Anesthesiology spinal canal puncture teaching system and working method thereof

Abstract

This invention proposes a teaching system for spinal canal puncture in anesthesiology and its working method, including a sampling and positioning module, a virtual-real interaction module, and a feedback and evaluation module. The sampling and positioning module includes capturing the trajectory of a reflective marker ball fixed on the puncture needle and model, and capturing the puncture needle tip in three-dimensional space. The virtual-real interaction module includes a replaceable layered biomimetic structure module, which constructs a virtual anatomical model. The virtual-real interaction module uses a virtual-real mapping calibration algorithm to match the model with the actual object. The virtual-real interaction module uses a ligamentum flavum breakthrough detection algorithm to detect the breakthrough point of the actual ligamentum flavum. The feedback and evaluation module is used to construct a multi-dimensional quantitative evaluation system, recording and analyzing in real time indicators including positioning accuracy deviation, needle insertion angle drift, total operation time, and path tortuosity.

Description

Technical Field

[0001] This invention proposes a teaching system for spinal puncture in anesthesiology and its working method, which relates to the field of medical teaching technology. Background Technology

[0002] Spinal anesthesia is a cornerstone technique for cesarean sections, lower limb surgeries, and labor analgesia, with over ten million cases performed annually nationwide. However, its operation heavily relies on the surgeon's "feel" and spatial imagination, and teaching and training have long faced three major challenges.

[0003] First, clinical teaching carries high risks. Young doctors are directly experimenting on patients, and mistakes can easily lead to serious complications such as headaches, nerve damage, or even paraplegia, further exacerbating the already tense doctor-patient relationship.

[0004] Secondly, physical models have significant limitations. Traditional silicone models show obvious needle marks after multiple punctures, suffer severe wear and tear, and cannot simulate the unique "sudden reduction in resistance" that occurs when the puncture needle breaks through the ligamentum flavum, making it difficult to reproduce the subtle changes in the actual anatomical layers.

[0005] Third, virtual simulation has low accuracy. Currently, training equipment on the market generally suffers from problems such as visual rendering delay and spatial positioning deviation, and completely lacks the tactile feedback required for physical operation, resulting in a significant gap between training effects and real clinical scenarios. Summary of the Invention

[0006] In view of this, in order to fill the gaps and deficiencies in the prior art, the present invention proposes an anesthesiology spinal canal puncture teaching system and its working method to solve the problems that have arisen in the background art.

[0007] This invention proposes a teaching system for spinal puncture in anesthesiology and its working method, including the following:

[0008] According to a first aspect of the present invention, the present invention provides a teaching system for spinal puncture in anesthesiology, characterized in that it includes a sampling and positioning module, a virtual-real interaction module, and a feedback and evaluation module;

[0009] The sampling and positioning module uses an image acquisition module, an active light source module, and a tracking marker module to achieve sampling and positioning. The sampling and positioning includes capturing the trajectory of the reflective marker ball fixed on the puncture needle and the model and realizing the capture of the puncture needle tip in three-dimensional space.

[0010] The virtual-real interaction module includes a replaceable layered biomimetic structure module, which constructs a virtual anatomical model; the virtual-real interaction module uses a virtual-real mapping calibration algorithm to match the model with the real object; and the virtual-real interaction module uses a ligamentum flavum breakthrough detection algorithm to detect the breakthrough point of the real ligamentum flavum.

[0011] The feedback and evaluation module is used to build a multi-dimensional quantitative evaluation system, which records and analyzes in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

[0012] Furthermore, the image acquisition module is used to filter out visible light interference, including surgical shadowless lamps; the image acquisition module includes an industrial camera, wherein a narrow-band infrared filter with a center wavelength of 850nm is installed in front of the lens of the industrial camera, allowing only the infrared light reflected by the marker ball to enter.

[0013] Furthermore, the active light source module emits invisible infrared light in the 850nm band into the operating area.

[0014] Furthermore, the tracking and marking module samples the puncture needle handle and rigidly connects the positioning rigid body of each working point through a 3D printed fixture. The positioning rigid body has four spherical marking points coated with a glass microsphere reflective coating, which has high reflectivity to 850nm infrared light.

