AI recognition-based automatic orthopedic rapid positioning bone drill robot and method

Abstract

This invention relates to the field of smart medical technology and discloses an automated orthopedic rapid positioning bone drill robot and method based on AI recognition. The robot includes: a mechanical fixation module comprising an upper fixation block and a lower fixation block connected by a hinged positioning post, the positioning post being adjustable to clamp the surgical site; an X-ray image acquisition interface for acquiring intraoperative X-ray fluoroscopy image data; a bone drill module, integrated with the mechanical fixation block, employing a 5-axis adjustment structure for performing drilling operations; an AI recognition unit for recognizing bone tissue features based on X-ray image data and calculating key drilling parameters; a motor control unit for driving the bone drill module; and a controller electrically connected to the AI ​​recognition unit, motor control unit, and bone drill module. This invention automates the entire orthopedic drilling process, improving positioning accuracy and surgical safety through precise calculation of drilling parameters, and is applicable to various orthopedic surgical scenarios such as limb and spine surgeries.

Description

Technical Field

[0001] This invention belongs to the field of smart medical technology, specifically relating to an AI-based automated orthopedic rapid positioning bone drill robot and method. Background Technology

[0002] In clinical medicine, the precise and rapid positioning of the bone drill during orthopedic surgery is a crucial step in ensuring surgical outcomes and patient safety. Whether it's fracture reduction and fixation, spinal fusion, or artificial joint replacement, all surgeries require drilling into specific bone tissue locations using a bone drill. The accuracy of this positioning directly affects the stability of internal fixation devices, the degree of surgical trauma, and postoperative recovery. Positioning errors can lead to serious complications such as nerve and blood vessel damage, loosening of internal fixation, and bone fragmentation. Therefore, achieving rapid and precise positioning of the bone drill is of great significance for improving the efficiency and safety of orthopedic surgeries.

[0003] Currently, bone drill positioning in orthopedic surgery mainly relies on two methods: one is for doctors to manually adjust the position of the bone drill based on clinical experience and intraoperative X-ray fluoroscopy. This method is greatly affected by the difference in the operator's skill level, and the positioning accuracy is unstable (the error is usually 2-5mm). Moreover, multiple fluoroscopy increases the radiation exposure risk for both doctors and patients. The other method is to use traditional navigation systems for positioning. These systems usually require CT or MRI scans to build a three-dimensional model before surgery, and then match it with real-time images during surgery. The operation is complicated, the preparation time is long, and the equipment is bulky and expensive, making it difficult to popularize in primary healthcare institutions. At the same time, the navigation accuracy is easily affected when bone tissue shifts during surgery, and dynamic and rapid adjustment cannot be achieved.

[0004] Furthermore, while some existing positioning systems incorporate robotic arm assistance, they mostly employ a 3-axis adjustment structure, enabling only simple translational movements and failing to adapt to the complex spatial angle requirements of bones. Moreover, they lack precise calculation logic for key drilling parameters (insertion position, exit position, and insertion depth), relying heavily on preoperative planning or manual judgment, and cannot dynamically adjust based on real-time intraoperative image data, thus limiting surgical adaptability and accuracy. Given the problems of insufficient accuracy, complex operation, high radiation risk, and limited adaptability of existing bone drill positioning methods, there is an urgent need for a solution that integrates intelligent recognition and multi-axis automated control technology. This solution should accurately acquire image data and calculate drilling parameters to achieve rapid and precise positioning of the bone drill in orthopedic surgery, reducing surgical trauma and complications, and improving surgical efficiency and adaptability. Summary of the Invention

[0005] The present invention aims to provide an automated orthopedic rapid positioning bone drill robot and method based on AI recognition, in order to solve the problems of low accuracy, complex operation and inability to dynamically adapt to bone status in existing orthopedic drilling positioning that rely on experience or traditional navigation.

