces can be manually sized to achieve precise segmentation of bone structures, and curved surfaces can be set to a curvature to better wrap around curved structures.

[0144] See also Figures 12A to 12H ,After continuous cutting of the complete spinal structure, multiple original images of single ,vertebral structures were obtained, including the upper cervical vertebrae (C1), upper cervical vertebrae (C2), lower cervical vertebrae (C3-C7), upper thoracic vertebrae (T1-T4), middle thoracic vertebrae (T3-T8), lower thoracic vertebrae (T9-T12), lumbar vertebrae (L1-L5) and sacral vertebrae.

[0145] In step S120 , masks are added to the subdivided substructures, which can be accomplished by an automatic segmentation method in the prior art, or by a manual automatic or semi-automatic method.

[0146] Regarding step S130, in specific practice, the single-spine multi-level multi-task model group includes:

[0147] a first-level cutting model, configured to perform first-level cutting on the single vertebral structure in the original image according to the substructure mask image, to obtain a coarse-divided image containing the first-level single vertebral structure;

[0148] A plurality of secondary cutting models, used for performing secondary cutting on the primary single vertebral structure in the coarse segmented image to obtain a segmented image containing the secondary single vertebral structure;

[0149] Each single vertebral structure corresponds to a first-level cutting model, each first-level cutting model corresponds to multiple second-level cutting models, and each second-level cutting model corresponds to a type of clinical indication and planning requirement.

[0150] Assume that the single vertebra structure to be segmented is the lumbar vertebra, see Figures 13A to 13C According to the traditional definition, a single lumbar vertebra is mainly divided into the vertebral body, pedicle, facet joint (including superior and inferior facet joints), lamina, transverse process and spinous process. According to the artificial preset rules (artificial preset rules can refer to spinal development, spinal biomechanics and clinical application significance), see Figure 14, the vertebral body can be classified as the anterior structure 101 , and the facet joints, lamina, transverse processes, and spinous processes can be classified as the posterior structure 102 .

[0151] It should be noted that, since subsequent navigation planning often requires the location near the pedicle as the surgical entry point, this embodiment proposes the concept of a pedicle entry point region to better facilitate entry point planning. The pedicle entry point region can be understood as consisting of the transverse process, superior articular process, and a portion of the pedicle. Its location intersects with all three, thus defining it as the intersection region. This definition has the advantage of identifying the pedicle screw entry point region, allowing for better entry point determination and reducing the risk of bone slippage.

[0152] Therefore, considering the importance of the entry point area in robot-assisted pedicle screw insertion, this embodiment uses the traditional classification method to classify the entry point area into the following categories: Figure 13A The intersection of the left pedicle 2, left transverse process 6, and left superior articular process 8 is defined as the left pedicle entry point area 4; the intersection of the right pedicle 3, right transverse process 7, and right superior articular process 9 is defined as the right pedicle entry point area 5. In the first-level cutting model, Figure 13A The left pedicle entry point area 4 and the right pedicle entry point area 5 are defined as Figure 14 The rear structure 102 in.

[0153] See also Figure 14 Assuming that the single vertebra structure to be segmented is a lumbar vertebra, the substructures of the lumbar vertebra after the first-level segmentation include: a vertebral body 101 , a posterior structure 102 and a pedicle 103 .

[0154] See also Figure 18 The lumbar spine corresponds to a first-level cutting model and multiple second-level cutting models, where the purpose of the first-level cutting model is to obtain Figure 14 The purpose of the secondary cutting model is to obtain the coarse image shown in the figure. Figures 13A to 13C The segmented image shown. Each secondary cutting model is trained to suit different clinical indications and planning requirements. Therefore, for the lumbar spine, there can be multiple secondary cutting models, each designed to perform different tasks.

[0155] See also Figure 13A The substructures after secondary cutting of the lumbar vertebrae include: vertebral body 1, left pedicle 2, right pedicle 3, left pedicle entry area 4, right pedicle entry area 5, left transverse process 6, right transverse process 7, spinous process 14, left superior articular process 8, left inferior articular process 12, right superior articular process 9, right inferior articular process 13, left vertebral lamina 10 and right vertebral lamina 11, a total of 14 secondary substructures, including 12 symmetrical structures and 2 independent structures (including vertebral body 1 and spinous process 14).