[0015] Furthermore, the sampling and positioning module includes the following:

[0016] The sampling method of the sampling and localization module includes the following: For each frame of image, extract the gray-level centroid within the connected component as the sub-pixel center coordinates of the marker point:

[0017] The positioning method of the sampling and positioning module includes the following:

[0018] Zhang's calibration method is used to obtain the camera intrinsic parameter K and the binocular extrinsic parameters R and t. Stereo calibration is completed by minimizing the reprojection error. The marker points in the left and right images are matched using epipolar geometric constraints, and mismatches are eliminated by combining the known relative positions of the marker points on the rigid body. The matched point pairs are triangulated and reconstructed to obtain the three-dimensional coordinates of the marker points. Then, singular value decomposition is used to solve the real-time pose of the needle tip.

[0019] Furthermore, the replaceable layered biomimetic structural module is manufactured using a multi-casting process; the replaceable layered biomimetic structural module also includes:

[0020] The underlying structure of the latex membrane is used to simulate the ligamentum flavum;

[0021] A middle layer structure with a hollow structure and an interior coated with colorless and transparent water-based gel is used to simulate the spinal canal cavity;

[0022] The outer layer structure uses platinum-silicone to simulate skin and subcutaneous tissue.

[0023] Furthermore, the virtual anatomical model includes the following:

[0024] The virtual anatomical model is constructed based on thin-slice CT data of the human thoracolumbar region, which is then segmented by thresholding and reconstructed in three dimensions. The final result is an STL mesh model that includes the skin, ligaments, and epidural space.

[0025] Furthermore, the virtual-real mapping calibration algorithm includes the following:

[0026] Feature point acquisition includes pre-setting at least 4 calibration dimples on the surface of the solid model, and the operator touching each dimple with the needle tip in turn to record the coordinates of the needle tip camera output by the binocular system.

[0027] Solving spatial transformations involves establishing rigid body transformation relationships and using four pairs of points to obtain the transformed coordinates.

[0028] Real-time mapping involves mapping the coordinates of the needle tip camera to the virtual space in each frame of operation according to the above formula, thereby driving the virtual needle tip to move synchronously.

[0029] Furthermore, the virtual-real mapping calibration algorithm includes the following:

[0030] Velocity calculation includes defining the puncture longitudinal direction as the z-axis and calculating the instantaneous velocity of the needle tip: the smoothed velocity is obtained by mean filtering through a sliding window;

[0031] Regional monitoring includes pre-marking the depth range of the ligamentum flavum layer in a virtual model, and initiating detection when the needle tip enters this range;

[0032] Breakthrough identification includes setting conditions for determining a breakthrough of the ligamentum flavum;

[0033] Feedback execution includes prompting the interface to display a message and recording the depth of the breakthrough point as a safe needle insertion boundary after the ligamentum flavum breakthrough is triggered.

[0034] According to a second aspect of the present invention, the present invention provides a method for operating an anesthesiology spinal puncture teaching system, which is implemented using an anesthesiology spinal puncture teaching system as described in the present invention, characterized in that it includes the following:

[0035] Step S1: The sampling and positioning module is used to capture the trajectory of the reflective marker ball fixed on the puncture needle and the model, as well as the puncture needle tip in three-dimensional space;

[0036] Step S2: Construct a virtual anatomical model using a virtual-real interaction module, match the model with the real object, and detect the breakthrough point of the ligamentum flavum in the real object;

[0037] Step S3: Construct a multi-dimensional quantitative evaluation system using the feedback and evaluation module, and record and analyze in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

[0038] The present invention has the following advantages:

[0039] This invention achieves a deep fusion of vision and touch. It organically integrates near-infrared binocular optical positioning technology (providing sub-millimeter-level visual navigation) with a highly realistic layered physical model (restoring the tactile sensation of real breakthroughs), thus solving the long-standing problem of VR training lacking tactile feedback and physical models lacking navigation.