[0006] To solve the above problems, the present invention adopts the following technical solution: Option 1: An AI-based automated orthopedic rapid positioning bone drill robot, including: The mechanical fixation module includes an upper fixation block and a lower fixation block, which are connected by a hinged positioning post. The positioning post can adjust the distance between the two to clamp the surgical site. Limiting holes are provided on both the upper and lower fixation blocks. The X-ray image acquisition interface is used to acquire intraoperative X-ray fluoroscopic image data, which includes the contour, thickness and spatial location information of the bone tissue at the surgical site. The bone drill module, connected to the mechanical fixation block as a whole, adopts a 5-axis adjustment structure for performing drilling operations. The 5-axis adjustment structure includes X, Y, and Z axis translation and A and B axis rotation. The bone drill module includes a drill bit that drills in from the insertion position in the upper fixing block limiting hole and extends along a preset path to the exit position in the lower fixing block limiting hole. The AI ​​recognition unit is used to process X-ray image data, identify the coordinates of core feature points of bone tissue, calculate the needle insertion position, needle exit position and needle insertion depth, and recommend the drilling rotation speed of the bone drill module. The motor control unit, including a stepper motor and drive circuit, is used to drive the bone drill module to move along five axes. The controller is electrically connected to the AI ​​recognition unit, the motor control unit, and the bone drill module, respectively, to control the equipment to work together.

[0007] Beneficial effects: The 5-axis adjustable structure enables multi-dimensional motion, adapting to the complex spatial angle requirements of the skeleton; the AI ​​recognition unit accurately calculates key drilling parameters based on real-time image data, combined with the limiting constraints of the mechanical fixation module, ensuring precise and controllable drilling path, and improving surgical safety and efficiency.

[0008] Preferably, the adjustment range of the positioning column is 0-20cm, and the spacing can be locked by the locking knob after adjustment. After locking, the relative displacement between the upper fixing block and the lower fixing block is ≤0.1mm, and the vertical spacing H between the upper fixing block and the lower fixing block is fed back to the AI ​​recognition unit in real time.

[0009] Beneficial effects: Wide range of adjustment to fit different limb thicknesses (such as palm, waist, thigh, etc.), locking function to ensure stable fixation during surgery and avoid drilling deviation caused by displacement; real-time spacing feedback provides accurate basic data for AI calculation.

[0010] Preferably, the AI ​​recognition unit extracts the bone tissue contour from the X-ray image data using the Canny edge detection algorithm, determines the bone tissue central axis L and the bone density distribution of the target drilling area, and, combined with the vertical distance H between the upper and lower fixation blocks and the bone tissue projection angle α, calculates the needle insertion position (X1, Y1, Z1), needle exit position (X2, Y2, Z2), and needle insertion depth D through three-dimensional coordinate transformation. The specific calculation formula is as follows: Needle insertion position X1 = midpoint of the projection of the bone tissue central axis L in the X direction + ΔX. Needle insertion position Y1 = midpoint of the projection of the bone tissue central axis L in the Y direction + ΔY. The needle insertion position Z1 = the Z-coordinate of the center of the limiting hole of the upper fixed block, and the needle exit position X2 = X1 + H × sinα. Needle exit position Y2 = Y1 + H × cosα, The needle exit position Z2 = the Z-coordinate of the center of the limiting hole of the lower fixed block. Needle depth D= ; The drilling rotation speed ranges from 800 to 3500 r / min, and is dynamically adjusted according to bone density.

[0011] Beneficial effects: Through mature edge detection algorithms and three-dimensional coordinate transformation models, key drilling parameters can be quantitatively calculated, avoiding errors in human judgment; by combining bone density distribution with needle insertion position and rotation speed, personalized drilling can be achieved, reducing bone tissue damage.

[0012] Preferably, the stepper motor of the motor control unit has a positioning accuracy of ≤0.05mm and a response time of ≤50ms. It can drive the bone drill module to translate along the X, Y, and Z axes and rotate along the A and B axes, with a rotation angle range of ±30°, ensuring that the coaxiality deviation between the drill bit axis and the path from the needle inlet position to the needle outlet position is ≤0.1°.

[0013] Beneficial effects: The high-precision, fast-response 5-axis drive system ensures precise adaptation of the bone drill in terms of spatial angle and position, breaking through the motion limitations of the traditional 3-axis structure and adapting to the drilling needs of complex orthopedic scenarios such as the spine and long bones.

