Method and system for automatically planning thoracolumbar spinal dorsal stenosis bony decompression path

Abstract

The invention aims to provide a method and system for automatically planning a bony decompression path of thoracolumbar spinal stenosis, and relates to the technical field of biomedical volume data set segmentation, decompression targets of thoracolumbar spinal stenosis and preoperative image data are combined, and a standardized decompression range is automatically formulated; taking the anatomical details as a planning path reference, and formulating a decompression path of a spinous process tip point C, a lower marginal point P1 in the left upper vertebral pedicle, a lower marginal point P2 in the right upper vertebral pedicle, an upper marginal point S1 in the left lower vertebral pedicle and an upper marginal point S2 in the right lower vertebral pedicle in sequence; the precision and the safety of the operation are improved, and the clinical application value is remarkable.

Description

Technical Field

[0001] This invention relates to the field of biomedical dataset segmentation technology, and in particular to a method and system for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis. Background Technology

[0002] Surgical treatment of thoracolumbar spinal stenosis (T / LSS) requires precise resection of dorsal bony structures such as the lamina and facet joints to decompress nerve tissue. Unilateral bilateral spinal decompression (ULBD) is a minimally invasive lumbar spinal decompression technique. This surgery only requires percutaneous exposure of bony structures on one side, thereby reducing iatrogenic damage to the midline tension band structures of the spinous process / supraspinous ligament and the contralateral paravertebral muscles. Currently, unilateral biportal endoscopic discectomy (UBE) is increasingly used in the treatment of lumbar lateral spinal stenosis, but the following technical challenges have also emerged: 1. Endoscopic ULBD requires a certain learning curve: spatial orientation and identification of endoscopic anatomical structures require high technical skills and long-term experience in handling cases. Furthermore, due to variations in anatomical structures, normal anatomical landmarks of the spine are difficult to identify, thus increasing the difficulty and risk of the surgical procedure. 2. Unstable control of decompression degree: In cases of posterior lumbar spinal stenosis, traditional surgery relies on the doctor's experience and touch to judge the decompression range, which is easily affected by subjectivity and error. This can lead to the removal of too much bony structure, which damages spinal stability, or too little, which makes it difficult to effectively decompress and can easily damage surrounding normal structures. 3. Overhead technique under navigation: For severe cases of ossification of the ligamentum flavum (OLF) in the thoracic spine, it is necessary to precisely control the depth of bone and calcification grinding. If the grinding is too shallow, more calcification will remain, and if the grinding is too deep, it will easily cause cerebrospinal fluid leakage and nerve damage. This is also particularly important for decompression on the contralateral side after overhead grinding. 4. Long operation time: Traditional surgery relies on the doctor's experience and touch to determine the decompression range, which takes time and prolongs the operation time. Therefore, accurately determining the bony decompression range during surgery is crucial for fully relieving nerve compression and improving surgical efficacy. This application designs an automatic planning method and system for bilateral decompression pathways via a posterior unilateral approach in thoracoacic / lumbar spinal stenosis (T / LSS) based on image navigation, to assist spinal surgeons in accurately identifying the decompression range and performing precise operations during spinal decompression surgery. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides the following technical solution: A method for automatically planning bony decompression pathways for dorsal thoracolumbar spinal stenosis includes the following steps: S1, Acquire and process patient image data; S2. The patient's imaging data is converted and reconstructed to establish a coordinate system and obtain the patient's key coordinate points. The coordinate system is set with the spinous process tip C as the origin, the left-right direction of the patient as the X-axis, the patient's abdominal-dorsal direction as the Y-axis, and the patient's head-to-foot direction as the Z-axis. The key coordinate points are P1 (lower medial border of the pedicle of the left upper vertebral body), P2 (lower medial border of the pedicle of the right upper vertebral body), S1 (upper medial border of the pedicle of the left lower vertebral body), and S2 (upper medial border of the pedicle of the right lower vertebral body). The decompression range is determined based on the patient's key coordinate points. S3, based on the defined decompression range, plans the decompression path.

[0004] Furthermore, the image data in S1 includes image data from CT, MRI, and X-ray machines; after the image data is acquired, noise reduction and contrast enhancement are performed to optimize image quality.