[0040] This invention quantifies subjective experience. The team independently developed a multi-dimensional quantitative evaluation algorithm for the characteristics of spinal canal blockade procedures, ending the history of vague evaluation relying solely on the subjective judgment of instructors in traditional teaching, and establishing a standardized teaching evaluation system that can be replicated and promoted.

[0041] This invention implements a modular cost-reduction design concept. The core area of ​​the puncture operation adopts a replaceable modular design, which can be quickly replaced when a single module is worn out, significantly reducing the long-term teaching material costs caused by repeated punctures of physical models. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the system structure of the present invention.

[0043] Figure 2 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0044] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0045] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0046] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0047] like Figures 1 to 2 As shown, this invention proposes an anesthesiology spinal puncture teaching system and its working method, including the following:

[0048] According to a first aspect of the present invention, in one embodiment of the present invention, the present invention provides a teaching system for spinal puncture in anesthesiology, characterized in that it includes a sampling and positioning module, a virtual-real interaction module, and a feedback and evaluation module.

[0049] The sampling and positioning module uses an image acquisition module, an active light source module, and a tracking marker module to achieve sampling and positioning. The sampling and positioning includes capturing the trajectory of the reflective marker ball fixed on the puncture needle and the model and realizing the capture of the puncture needle tip in three-dimensional space.

[0050] The virtual-real interaction module includes a replaceable layered biomimetic structure module, which constructs a virtual anatomical model; the virtual-real interaction module uses a virtual-real mapping calibration algorithm to match the model with the real object; and the virtual-real interaction module uses a ligamentum flavum breakthrough detection algorithm to detect the breakthrough point of the real ligamentum flavum.

[0051] The feedback and evaluation module is used to build a multi-dimensional quantitative evaluation system, which records and analyzes in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

[0052] Furthermore, in one embodiment of the present invention, the image acquisition module is used to filter out visible light interference, including surgical shadowless lamps; wherein the image acquisition module includes an industrial camera, wherein a narrow-band infrared filter with a center wavelength of 850nm is installed in front of the lens of the industrial camera, allowing only infrared light reflected by the marker ball to enter.

[0053] Furthermore, in one embodiment of the present invention, the active light source module emits invisible infrared light in the 850nm band toward the operating area.

[0054] Furthermore, in one embodiment of the present invention, the tracking marker module samples the puncture needle handle and is rigidly connected to the positioning rigid body of each working point by a 3D printed fixture. The positioning rigid body has four spherical markers coated with a glass microsphere reflective coating, which has high reflectivity to 850nm infrared light.

[0055] Furthermore, in one embodiment of the present invention, the sampling and positioning module includes the following:

[0056] The sampling method of the sampling and localization module includes the following: For each frame of image, the gray-level centroid within the connected component Ω is extracted as the sub-pixel center coordinates of the marker point:

[0057] The positioning method of the sampling and positioning module includes the following:

[0058] Zhang's calibration method is used to obtain the camera intrinsic parameter K and the binocular extrinsic parameters R and t. Stereo calibration is completed by minimizing the reprojection error. The marker points in the left and right images are matched using epipolar geometric constraints, and mismatches are eliminated by combining the known relative positions of the marker points on the rigid body. The matched point pairs are triangulated and reconstructed to obtain the three-dimensional coordinates of the marker points. Then, singular value decomposition is used to solve the real-time pose of the needle tip.

[0059] Furthermore, in one embodiment of the present invention, the replaceable layered biomimetic structural module is manufactured using a multi-casting process; wherein the replaceable layered biomimetic structural module further includes:

[0060] The underlying structure of the latex membrane is used to simulate the ligamentum flavum;

[0061] A middle layer structure with a hollow structure and an interior coated with colorless and transparent water-based gel is used to simulate the spinal canal cavity;

[0062] The outer layer structure uses platinum-silicone to simulate skin and subcutaneous tissue.

[0063] Furthermore, in one embodiment of the present invention, the virtual anatomical model includes the following:

[0064] The virtual anatomical model is constructed based on thin-slice CT data of the human thoracolumbar region, which is then segmented by thresholding and reconstructed in three dimensions. The final result is an STL mesh model that includes the skin, ligaments, and epidural space.