[0014] Preferably, the gap between the diameter of the drill bit of the bone drill module and the diameter of the limiting hole of the upper and lower fixing blocks is ≤0.2mm. The drill bit is made of medical stainless steel or titanium alloy, and its surface is sterile. The length of the drill bit is adapted to the needle insertion depth range of 0-50mm.

[0015] Beneficial effects: The small gap prevents drill bit deviation during drilling, and medical-grade materials and sterile treatment ensure surgical safety; it adapts to different needle insertion depth requirements, expanding the scope of surgical applications.

[0016] Preferably, it also includes a data storage unit for storing X-ray image data, drilling parameters calculated by the AI ​​recognition unit, and surgical procedure data.

[0017] Beneficial effects: The data storage unit supports postoperative traceability and model optimization, enabling full-process recording of surgical data and providing a basis for postoperative review and complication analysis. At the same time, the accumulated data can be used to optimize the AI ​​recognition model and continuously improve positioning accuracy.

[0018] Option 2: An automated orthopedic rapid positioning bone drill method based on AI recognition, using the robot described above, includes the following steps: S1. Based on the thickness of the surgical site, adjust the distance between the upper and lower fixing blocks of the mechanical fixation module using the positioning column to clamp and lock the surgical site, and record the vertical distance H between them; S2. Acquire X-ray fluoroscopic image data containing bone tissue of the surgical site through the X-ray image acquisition interface; The S3.AI recognition unit enhances X-ray image data, extracts bone tissue contours and feature point coordinates, and calculates the needle insertion position, needle exit position and needle insertion depth by combining the vertical spacing H and bone tissue projection angle α, and recommends the drilling rotation speed. S4. Based on the needle entry position, needle exit position, and drilling path, the controller drives the 5-axis structure of the bone drill module through the motor control unit to calibrate the consistency between the drill bit axis and the preset path, so that the drill bit is aligned with the needle entry position. S5. The bone drill module is started. The drill bit drills into the upper fixed block limit hole along the preset path and completes the drilling at the needle exit position.

[0019] Beneficial effects: The fully automated process of image acquisition, AI calculation, 5-axis positioning and automatic drilling replaces manual alignment and judgment, improving surgical efficiency; the drilling parameters are dynamically calculated based on real-time image data to ensure positioning accuracy and adaptation to bone condition.

[0020] Preferably, in step S3, the processing of X-ray image data by the AI ​​recognition unit includes: using a Frangi filter to enhance the contrast between bone tissue and surrounding tissue, locating the central axis L of bone tissue through Hough transform, analyzing bone density distribution based on grayscale threshold segmentation method, where grayscale value > 200 is a high-density area, grayscale value 100-200 is a medium-density area, and grayscale value < 100 is a low-density area, and the needle insertion position is preferentially selected in medium and high-density areas and avoids areas with bone tissue edges ≥ 2mm.

[0021] Beneficial effects: By using multiple algorithms to collaboratively process image data, the accuracy and stability of bone tissue feature recognition are improved; the needle insertion position selection logic takes into account both drilling stability and bone tissue protection, reducing the risk of complications.

[0022] Preferably, in step S4, when the controller drives the bone drill module to move, the drill bit angle is first calibrated by rotating the A and B axes so that the deviation between the drill bit axis and the path from the needle entry position to the needle exit position is ≤0.1°. Then, the drill bit is precisely positioned to the needle entry position by translating the X, Y, and Z axes, with a positioning time of ≤3 seconds.

[0023] Beneficial effects: The motion logic of first calibrating the angle and then positioning ensures that the drill bit is completely aligned with the preset path, avoiding drilling deviation caused by angle deviation; rapid positioning shortens the operation time and reduces the risk of anesthesia and radiation exposure for patients.

[0024] Preferably, in step S3, the drilling rotation speed set in the AI ​​recognition unit is: 2500-3500 r / min in the high-density area of ​​bone tissue, 1500-2500 r / min in the medium-density area, and 800-1500 r / min in the low-density area. For every 10 mm increase in needle depth, the rotation speed adaptability is increased by 5%-10% to compensate for changes in bone tissue resistance.

[0025] Beneficial effects: Based on the dual-dimensional speed adjustment of bone density and needle depth, drilling efficiency is ensured while minimizing mechanical damage to bone tissue and improving postoperative recovery.