[0156] In summary, taking the lumbar vertebrae as an example, the substructures after the first-level cutting currently mainly include: vertebral body, pedicle and posterior structure; the substructures after the second-level cutting are mainly the subdivision of the posterior structure, mainly including: transverse process, pedicle entry area, superior articular process, inferior articular process, lamina and spinous process.

[0157] As you can understand, the technical solution provided in this embodiment utilizes a multi-level, multi-task fine segmentation model cluster strategy tailored to clinical indications and planning requirements. This strategy decomposes the complex task of spinal fine segmentation into multiple sub-models to meet indication-based planning requirements. This reduces the complexity of each model and improves its responsiveness. Users can select different secondary segmentation models based on different tasks, resulting in a wide range of applications and strong scalability.

[0158] In specific practice, the method further includes:

[0159] Connecting the disconnected areas of the coarse-divided image with the surrounding connected areas through a connected domain analysis algorithm, and inputting the connected coarse-divided image into a secondary cutting model;

[0160] The disconnected areas of the subdivided image are connected with the surrounding connected areas through a connected domain analysis algorithm, and the connected subdivided image is used as an output result.

[0161] In specific practice, the method further includes:

[0162] extracting morphological features of the secondary single vertebral structure in the segmented image;

[0163] According to the morphological characteristics, multiple single vertebral structures are fused to generate bony channels formed between the multiple single vertebral structures; the bony channels at least include: a spinal canal, an intervertebral foramen, and an intervertebral disc area.

[0164] See also Figures 15A to 15E From the perspective of clinical application, bony channels are of great significance. Therefore, after the substructures are subdivided, bony channels need to be generated. The bony channels include but are not limited to: the upper endplate of the vertebral body, the lower endplate of the vertebral body, the spinal canal, the intervertebral foramen, the facet joints and the lateral recess.

[0165] For example, the spinal canal is a bony channel that is naturally formed when multiple single lumbar vertebrae are combined upper and lower.

[0166] To understand the specific location of the bony tunnel, see Figures 15A to 15E , Figure 15A Indicates the upper end plate (the part within the upper dotted line in the figure) and the lower end plate (the part within the lower dotted line in the figure) of the vertebral body; Figure 15BIndicates the vertebral canal (the dotted line in the figure contains the spinal cord); Figure 15C Indicates the intervertebral foramen (within the dotted line in the figure, where nerve roots enter and exit); Figure 15D Indicates the facet joint and articular surface (within the dotted line in the figure); Figure 15E Indicates the lateral recess (within the dotted line in the figure).

[0167] It can be understood that the technical solution provided in this embodiment takes into account the bony channels between multiple single vertebral structures after cutting, which provides an accurate substructure segmentation map for subsequent clinical surgical planning, and lays a solid foundation for subsequent fast, reliable and stable initial planning, ultimately improving surgical efficiency, enhancing intraoperative decision-making assistance and safety reminders.

[0168] As mentioned above, the single-spine multi-level multi-task model group includes a primary cutting model and multiple secondary cutting models, and the primary cutting model and multiple secondary cutting models are all pre-trained.

[0169] Before model training, image preprocessing is required, which includes:

[0170] 1. Adjust the pixel spacing in the X, Y, and Z directions of the original image of each single vertebral structure to be consistent to obtain isotropic pixel volumes and masks;

[0171] 2. Divide the image grayscale into different levels and perform grayscale processing on the original image;

[0172] 3. Construct a three-dimensional sample based on a single vertebral structure. The size of the sample should at least cover the complete image of the bone structure to be segmented, so that the 3D U-net network model can learn the overall structural morphological characteristics.

[0173] In addition, during the image preprocessing stage, osteophytes need to be manually segmented from the spinal bone structure using grid components (after the osteophytes are segmented, the segmented osteophytes are not masked, and the model training will not process the osteophyte image) to avoid the osteophytes affecting the model training results. Due to the variable morphology of osteophytes, their segmentation is not currently performed automatically.

[0174] In addition, sample data enhancement during network model training can include adding random noise, rotating around the three axes XYZ within a certain angle range, scaling within a certain magnification range, etc.