[0065] Furthermore, in one embodiment of the present invention, the virtual-real mapping calibration algorithm includes the following:

[0066] Feature point acquisition includes pre-setting at least 4 calibration dimples on the surface of the solid model, and the operator touching each dimple with the needle tip in turn to record the coordinates of the needle tip camera output by the binocular system.

[0067] Solving spatial transformations involves establishing rigid body transformation relationships and using four pairs of points to obtain the transformed coordinates.

[0068] Real-time mapping involves mapping the coordinates of the needle tip camera to the virtual space in each frame of operation according to the above formula, thereby driving the virtual needle tip to move synchronously.

[0069] Furthermore, in one embodiment of the present invention, the virtual-real mapping calibration algorithm includes the following:

[0070] Velocity calculation includes defining the puncture longitudinal direction as the z-axis and calculating the instantaneous velocity of the needle tip: the smoothed velocity is obtained by mean filtering through a sliding window;

[0071] Regional monitoring includes pre-marking the depth range of the ligamentum flavum layer in a virtual model, and initiating detection when the needle tip enters this range;

[0072] Breakthrough identification includes setting conditions for determining a breakthrough of the ligamentum flavum;

[0073] Feedback execution includes displaying a prompt on the interface and recording the depth of the breakthrough point as a safe needle insertion boundary after the ligamentum flavum breakthrough is triggered. According to a second aspect of the present invention, in one embodiment, the present invention proposes a method for operating an anesthesiology spinal puncture teaching system, implemented using an anesthesiology spinal puncture teaching system as described in the present invention, characterized by including the following:

[0074] Step S1: The sampling and positioning module is used to capture the trajectory of the reflective marker ball fixed on the puncture needle and the model, as well as the puncture needle tip in three-dimensional space;

[0075] Step S2: Construct a virtual anatomical model using a virtual-real interaction module, match the model with the real object, and detect the breakthrough point of the ligamentum flavum in the real object;

[0076] Step S3: Construct a multi-dimensional quantitative evaluation system using the feedback and evaluation module, and record and analyze in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

[0077] In addition to the above, the present invention also has related embodiments, including the following:

[0078] In one embodiment of the present invention, the image acquisition module consists of two high frame rate (≥60fps) industrial cameras. A narrow-band infrared filter with a center wavelength of 850nm is installed in front of the camera lens, which only allows infrared light reflected by the marker ball to enter, effectively filtering out visible light interference such as surgical shadowless lamps and indoor fluorescent lamps.

[0079] In one embodiment of the present invention, an active light source module is coaxially mounted with a ring-shaped near-infrared LED array around each camera lens to emit invisible infrared light in the 850nm band into the operating area.

[0080] In one embodiment of the invention, the tracking marker module further includes a puncture needle handle, at which a positioning rigid body is rigidly connected via a 3D-printed fixture. This rigid body has four spherical marker dots coated with a glass microsphere reflective coating. The coating has high reflectivity to 850nm infrared light.

[0081] In one embodiment of the present invention, the sampling and positioning module further includes the following:

[0082] The sampling methods include the following:

[0083] The left and right cameras synchronously acquire grayscale images at a frequency of no less than 60fps. For each frame, the grayscale centroid within the connected component Ω is extracted as the sub-pixel center coordinates of the marker point.

[0084] ; ;

[0085] Where I(x,y) is the pixel grayscale value.

[0086] The location methods include the following:

[0087] Zhang's calibration method was used to obtain the camera's intrinsic parameters K and the binocular extrinsic parameters R and t. Stereo calibration was completed by minimizing the reprojection error. Epipolar geometric constraints were used to match marker points in the left and right images, and mismatches were eliminated by combining the known relative positions of marker points on the rigid body. Triangulation reconstruction was performed on the matched point pairs to obtain the 3D coordinates of the marker points. Then, singular value decomposition (SVD) was used to solve for the real-time pose of the needle tip, i.e., minimizing:

[0088] ;

[0089] in The a priori calibration coordinates of the needle tip in the local coordinate system of the rigid body are given.