[0026] The working principle and advantages of this invention: Working principle: This invention achieves rapid and accurate positioning of orthopedic bone drills through a collaborative process of mechanical fixation, image acquisition, AI calculation, 5-axis positioning, and automatic drilling. 1. The mechanical fixation module clamps the surgical site with adjustable upper and lower fixing blocks, and uses limiting holes to constrain the movement trajectory of the drill bit to ensure drilling stability. At the same time, it records the spacing H of the fixing blocks and feeds it back to the AI ​​recognition unit. 2. The X-ray image acquisition interface acquires real-time fluoroscopic image data containing bone tissue, providing complete spatial information of bone tissue for AI calculation; 3. The AI ​​recognition unit processes the image data using multiple algorithms to extract the bone tissue contour, central axis, and bone density distribution. Combining the spacing H and the bone projection angle α, it accurately calculates the needle insertion position, needle exit position, and needle insertion depth through a three-dimensional coordinate transformation model. At the same time, it recommends an appropriate rotation speed based on bone density. 4. The controller receives the AI ​​calculation results and drives the 5-axis bone drill module through the motor control unit. First, it calibrates the drill bit angle by rotating along the A and B axes, and then it moves and positions the drill bit to the insertion position by translating along the X, Y, and Z axes to ensure that the drill bit is coaxial with the preset path. 5. The bone drill module starts and completes drilling along the preset path from the needle inlet to the needle outlet, without the need for manual intervention.

[0027] advantage: 1. High positioning accuracy: By using AI to quantitatively calculate key drilling parameters, combined with 5-axis high-precision drive and limit hole constraints, the drilling path deviation is ≤0.1mm and the needle depth error is ≤0.3mm, which is far superior to traditional manual positioning (error 2-5mm). 2. Strong adaptability to movement: The 5-axis adjustment structure (3 translations, 2 rotations) breaks through the limitations of the traditional 3-axis structure and can adapt to complex spatial angle requirements such as the pedicle abduction angle of the spine and oblique drilling of long bones. Its application range covers a variety of orthopedic scenarios such as limbs and spine. 3. High efficiency: The entire process from image acquisition to drilling completion is automated, with a single hole operation time of ≤15 seconds, which is significantly shorter than traditional methods. Moreover, only one X-ray image acquisition is required, effectively reducing radiation exposure. 4. High safety: Based on dynamic adjustment of rotation speed and needle insertion position according to bone density, combined with the small gap constraint between the drill bit and the limiting hole, the risk of complications such as bone tissue splitting and nerve and blood vessel damage is significantly reduced; 5. Data traceability: The entire surgical process is recorded through data storage units, supporting postoperative review and AI model iteration and optimization to continuously improve surgical accuracy and adaptability. Attached Figure Description

[0028] Figure 1 This is a logic block diagram of the robot of the present invention.

[0029] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0030] The following detailed description illustrates the specific implementation method: The AI-based automated orthopedic rapid positioning bone drill robot of this invention, such as... Figure 1 As shown, it includes a controller and a mechanical fixation module, an X-ray image acquisition interface, a bone drill module, an AI recognition unit, a motor control unit, and a data storage unit, all connected to the controller.

[0031] The mechanical fixation module includes an upper fixation block and a lower fixation block, which are connected by a hinged positioning post. The positioning post can adjust the distance between the two blocks to clamp the surgical site. The adjustment range of the positioning post is 0-20cm, and the distance can be locked by a locking knob after adjustment. After locking, the relative displacement between the upper and lower fixation blocks is ≤0.1mm. The vertical distance H between the upper and lower fixation blocks is fed back to the AI ​​recognition unit in real time. Both the upper and lower fixation blocks have limit holes to constrain the movement trajectory of the drill bit.

[0032] The X-ray image acquisition interface is used to acquire intraoperative X-ray fluoroscopic image data. The image data includes the contour, thickness and spatial location information of the bone tissue at the surgical site. It uses a Dexela 1211 digital X-ray detector with a pixel size of 100μm and transmits data quickly through a USB 3.0 interface to ensure image clarity and transmission efficiency.