[0005] Furthermore, the specific method in S2 is as follows: S2.1, convert the two-dimensional sequence of images of the image data into a three-dimensional model; S2.2, based on a U-Net-improved 3D segmentation network, separates the spinal bony structures from the surrounding tissues and extracts the multiplanar reconstruction combined MPR of the spinal bony structure surface; S2.3, Calibrate key coordinate points: With the spinous process tip C as the origin of the coordinate system, the left-right direction of the patient is the X-axis, the ventral-dorsal direction of the patient is the Y-axis, and the head-to-foot direction of the patient is the Z-axis. Set the coordinate system; obtain the coordinates of the key coordinate points, which are: P1, P2, S1, and S2 of the inner lower edge of the pedicle of the left upper vertebral body; S2.4, Mark key coordinate points on several previous cases and build a database based on the stenosis of previous cases and the corresponding decompression degree; S2.5, compare the identified narrowing conditions with the narrowing conditions in the database, and match the corresponding decompression degree; S2.6, the decompression range is formed by the four points enclosing the key coordinate points: P1, P2, S1, and S2, which are the inner lower edges of the pedicles of the upper left vertebral body, the inner lower edges of the pedicles of the upper right vertebral body, the inner upper edges of the pedicles of the lower left vertebral body, and the inner upper edges of the pedicles of the lower right vertebral body.

[0006] Furthermore, the bony structures include the lamina, facet joints, and pedicles, and the multiplanar reconstruction is combined into coronal, sagittal, and axial views.

[0007] Furthermore, in S3, the decompression path is as follows: the tip of the spinous process C, the lower inner edge of the pedicle of the upper left vertebral body P1, the lower inner edge of the pedicle of the upper right vertebral body P2, the upper inner edge of the pedicle of the lower left vertebral body S1, and the upper inner edge of the pedicle of the lower right vertebral body S2.

[0008] A system for automatically planning bony decompression pathways for dorsal thoracolumbar spinal canal stenosis includes an image processing module, a reconstruction module, a coordinate calibration module, a comparison module, and a decompression planning module. The image processing module includes an acquisition unit for acquiring raw medical image data and an image processing unit for optimizing the quality of the raw images in preparation for subsequent processing. The reconstruction module includes a reconstruction unit for converting acquired two-dimensional sequence images into a three-dimensional model, and a segmentation and extraction unit for separating spinal bony structures from surrounding tissues and extracting multiplanar reconstruction combined with MPR from the surface of spinal bony structures. The coordinate calibration module includes an identification coordinate unit for identifying key coordinate points and a calibration coordinate unit for calibrating key coordinate points. The comparison module includes an image recognition unit, a recording unit, a comparison unit, and a recall unit. The image recognition unit is used to identify the features of the bony structures extracted by the reconstruction module and determine the stenosis status. The recording unit is used to record key coordinate points calibrated by the coordinate calibration module, the identified stenosis status, and the stenosis status and corresponding decompression degree of previous cases. The comparison unit is used to compare and analyze the identified stenosis status with the recorded stenosis status and transmit the comparison results to the recall unit. The recall unit is used to receive the comparison results and search for the decompression degree corresponding to the recorded previous stenosis status. The decompression planning module includes an intervention unit and a generation unit. The intervention unit is used to input the decompression range and decompression path based on the comparison. The generation unit is used to generate the decompression range and decompression path based on the degree of decompression corresponding to the narrowing condition searched by the calling unit, or to generate the decompression range and decompression path based on the values ​​input by the intervention unit.

[0009] Compared with the prior art, the technical solution of this application has the following beneficial effects: This invention combines decompression target points for thoracolumbar spinal stenosis with preoperative imaging data to establish a standardized decompression range. It also automatically customizes the decompression range and path based on previous cases, solving the problem of relying on surgeons' experience and touch to determine the decompression range in traditional surgery, which is time-consuming and prone to inaccurate decompression. By using anatomical details as a reference for path planning, it establishes a decompression path sequentially at the spinous process tip C, the lower medial border of the left upper vertebral pedicle P1, the lower medial border of the right upper vertebral pedicle P2, the upper medial border of the left lower vertebral pedicle S1, and the upper medial border of the right lower vertebral pedicle S2, thus improving the precision and safety of the surgery. Attached Figure Description