[0175] It can be understood that image preprocessing can ensure that the data input into the subsequent first-level cutting model and second-level cutting model can be accurately identified and effectively processed, thereby improving the reliability and accuracy of model training.

[0176] In practice, the first-level cutting model is trained based on a 3D U-net network model and includes:

[0177] Obtaining training samples: obtaining an original image containing a complete vertebral structure, continuously cutting the vertebral structure in the original image containing the complete vertebral structure to obtain an original image containing multiple single vertebral structures, selecting an original image of the single vertebral structure to be trained from the multiple cut original images of the single vertebral structures; adding a mask to the first-level subdivision structure on the original image of the single vertebral structure to be trained to obtain a ground truth mask (GT mask); using the original image of the single vertebral structure to be trained and the GT mask as training samples;

[0178] Build the model architecture: Define the network parameters of the 3D U-net network model, including the number of convolutional layers, number of channels, loss function, and number of model optimization iterations; define two input channels, one for receiving the original image of the single vertebra structure to be trained and the ground truth mask; define one output channel for outputting the coarse-separated image;

[0179] Model training: The training samples are input into the constructed 3D U-net network model, and iterative optimization is performed. The loss function of the first-level substructure segmentation mask and the GT mask predicted by the model is calculated, and the network parameters are adjusted according to the loss function. The model converges until the number of optimization iterations reaches a preset number and the loss function value is within a preset threshold range and tends to be stable. The model is then marked as a first-level segmentation model.

[0180] Model prediction: The re-acquired training samples are input into the first-level cutting model to obtain a coarse-resolved image of the single vertebral structure to be trained.

[0181] Preferably, the training sample of the secondary cutting model is a coarse-divided image output by the primary cutting model, and the coarse-divided image carries mask information of the secondary single vertebral structure to be cut;

[0182] The network model structure and training method of the secondary cutting model are the same as those of the primary cutting model, with only the network model parameters being different (network model parameters include but are not limited to: convolution type, number of convolution layers, weight distribution in the loss function, number of channels, and fine-tuning of the number of optimization iterations of the model).

[0183] Taking the lumbar spine as an example, see Figure 16 The input of the 3D U-net network model is the original image of a single lumbar vertebra structure and the substructure mask map. The output of the first-level cutting model is a coarse image including the background, vertebral body, pedicle and posterior structure. The output effect of the model can be seen in Figure 17shown.

[0184] The loss function can be any of various methods, such as weighted cross entropy or Dice loss function, or a combination thereof. In practice, given that the volume sizes of different categories may differ significantly during substructure segmentation, the loss function of this embodiment preferably uses weighted cross entropy. For example, the weights of [background, vertebral body, pedicle, posterior structure] are set to [1, 1, 6, 1], respectively, to reduce the problem of volume imbalance between categories and increase the weight of the smaller pedicles in the loss function.

[0185] It can be understood that the technical solution provided in this embodiment can be applied to various application scenarios, including but not limited to processing complex and changeable substructure segmentation from the cervical spine to the thoracic and lumbar spine, and substructure segmentation of any segment of the lumbar spine and thoracic spine, which only requires different classification methods to be used for subdividing the substructure.

[0186] In addition, this embodiment also proposes a fine segmentation model group strategy based on multi-level and multi-task, which can better improve the segmentation accuracy from coarse to fine segmentation substructures. At the same time, according to different clinical indications and planning requirements, secondary cutting models are selectively trained to complete different tasks, which reduces the complexity and mutual interference of the model's fine substructure segmentation and can improve the accuracy of subsequent planning.

[0187] It is understandable that the introduction of multi-level models also reduces the difficulty of network model training. For a network model, the more categories that need to be segmented, the more complex the network structure itself may be, the larger the amount of training data required, and the easier it is to confuse categories. Especially for delicate structures such as the spine, each substructure is relatively small, there is no obvious grayscale difference between substructures, and it becomes more difficult to distinguish after too many classifications. The technical solution provided by this implementation selectively trains secondary cutting models for different tasks according to different clinical indications and planning needs, reducing the complexity and mutual interference of the segmentation of fine substructures of the model, and can improve the accuracy of subsequent planning.