[0033] The bone drill module, connected to the mechanical fixation block as a whole, adopts a 5-axis adjustment structure (X, Y, Z axis translation and A, B axis rotation) for performing drilling operations. The bone drill module includes a drill bit, which drills into the upper fixation block from the insertion position in the limiting hole and extends along a preset path to the exit position in the lower fixation block from the limiting hole. The gap between the diameter of the drill bit and the diameter of the limiting holes of the upper and lower fixation blocks is ≤0.2mm. The drill bit is made of medical-grade stainless steel, and its surface is sterilized. The length of the drill bit is suitable for a needle insertion depth range of 0-50mm.

[0034] The AI ​​recognition unit, with NVIDIA Jetson Xavier NX as its core processor and equipped with a customized deep learning model, is used to process X-ray image data: it uses a Frangi filter to enhance bone tissue contrast, extracts bone tissue contours through the Canny edge detection algorithm, locates the central axis L of bone tissue using Hough transform, analyzes bone density distribution based on grayscale threshold segmentation, and calculates the needle insertion position, needle exit position, and needle insertion depth by combining the vertical spacing H and the bone tissue projection angle α, and recommends the drilling rotation speed.

[0035] The motor control unit, including the TMC2209 stepper motor driver and the STM32 microcontroller, has a stepper motor positioning accuracy of ≤0.05mm and a response time of ≤50ms. It can drive the bone drill module to translate along the X, Y, and Z axes and rotate along the A and B axes (rotation angle range ±30°), ensuring that the coaxiality deviation between the drill bit axis and the preset path is ≤0.1°.

[0036] The controller uses an Advantech IPC-610L industrial computer, equipped with an Intel Core i7 processor, and is electrically connected to the AI ​​recognition unit, motor control unit, and bone drill module to control the equipment to work together.

[0037] The data storage unit uses a Samsung 870 EVO 500GB solid-state drive to store X-ray image data, drilling parameters calculated by the AI ​​recognition unit, and surgical procedure data, supporting postoperative traceability and model optimization.

[0038] like Figure 2 As shown, the steps involved in rapid bone drilling in orthopedics using the above robot include: S1. Based on the thickness of the surgical site, adjust the distance between the upper and lower fixing blocks of the mechanical fixation module using the positioning column to clamp and lock the surgical site, and record the vertical distance H between them; S2. Acquire X-ray fluoroscopic image data containing bone tissue of the surgical site through the X-ray image acquisition interface; The S3.AI recognition unit enhances X-ray image data, extracts bone tissue contours and feature point coordinates, and calculates the needle insertion position, needle exit position and needle insertion depth by combining the vertical spacing H and bone tissue projection angle α, and recommends the drilling rotation speed. S4. Based on the needle entry position, needle exit position, and drilling path, the controller drives the 5-axis structure of the bone drill module through the motor control unit. First, the drill bit angle is calibrated by rotating the A and B axes to ensure that the deviation between the drill bit axis and the preset path is ≤0.1°. Then, the drill bit is precisely positioned to the needle entry position by translating the X, Y, and Z axes. S5. The bone drill module is started. The drill bit drills into the upper fixed block limit hole along the preset path and completes the drilling at the needle exit position.

[0039] In step S3, when the AI ​​recognition unit identifies the coordinates of bone tissue feature points, it uses the edge of the bone tissue contour as a reference, with an identification accuracy of ≤0.1mm to ensure the accuracy of the calculation parameters.

[0040] In step S3, the drilling rotation speed set in the AI ​​recognition unit is as follows: 2500-3500 r / min is recommended for high-density bone tissue (grayscale value > 200), 1500-2500 r / min is recommended for medium-density bone tissue (grayscale value 100-200), and 800-1500 r / min is recommended for low-density bone tissue (grayscale value < 100). For every 10 mm increase in needle depth, the rotation speed adaptability increases by 5%-10%.

[0041] This method combines image acquisition with AI calculation to achieve dynamic and precise planning of drilling parameters. With 5-axis automated drive, it ensures stable and controllable drilling throughout the process, improving surgical efficiency and safety.