[0010] Figure 1 This is a functional block diagram of the system for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis as described in this invention; Figure 2 This is a system flowchart for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis as described in this invention. Figure 3 This is a flowchart illustrating the steps of the method for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis according to the present invention. Figure 4 This is a detailed step diagram of S2 as described in the present invention; Figure 5 This is a schematic diagram of the system interface of the present invention; Figure 6 This is a schematic diagram of the coordinate calibration module interface of the system of the present invention; Figure 7 This is a schematic diagram of the decompression planning module interface of the system of the present invention; Figure 8 This is a schematic diagram of the working interface of the machine grinding module of the present invention; Figure 9 This is a schematic diagram of the coordinate system used for polishing in this invention; Figure 10 This is a reverse-printed model of the vertebral body used for decompression and grinding in this invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] Example 1: Please see Figure 1-10 A method for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis, comprising the following steps: S1, Acquire and process patient image data; S2. The patient's imaging data is converted and reconstructed to establish a coordinate system and obtain the patient's key coordinate points. The coordinate system is set with the spinous process tip C as the origin, the left-right direction of the patient as the X-axis, the patient's abdominal-dorsal direction as the Y-axis, and the patient's head-to-foot direction as the Z-axis. The key coordinate points are P1 (lower medial border of the pedicle of the left upper vertebral body), P2 (lower medial border of the pedicle of the right upper vertebral body), S1 (upper medial border of the pedicle of the left lower vertebral body), and S2 (upper medial border of the pedicle of the right lower vertebral body). The decompression range is determined based on the patient's key coordinate points. S3, Based on the defined decompression range, plan the decompression path; In this embodiment, the specific method in step S2 is as follows: S2.1, convert the two-dimensional sequence of images of the image data into a three-dimensional model; S2.2, based on a U-Net-improved 3D segmentation network, separates the spinal bony structures from the surrounding tissues and extracts the multiplanar reconstruction combined MPR of the spinal bony structure surface; S2.3, Calibrate key coordinate points: With the spinous process tip C as the origin of the coordinate system, the left-right direction of the patient is the X-axis, the ventral-dorsal direction of the patient is the Y-axis, and the head-to-foot direction of the patient is the Z-axis. Set the coordinate system; obtain the coordinates of the key coordinate points, which are: P1, P2, S1, and S2 of the inner lower edge of the pedicle of the left upper vertebral body; S2.4, Mark key coordinate points on several previous cases and build a database based on the stenosis of previous cases and the corresponding decompression degree; S2.5, compare the identified narrowing conditions with the narrowing conditions in the database, and match the corresponding decompression degree; S2.6, the decompression range is formed by the four points enclosing the key coordinate points: P1, P2, S1, and S2, which are the inner lower edges of the pedicles of the upper left vertebral body, the inner lower edges of the pedicles of the upper right vertebral body, the inner upper edges of the pedicles of the lower left vertebral body, and the inner upper edges of the pedicles of the lower right vertebral body.