[0188] Example 3

[0189] According to an exemplary embodiment, an electronic device includes:

[0190] A processor, and a memory connected to the processor;

[0191] The memory is used to store computer programs;

[0192] The processor is configured to call and execute the computer program in the memory to perform the above method.

[0193] It can be understood that the technical solution provided in this embodiment accurately generates the soft tissue structure in the MRI image in the bony channel of the CT image through the registration and fusion of the CT image and the MRI image, and then analyzes the position and morphology of the soft tissue structure in the bony channel based on the registered and fused image to see whether there is any pathology, thereby realizing automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the prior art, since the problems of missed judgment and wrong judgment caused by manual positioning are reduced, a solid foundation is laid for the accurate planning of the decompression path, which can improve the success rate of the operation, reduce the postoperative revision rate, and enhance the patient experience.

[0194] Example 4

[0195] According to an exemplary embodiment, a non-transitory computer-readable storage medium storing computer instructions is shown, where the computer instructions are used to enable a computer to execute the above method.

[0196] It can be understood that the technical solution provided in this embodiment accurately generates the soft tissue structure in the MRI image in the bony channel of the CT image through the registration and fusion of the CT image and the MRI image, and then analyzes the position and morphology of the soft tissue structure in the bony channel based on the registered and fused image to see whether there is any pathology, thereby realizing automatic positioning of the target decompression area. Compared with the manual determination method of the target decompression area in the prior art, since the problems of missed judgment and wrong judgment caused by manual positioning are reduced, a solid foundation is laid for the accurate planning of the decompression path, which can improve the success rate of the operation, reduce the postoperative revision rate, and enhance the patient experience.

[0197] Of course, those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a memory, a magnetic disk, an optical disk, etc.

[0198] The specific embodiments of the present invention described above do not limit the scope of protection of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for determining a target decompression area of ​​a bone structure, characterized in that: include: Obtaining a CT image of the vertebral structure within a preset marked area, and an MRI image of the vertebral structure corresponding to the preset marked area, wherein the CT image carries substructure information of the vertebral structure and information about bony channels formed between adjacent single vertebral structures; registering the CT image with the MRI image, extracting a soft tissue structure from the MRI image based on a boundary position of the bony channel, and generating the soft tissue structure within the bony channel; The morphological features of each soft tissue structure are extracted, and whether the soft tissue structure has a lesion is determined based on the morphological features, and the area where the lesion occurs is determined as the target decompression area.

2. The method according to claim 1, characterized in that The bony channel information includes at least: the spinal canal, the intervertebral foramen, and the intervertebral disc area. The soft tissue structure is extracted from the MRI image and generated in the bony channel, including: The MRI image within the spinal canal and nerve root range is cropped out, and the grayscale of the soft tissue structure in the cropped MRI image is threshold analyzed to extract the soft tissue structure, including: neural structure, yellow ligament, and possible herniated intervertebral disc; The MRI image within the intervertebral disc area is cropped out, and the grayscale of the soft tissue structure in the cropped MRI image is threshold analyzed to extract the soft tissue structure, including the intervertebral disc and the possible protruding nucleus pulposus; Generating the neural structure in the spinal canal, wherein the nerve roots branching from the spinal cord exit through the intervertebral foramina on both sides of the spinal canal; Generating the yellow ligament in the spinal canal, the yellow ligament runs from the lower edge and inner surface of the upper lumbar vertebra to the upper edge and outer edge of the lower lumbar vertebra, participating in forming the posterior wall and posterolateral wall of the spinal canal, and gradually thickening from top to bottom; The intervertebral disc and nucleus pulposus are generated in the intervertebral disc area.

3. The method according to claim 2, characterized in that The morphological characteristics include at least size and position. The determining whether the soft tissue structure has a lesion based on the morphological characteristics and determining the area where the lesion occurs as the target decompression area includes: If the position of the intervertebral disc exceeds the preset position and / or the size exceeds the preset size, it is determined that the intervertebral disc is herniated, and the herniated intervertebral disc is determined as a target decompression area; If the position of the ligamentum flavum exceeds the preset position and / or the size exceeds the preset size, it is determined that the ligamentum flavum is hypertrophic, and the hypertrophic ligamentum flavum is determined as the target decompression area; The dimensions include: length, width, height, and volume converted from the length, width, and height.