[0042] Example 1: Metacarpal Fracture Screw Fixation (Limb Orthopedics Scenario) Parameter settings: Target bone: Fracture end of the 3rd metacarpal bone, requiring drilling holes on both sides of the fracture line to insert two 2.5mm diameter screws; Mechanical fixation module: The distance between the upper and lower fixation blocks is adjusted to 5cm (adapting to the thickness of the palm), the limiting hole diameter is 2.8mm, and the distance H=50mm after locking; X-ray image data: One intraoperative anteroposterior fluoroscopy, resolution 1024×1024 pixels, clearly showing the metacarpal contour and fracture line; AI recognition unit: Extracting the metacarpal central axis L and bone tissue projection. Angle α=0° (horizontal state), bone density gray value 180 (medium density zone), calculate needle insertion position (X1=25mm, Y1=15mm, Z1=10mm), needle exit position (X2=25mm, Y2=15mm, Z2=60mm), needle depth D=50mm, recommended drilling speed 1800r / min; motor control: drive bone drill module A and B axes to rotate 0°, X, Y, and Z axes to translate and position to needle insertion position, positioning time 2 seconds.

[0043] Implementation process: Adjust the mechanical fixation module to clamp the palm, ensuring the metacarpals are in a horizontal and stable state, and record the interval H=50mm; The X-ray image acquisition interface acquires metacarpal anteroposterior fluoroscopic image data and transmits it to the AI ​​recognition unit. The AI ​​recognition unit processes the image data and calculates the needle insertion position, needle exit position, and needle insertion depth. The recommended rotation speed is 1800 r / min. After the controller drives the 5-axis bone drill module to calibrate the angle, it is positioned at the needle insertion position; The bone drill rotates at 1800 r / min and completes drilling of two screw holes in 8 seconds per hole.

[0044] Results: The perpendicularity error between the drilling axis and the long axis of the metacarpal bone was ≤0.8°, the needle insertion depth error was ≤0.2mm, the fracture ends were well aligned after screw placement, the operation time was shortened to 10 minutes compared to the traditional method (25 minutes), and there were no complications such as metacarpal splitting and nerve damage.

[0045] Example 2: Lumbar pedicle screw placement (spinal orthopedics scenario) Parameter settings: Target bone: L4 pedicle, requiring drilling and insertion of 3.5mm diameter pedicle screws; Mechanical fixation module: Adjust the distance between the upper and lower fixation blocks to 12cm (adapting to lumbar thickness), limiting hole diameter 3.8mm, locking distance H=120mm; X-ray imaging data: Intraoperative lateral fluoroscopy, resolution 1024×1024 pixels, clearly showing the pedicle axis and surrounding tissues; AI recognition unit: Extract the pedicle central axis L, bone tissue projection angle α=15° (abduction angle), bone density gray value 220 (high density area), calculate the needle insertion position (X1=40). mm, Y1=30mm, Z1=15mm), exit position (X2=40+120×sin15°≈71.06mm, Y2=30+120×cos15°≈146.39mm, Z2=135mm), needle depth D=√[(31.06)²+(116.39)²+(120)²]≈175.2mm, recommended drilling speed 2800r / min; motor control: drive bone drill module A axis rotate 15°, B axis rotate 0°, after calibrating drill bit angle, X, Y, Z axis translate to position to needle entry position, positioning time 3 seconds.

[0046] Implementation process: The mechanical fixation module stabilizes the waist to prevent intraoperative positional movement, and the recording interval H=120mm; The X-ray image acquisition interface acquires lateral fluoroscopic image data of the L4 pedicle; The AI ​​recognition unit processes the image data and calculates the drilling parameters by combining the pedicle abduction angle, recommending a rotation speed of 2800 r / min. The controller drives the 5-axis bone drill module to first calibrate the 15° abduction angle, and then position it to the needle insertion position; The bone drill rotates at 2800 r / min and completes drilling on one side in 12 seconds.

[0047] Results: The drilling process did not break through the pedicle cortex (the breakthrough rate of traditional manual drilling is about 12%), the screw placement accuracy was 100%, the needle depth error was ≤0.3mm, the operation time was shortened to 15 minutes compared with the traditional navigation method (40 minutes), and the patient's intraoperative radiation exposure was reduced by 70% (only 1 fluoroscopy).