[0013] Example 2: Please see Figure 1-10 A system for automatically planning bony decompression pathways for dorsal thoracolumbar spinal canal stenosis includes an image processing module, a reconstruction module, a coordinate calibration module, a comparison module, and a decompression planning module. In this embodiment, the image processing module includes an acquisition unit for acquiring raw medical image data and an image processing unit for optimizing the quality of the raw images in preparation for subsequent processing; the image inputs of the acquisition unit include CT, MRI, and X-ray machines; In this embodiment, the reconstruction module includes a reconstruction unit for converting acquired two-dimensional sequence images into a three-dimensional model, and a segmentation and extraction unit for separating spinal bony structures from surrounding tissues and extracting multiplanar reconstruction combined MPR of the spinal bony structure surface; the reconstruction unit generates stereoscopic views of structures such as vertebrae, nerves, and blood vessels; and provides coronal, sagittal, and axial views by segmenting and extracting multiplanar reconstruction combined MPR of specific tissue surfaces. In this embodiment, the coordinate calibration module includes an identification coordinate unit for identifying key coordinate points and a calibration coordinate unit for calibrating key coordinate points. The coordinate calibration module sets the coordinate system with the spinous process tip C as the origin, the left-right direction of the patient as the X-axis, the patient's abdominal-dorsal direction as the Y-axis, and the patient's head-to-foot direction as the Z-axis. The coordinates of the key coordinate points are calibrated. The key coordinate points are the lower inner edge of the pedicle of the left upper vertebral body P1, the lower inner edge of the pedicle of the right upper vertebral body P2, the upper inner edge of the pedicle of the left lower vertebral body S1, and the upper inner edge of the pedicle of the right lower vertebral body S2. In this embodiment, the comparison module includes an image recognition unit, a recording unit, a comparison unit, and a recall unit. The image recognition unit is used to identify the features of the bony structures extracted by the reconstruction module and determine the stenosis. The recording unit is used to record the key coordinate points calibrated by the coordinate calibration module, the identified stenosis, and the stenosis of previous cases, as well as the corresponding decompression degree. The comparison unit is used to compare and analyze the identified stenosis with the recorded stenosis and transmit the comparison results to the recall unit. The recall unit is used to receive the comparison results and search for the decompression degree corresponding to the recorded previous stenosis. In this embodiment, the decompression planning module includes an intervention unit and a generation unit; the intervention unit is used to input the decompression range and decompression path according to the comparison situation; the generation unit is used to generate the decompression range and decompression path according to the decompression degree corresponding to the narrowing situation searched by the calling unit, or to generate the decompression range and decompression path according to the value input by the intervention unit. In clinical use, the decompression path is imported into the existing grinding robot. After fixation, registration, and alignment, the machine's grinding module performs grinding according to the imported decompression path, specifically at the spinous process tip point C, the lower inner edge of the pedicle of the left upper vertebral body P1, the lower inner edge of the pedicle of the right upper vertebral body P2, the upper inner edge of the pedicle of the left lower vertebral body S1, and the upper inner edge of the pedicle of the right lower vertebral body S2. The grinding strategy is as follows: (1) The lower edge of the upper vertebral lamina, the insertion point of the head of the ligamentum flavum; (2) Expose the safe nest; (3) The medial border of the inferior joint is exposed to reveal the articular surface of the superior joint on the same side; (4) The upper edge of the lower vertebral lamina, the distal insertion point of the ligamentum flavum; (5) The medial border of the superior articular process on this side, and the lateral insertion of the ligamentum flavum; (6) The base of the spinous process at the cephalic end; (7) On the ventral side of the contralateral inferior articular process, the articular surface of the contralateral superior articular process is exposed; (8) The base of the distal spinous process; (9) The medial border of the contralateral superior articular process; (10) Contralateral safety nest; In this embodiment, the operational logic flow of the system for automatically planning bony decompression pathways for dorsal thoracolumbar spinal stenosis is as follows: First, a database was built based on previous case data: The image processing module collects raw medical image data from previous cases and optimizes the quality of the raw images. The original medical imaging data of the case is converted into a three-dimensional model through the reconstruction module; and based on the U-Net improved 3D segmentation network, the spinal bony structures are separated from the surrounding tissues, and the multiplanar reconstruction combined MPR of the surface of the spinal bony structures is extracted. The coordinate calibration module identifies key coordinate points in the original medical imaging data of the case. The coordinate system is set with the spinous process tip C as the origin, the left-right direction of the patient as the X-axis, the abdominal-dorsal direction of the patient as the Y-axis, and the head-to-foot direction of the patient as the Z-axis. The coordinates of the key coordinate points are obtained. The key coordinate points are: P1, P2, S1, and S2 of the inner lower edge of the pedicle of the left upper vertebral body; P2, P2, S1, S2; and S2 of the inner upper edge of the pedicle of the right lower vertebral body. The database is constructed by comparing the stenosis conditions and recording the key coordinates, stenosis conditions, and corresponding decompression degrees (i.e., the abrasion conditions during clinical decompression) of the identified cases. Secondly, based on the patient's medical imaging data, a decompression path is automatically planned: Import the patient's CT, MRI, or X-ray medical image data into the image processing module to optimize the quality of the original images; The reconstruction module converts patient medical image data into a three-dimensional model; and based on the U-Net improved 3D segmentation network, it separates the spinal bony structures from the surrounding tissues and extracts the multiplanar reconstruction combined MPR of the spinal bony structure surface. The narrowing situation is identified by the comparison module, and the coordinates of key coordinate points are identified by the coordinate calibration module. The comparison module searches the database for stenosis cases from previous cases, compares them with the identified stenosis cases, generates a comparison report and a decompression report for the most similar case; The decompression range and decompression path can be generated by the generation module based on the comparison report and decompression report; or the doctor can directly input the values ​​after viewing the comparison report and decompression report to generate the decompression range and decompression path. Import the generated decompression range and decompression path into the machine grinding module, prioritizing the decompression range and decompression path generated by the doctor's input values; The robotic arm grinds along the decompression path by aligning the machine with the intraoperative images and the planned decompression path using the machine grinding module.

[0014] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for automatically planning bony decompression pathways for dorsal thoracolumbar spinal stenosis, characterized in that, Includes the following steps: S1, Acquire and process patient image data; S2. The patient's imaging data is converted and reconstructed to establish a coordinate system and obtain the patient's key coordinate points. The coordinate system is set with the spinous process tip C as the origin, the left-right direction of the patient as the X-axis, the patient's abdominal-dorsal direction as the Y-axis, and the patient's head-to-foot direction as the Z-axis. The key coordinate points are P1 (lower medial border of the pedicle of the left upper vertebral body), P2 (lower medial border of the pedicle of the right upper vertebral body), S1 (upper medial border of the pedicle of the left lower vertebral body), and S2 (upper medial border of the pedicle of the right lower vertebral body). The decompression range is determined based on the patient's key coordinate points. S3, based on the defined decompression range, plans the decompression path.