4. The method according to claim 3, characterized in that Also includes: According to the characteristic data of each sub-molecule structure, it is determined whether there is osteophyte on each sub-molecule structure; If it is determined that osteophytes exist, extracting characteristic data of the osteophytes, the characteristic data of the osteophytes including at least: position and size information of the osteophytes; Based on the position and size information of the osteophyte, it is determined whether the osteophyte intersects with the adjacent neural structure. If so, the area of ​​the intersection region between the osteophyte and the neural structure is calculated, and the intersection region is determined as the target decompression area.

5. A method for automatically planning a bone structure decompression path, characterized in that: include: Determine a target decompression area according to the method according to any one of claims 1 to 4, and determine the target decompression area as a target area; Determine the indication for bone structure based on the location of the target area; Determine the entry area of ​​the decompression pathway based on the indication, the location of the target area, and the morphological characteristics of the substructures surrounding the target area; The surgical procedure and decompression path are determined based on the target area and approach area.

6. The method according to claim 5, characterized in that If the bone structure is a lumbar vertebra, the indications for the bone structure are determined based on the location of the target area, including: If the target area is located in the center of the posterior part of the spinal canal, the indication for determining the bone structure is central spinal stenosis; if the target area is located in the lateral recess, the indication for determining the bone structure is lateral recess spinal stenosis; if the target area is located in the intervertebral foramen, the indication for determining the bone structure is intervertebral foramen spinal stenosis; If the target area is located in the center of the back of the intervertebral disc, the indication for determining the bone structure is central type intervertebral disc herniation; if the target area is located in the intervertebral foramen but does not compress the nerve, the indication for determining the bone structure is intervertebral foraminal type intervertebral disc herniation.

7. The method according to claim 6, characterized in that If the bone structure is a lumbar vertebra, determining the approach area of ​​the decompression path based on the indication, the location of the target area, and the morphological characteristics of the substructures surrounding the target area includes: If the indication is central disc herniation, the target area is the posterior aspect of the disc, and the approach area is defined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower vertebra; If the indication is lateral recess type intervertebral disc herniation, the target area is the lateral recess area of ​​the intervertebral disc, and the approach area is determined as the ipsilateral intervertebral foramen area; If the indication is foraminal disc herniation, the target area is the foraminal area, and the approach area is determined as the ipsilateral foraminal area; If the indication is central spinal stenosis, the target area is the posterior part of the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine; If the indication is lateral recess type spinal stenosis, the target area is the unilateral or bilateral lateral recess area in the spinal canal, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine; If the indication is foraminal spinal stenosis, the target area is unilateral or bilateral foraminal area, and the approach area is determined as the lower 1 / 3 lamina area and the upper 1 / 3 lamina area of ​​the lower spine.

8. The method according to claim 7, characterized in that Based on the target area and approach area, determine the surgical procedure and decompression path, including: If the indication is central intervertebral disc herniation, or if the indication is central spinal stenosis, or if the indication is lateral recess spinal stenosis and it is unilateral lateral recess spinal stenosis; or if the indication is foraminal spinal stenosis and it is unilateral foraminal spinal stenosis, the surgical procedure is determined to be the interlaminar approach, and the decompression path is the outer circumferential tangent line between the target area and the approach area; If the indication is lateral recess type disc herniation, or if the indication is foraminal type disc herniation, the surgical procedure is determined to be a transforaminal approach, and the decompression path is the outer circumferential tangent line between the target area and the approach area; If the indication is lateral recess spinal stenosis and the lateral recess spinal stenosis is bilateral; or if the indication is intervertebral foraminal spinal stenosis and the intervertebral foraminal spinal stenosis is bilateral, the surgical procedure is determined to be ULBD, and the decompression path is the outer circumferential tangent line between the target area and the approach area.

9. An electronic device, characterized in that: include: A processor, and a memory connected to the processor; The memory is used to store computer programs; The processor is configured to call and execute the computer program in the memory to perform the method according to any one of claims 1 to 8.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 8.

Images