[0048] Existing technologies mostly employ a 3-axis translation structure, which can only achieve simple position adjustments. This invention uses a 5-axis adjustment structure (3 translations, 2 rotations), which can adapt to complex spatial angle requirements such as spinal pedicle abduction angle and oblique drilling of long bones, significantly improving motion adaptability. Existing technologies lack quantitative calculation of key drilling parameters, relying mostly on preoperative planning or manual judgment. This invention integrates edge detection, 3D coordinate transformation, and other algorithms through an AI recognition unit, dynamically calculating the needle insertion position, needle exit position, and needle insertion depth based on real-time image data, achieving precise parameter quantification. Existing technologies mostly rely on image detection and manual alignment, or preoperative images and intraoperative matching. This invention clearly defines the core as image data acquisition and real-time AI calculation, eliminating the need for preoperative 3D modeling. Dynamic parameter adjustment can be completed with only one intraoperative image acquisition, making the operation more efficient and reducing radiation.

[0049] Traditionally, precise drilling in orthopedics relies on preoperative CT 3D navigation and 3-axis robotic arm positioning. However, this invention breaks through this technological bias by innovatively combining a 5-axis motion structure with real-time AI calculations. It uses 2D X-ray image data to deduce 3D drilling parameters, eliminating the need for complex 3D reconstruction and solving the problem of mismatch between preoperative planning and intraoperative bone displacement in existing technologies. It establishes a multi-dimensional adaptation model for bone density, needle insertion position, rotation speed, and Pfizer needle insertion depth. This is not a simple superposition of existing fixation devices and image recognition, but rather a breakthrough in both accuracy and efficiency achieved through parameter closed-loop control. The introduction of the 5-axis structure is not simply an increase in motion dimensions, but a targeted design for the complex spatial angles of orthopedic bones. Combined with limiting hole constraints and AI angle calibration, it achieves high-precision positioning in complex scenarios. Its technological combination is significantly innovative.

[0050] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. An automated orthopedic rapid positioning bone drill robot based on AI recognition, characterized in that, include: The mechanical fixation module includes an upper fixation block and a lower fixation block, which are connected by a hinged positioning post. The positioning post can adjust the distance between the two to clamp the surgical site. Limiting holes are provided on both the upper and lower fixation blocks. The X-ray image acquisition interface is used to acquire intraoperative X-ray fluoroscopic image data, which includes the contour, thickness and spatial location information of the bone tissue at the surgical site. The bone drill module, connected to the mechanical fixation block as a whole, adopts a 5-axis adjustment structure for performing drilling operations. The 5-axis adjustment structure includes X, Y, and Z axis translation and A and B axis rotation. The bone drill module includes a drill bit that drills in from the insertion position in the upper fixing block limiting hole and extends along a preset path to the exit position in the lower fixing block limiting hole. The AI ​​recognition unit is used to process X-ray image data, identify the coordinates of core feature points of bone tissue, calculate the needle insertion position, needle exit position and needle insertion depth, and recommend the drilling rotation speed of the bone drill module. The motor control unit, including a stepper motor and drive circuit, is used to drive the bone drill module to move along five axes. The controller is electrically connected to the AI ​​recognition unit, the motor control unit, and the bone drill module, respectively, to control the equipment to work together.

2. The AI-based automated orthopedic rapid positioning bone drill robot according to claim 1, characterized in that, The positioning column can be adjusted from 0 to 20 cm. After adjustment, the spacing can be locked by the locking knob. After locking, the relative displacement between the upper and lower fixing blocks is ≤0.1 mm. The vertical spacing H between the upper and lower fixing blocks is fed back to the AI ​​recognition unit in real time.

3. The AI-based automated orthopedic rapid positioning bone drill robot according to claim 1, characterized in that, The AI ​​recognition unit extracts the bone tissue contour from the X-ray image data using the Canny edge detection algorithm, determines the bone tissue central axis L and the bone density distribution in the target drilling area, and, combined with the vertical distance H between the upper and lower fixation blocks and the bone tissue projection angle α, calculates the needle insertion position (X1, Y1, Z1), needle exit position (X2, Y2, Z2), and needle insertion depth D through three-dimensional coordinate transformation. The specific calculation formula is as follows: Needle insertion position X1 = midpoint of the projection of the bone tissue central axis L in the X direction + ΔX. Needle insertion position Y1 = midpoint of the projection of the bone tissue central axis L in the Y direction + ΔY. The needle insertion position Z1 = the Z-coordinate of the center of the limiting hole of the upper fixed block, and the needle exit position X2 = X1 + H × sinα. Needle exit position Y2 = Y1 + H × cosα, The needle exit position Z2 = the Z-coordinate of the center of the limiting hole of the lower fixed block. Needle depth D= ; The drilling rotation speed ranges from 800 to 3500 r / min, and is dynamically adjusted according to bone density.