[0090] In one embodiment of the present invention, the sampling and positioning module fixes the positioning rigid body to a precision displacement platform (repeat positioning accuracy ≤ 0.02 mm), selects 25 verification points in an operating space of 300×300×300 mm³, and measures the root mean square error of positioning to be ≤ 0.2 mm.

[0091] In one embodiment of the present invention, the virtual-real interaction module further includes the following:

[0092] The virtual anatomical model is based on thin-slice CT data of the human thoracolumbar region. Through threshold segmentation and 3D reconstruction, an STL mesh model containing structures such as skin, ligaments, and epidural space is generated. The physical training model is made using a layered mold consistent with the virtual model. The key ligament layer uses a highly elastic material to simulate resistance changes, and the epidural space is reserved with gaps to ensure a sense of non-vibration. The puncture area is designed as a replaceable module, which is connected to the overall model by clips.

[0093] In one embodiment of the present invention, the virtual-real interaction module further includes the following:

[0094] The virtual-real mapping calibration algorithm also includes the following:

[0095] Step A: Feature point acquisition, including the following: Pre-set at least 4 calibration dimples on the surface of the solid model. The operator touches each dimple with the needle tip in sequence and records the needle tip camera coordinates output by the binocular system. .

[0096] The virtual model has already defined corresponding reference points. .

[0097] Step B: Solve the spatial transformation, including the following: Establish rigid body transformation relationships: ;

[0098] Using four pairs of points to obtain SVD and .

[0099] Step C: Real-time mapping, including the following: During operation, the coordinates of the needle tip camera are mapped to the virtual space according to the above formula, driving the virtual needle tip to move synchronously.

[0100] In one embodiment of the present invention, after the virtual-real interaction module completes calibration, 10 points are randomly selected on the model surface, and a displacement platform is used to drive the needle tip to precisely touch them, recording the deviation displayed in the virtual space. The measured maximum positional deviation is ≤0.7mm.

[0101] In one embodiment of the present invention, the virtual-real interaction module further includes the following:

[0102] The ligamentum flavum breakthrough detection algorithm also includes the following: Step A: Velocity calculation, including the following: Define the puncture longitudinal direction as the z-axis, and the instantaneous velocity of the needle tip:

[0103] ;

[0104] Smooth speed obtained by sliding window mean filtering (A horizontal line above the "V"). The window width can be 5 frames.

[0105] Step B: Area monitoring, including the following:

[0106] Pre-mark the depth range of the ligamentum flavum layer in the virtual model. Detection is initiated when the needle tip enters this range.

[0107] Step C: Breakthrough identification, including the following: setting a speed threshold (Typical value 10~15 mm / s). A ligamentum flavum breach is determined when both of the following conditions are met simultaneously: The peak velocity dropped to below 100ms within 100ms. ;

[0108] Step D: Feedback execution, including the following:

[0109] After the event is triggered, a prompt will pop up on the screen, a sound effect of the needle falling will play, and the depth of the breakthrough point will be recorded as the safe needle insertion boundary.

[0110] In one embodiment of the present invention, several anesthesiologists each perform 10 punctures on the system. The algorithm automatically detects breakthroughs, while the physicians manually record the gold standard by pressing buttons. The statistical accuracy rate is ≥96%, with no false alarms.

[0111] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A teaching system for spinal puncture in anesthesiology, characterized in that, It includes a sampling and positioning module, a virtual-real interaction module, and a feedback and evaluation module; The sampling and positioning module uses an image acquisition module, an active light source module, and a tracking marker module to achieve sampling and positioning. The sampling and positioning includes capturing the trajectory of the reflective marker ball fixed on the puncture needle and the model and realizing the capture of the puncture needle tip in three-dimensional space. The virtual-real interaction module includes a replaceable layered biomimetic structure module, which constructs a virtual anatomical model; the virtual-real interaction module uses a virtual-real mapping calibration algorithm to match the model with the real object; and the virtual-real interaction module uses a ligamentum flavum breakthrough detection algorithm to detect the breakthrough point of the real ligamentum flavum. The feedback and evaluation module is used to build a multi-dimensional quantitative evaluation system, which records and analyzes in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

2. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The image acquisition module is used to filter out visible light interference, including surgical shadowless lamps; the image acquisition module includes an industrial camera, in which a narrow-band infrared filter with a center wavelength of 850nm is installed in front of the lens of the industrial camera, allowing only the infrared light reflected by the marker ball to enter.

3. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The active light source module emits invisible infrared light in the 850nm band into the operating area.

4. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The tracking and marking module samples the puncture needle handle and is rigidly connected to the positioning rigid body of each working point through a 3D printed fixture. The positioning rigid body has four spherical marking points coated with a glass microsphere reflective coating, which has high reflectivity to 850nm infrared light.

5. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The sampling and positioning module includes the following: The sampling method of the sampling and localization module includes the following: For each frame of image, extract the gray-level centroid within the connected component as the sub-pixel center coordinates of the marker point: The positioning method of the sampling and positioning module includes the following: Zhang's calibration method is used to obtain the camera intrinsic parameter K and the binocular extrinsic parameters R and t. Stereo calibration is completed by minimizing the reprojection error. The marker points in the left and right images are matched using epipolar geometric constraints, and mismatches are eliminated by combining the known relative positions of the marker points on the rigid body. The matched point pairs are triangulated and reconstructed to obtain the three-dimensional coordinates of the marker points. Then, singular value decomposition is used to solve the real-time pose of the needle tip.

6. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The replaceable layered biomimetic structural modules are manufactured using a multi-casting process. The replaceable layered biomimetic structure module also includes: The underlying structure of the latex membrane is used to simulate the ligamentum flavum; A middle layer structure with a hollow structure and an interior coated with colorless and transparent water-based gel is used to simulate the spinal canal cavity; The outer layer structure uses platinum-silicone to simulate skin and subcutaneous tissue.

7. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The virtual anatomical model includes the following: The virtual anatomical model is constructed based on thin-slice CT data of the human thoracolumbar region, which is then segmented by thresholding and reconstructed in three dimensions. The final result is an STL mesh model that includes the skin, ligaments, and epidural space.

8. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The virtual-real mapping calibration algorithm includes the following: Feature point acquisition includes pre-setting at least 4 calibration dimples on the surface of the solid model, and the operator touching each dimple with the needle tip in turn to record the coordinates of the needle tip camera output by the binocular system. Solving spatial transformations involves establishing rigid body transformation relationships and using four pairs of points to obtain the transformed coordinates. Real-time mapping involves mapping the coordinates of the needle tip camera to the virtual space in each frame of operation according to the above formula, thereby driving the virtual needle tip to move synchronously.

9. The anesthesiology spinal puncture teaching system according to claim 1, characterized in that, The virtual-real mapping calibration algorithm includes the following: Velocity calculation includes defining the puncture longitudinal direction as the z-axis and calculating the instantaneous velocity of the needle tip: the smoothed velocity is obtained by mean filtering through a sliding window; Regional monitoring includes pre-marking the depth range of the ligamentum flavum layer in a virtual model, and initiating detection when the needle tip enters this range; Breakthrough identification includes setting conditions for determining a breakthrough of the ligamentum flavum; Feedback execution includes prompting the interface to display a message and recording the depth of the breakthrough point as a safe needle insertion boundary after the ligamentum flavum breakthrough is triggered.

10. A method for operating an anesthesiology spinal puncture teaching system, implemented using the anesthesiology spinal puncture teaching system as described in claims 1 to 9, characterized in that, Includes the following: Step S1: The sampling and positioning module is used to capture the trajectory of the reflective marker ball fixed on the puncture needle and the model, as well as the puncture needle tip in three-dimensional space; Step S2: Construct a virtual anatomical model using a virtual-real interaction module, match the model with the real object, and detect the breakthrough point of the ligamentum flavum in the real object; Step S3: Construct a multi-dimensional quantitative evaluation system using the feedback and evaluation module, and record and analyze in real time indicators including positioning accuracy deviation, needle angle drift, total operation time, and path tortuosity.

Images