2. The method for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis according to claim 1, characterized in that, The image data in S1 includes CT, MRI, and X-ray images; after the image data is acquired, noise reduction and contrast enhancement are performed to optimize image quality.

3. The method for automatically planning a bony decompression path for dorsal thoracic and lumbar spinal canal stenosis according to claim 1, characterized in that, The specific method in S2 is as follows: S2.1, convert the two-dimensional sequence of images of the image data into a three-dimensional model; S2.2, based on a U-Net-improved 3D segmentation network, separates the spinal bony structures from the surrounding tissues and extracts the multiplanar reconstruction combined MPR of the spinal bony structure surface; S2.3, Calibrate key coordinate points: With the spinous process tip C as the origin of the coordinate system, the left-right direction of the patient is the X-axis, the ventral-dorsal direction of the patient is the Y-axis, and the head-to-foot direction of the patient is the Z-axis. Set the coordinate system; obtain the coordinates of the key coordinate points, which are: P1, P2, S1, and S2 of the inner lower edge of the pedicle of the left upper vertebral body; S2.4, Mark key coordinate points on several previous cases and build a database based on the stenosis of previous cases and the corresponding decompression degree; S2.5, compare the identified narrowing conditions with the narrowing conditions in the database, and match the corresponding decompression degree; S2.6, the decompression range is formed by the four points enclosing the key coordinate points: P1, P2, S1, and S2, which are the inner lower edges of the pedicles of the upper left vertebral body, the inner lower edges of the pedicles of the upper right vertebral body, the inner upper edges of the pedicles of the lower left vertebral body, and the inner upper edges of the pedicles of the lower right vertebral body.

4. The method for automatically planning a bony decompression path for dorsal thoracolumbar spinal canal stenosis according to claim 3, characterized in that, The bony structures include the lamina, facet joints, and pedicles, and the multiplanar reconstruction is combined into coronal, sagittal, and axial views.

5. The method for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis according to claim 2, characterized in that, In S3, the decompression path is sequentially: the tip of the spinous process C, the lower inner edge of the pedicle of the upper left vertebra P1, the lower inner edge of the pedicle of the upper right vertebra P2, the upper inner edge of the pedicle of the lower left vertebra S1, and the upper inner edge of the pedicle of the lower right vertebra S2.

6. A system for automatically planning bony decompression pathways for dorsal thoracolumbar spinal stenosis, characterized in that, The system employs the method described in any one of claims 1-5.

7. The system for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis according to claim 6, characterized in that, The system includes an image processing module, a reconstruction module, a coordinate calibration module, a comparison module, and a decompression planning module.

8. The system for automatically planning bony decompression pathways for dorsal thoracic and lumbar spinal canal stenosis according to claim 7, characterized in that, The image processing module includes an acquisition unit for acquiring raw medical image data and an image processing unit for optimizing the quality of the raw images in preparation for subsequent processing. The reconstruction module includes a reconstruction unit for converting acquired two-dimensional sequence images into a three-dimensional model, and a segmentation and extraction unit for separating spinal bony structures from surrounding tissues and extracting multiplanar reconstruction combined with MPR from the surface of spinal bony structures. The coordinate calibration module includes an identification coordinate unit for identifying key coordinate points and a calibration coordinate unit for calibrating key coordinate points. The comparison module includes an image recognition unit, a recording unit, a comparison unit, and a recall unit. The image recognition unit is used to identify the features of the bony structures extracted by the reconstruction module and determine the stenosis status. The recording unit is used to record key coordinate points calibrated by the coordinate calibration module, the identified stenosis status, and the stenosis status and corresponding decompression degree of previous cases. The comparison unit is used to compare and analyze the identified stenosis status with the recorded stenosis status and transmit the comparison results to the recall unit. The recall unit is used to receive the comparison results and search for the decompression degree corresponding to the recorded previous stenosis status. The decompression planning module includes an intervention unit and a generation unit. The intervention unit is used to input the decompression range and decompression path based on the comparison. The generation unit is used to generate the decompression range and decompression path based on the degree of decompression corresponding to the narrowing condition searched by the calling unit, or to generate the decompression range and decompression path based on the values ​​input by the intervention unit.

Images