4. The AI-based automated orthopedic rapid positioning bone drill robot according to claim 1, characterized in that, The stepper motor of the motor control unit has a positioning accuracy of ≤0.05mm and a response time of ≤50ms. It can drive the bone drill module to translate along the X, Y, and Z axes and rotate along the A and B axes with a rotation angle range of ±30°, ensuring that the coaxiality deviation between the drill bit axis and the path from the needle inlet position to the needle outlet position is ≤0.1°.

5. The AI-based automated orthopedic rapid positioning bone drill robot according to claim 1, characterized in that, The gap between the diameter of the drill bit of the bone drill module and the diameter of the limiting hole of the upper and lower fixing blocks is ≤0.2mm. The drill bit is made of medical stainless steel or titanium alloy, and its surface is sterile. The length of the drill bit is suitable for a needle insertion depth range of 0-50mm.

6. The AI-based automated orthopedic rapid positioning bone drill robot according to claim 1, characterized in that, It also includes a data storage unit for storing X-ray image data, drilling parameters calculated by the AI ​​recognition unit, and surgical procedure data.

7. An automated orthopedic rapid positioning bone drill method based on AI recognition, characterized in that, The robot described in claim 1 is used in the following steps: S1. Based on the thickness of the surgical site, adjust the distance between the upper and lower fixing blocks of the mechanical fixation module using the positioning column to clamp and lock the surgical site, and record the vertical distance H between them; S2. Acquire X-ray fluoroscopic image data containing bone tissue of the surgical site through the X-ray image acquisition interface; The S3.AI recognition unit enhances X-ray image data, extracts bone tissue contours and feature point coordinates, and calculates the needle insertion position, needle exit position and needle insertion depth by combining the vertical spacing H and bone tissue projection angle α, and recommends the drilling rotation speed. S4. Based on the needle entry position, needle exit position, and drilling path, the controller drives the 5-axis structure of the bone drill module through the motor control unit to calibrate the consistency between the drill bit axis and the preset path, so that the drill bit is aligned with the needle entry position. S5. The bone drill module is started. The drill bit drills into the upper fixed block limit hole along the preset path and completes the drilling at the needle exit position.

8. The method for automated orthopedic rapid positioning bone drill based on AI recognition according to claim 7, characterized in that, In step S3, the AI ​​recognition unit processes the X-ray image data as follows: using a Frangi filter to enhance the contrast between bone tissue and surrounding tissue, locating the central axis L of bone tissue through Hough transform, analyzing bone density distribution based on grayscale threshold segmentation, defining grayscale values ​​> 200 as high-density areas, grayscale values ​​between 100 and 200 as medium-density areas, and grayscale values ​​< 100 as low-density areas, prioritizing medium and high-density areas for needle insertion and avoiding areas with bone tissue edges ≥ 2mm.

9. The AI-based automated orthopedic rapid positioning bone drill method according to claim 7, characterized in that, In step S4, when the controller drives the bone drill module to move, it first calibrates the drill bit angle by rotating along the A and B axes so that the deviation between the drill bit axis and the path from the needle entry position to the needle exit position is ≤0.1°. Then, it precisely positions the drill bit to the needle entry position by translating along the X, Y, and Z axes, with a positioning time of ≤3 seconds.

10. The method for automated orthopedic rapid positioning bone drill based on AI recognition according to claim 7, characterized in that, In step S3, the drilling rotation speed set in the AI ​​recognition unit is: 2500-3500 r / min in the high-density area of ​​bone tissue, 1500-2500 r / min in the medium-density area, and 800-1500 r / min in the low-density area. For every 10 mm increase in needle depth, the rotation speed adaptability is increased by 5%-10% to compensate for changes in bone tissue resistance.

Images