An all-orthopedic cold region surgical robot intelligent navigation positioning system and method

Abstract

The application provides a full-orthopedic cold region surgery robot intelligent navigation positioning system and method, the system comprises: an intelligent evaluation module for generating auxiliary evaluation results; a joint replacement planning and execution module for targeted surgery planning, registration mapping and control of the mechanical arm to execute the surgery planning; a spine surgery planning and execution module for targeted surgery planning, registration mapping and control of the mechanical arm to execute the surgery planning; a trauma surgery planning and execution module for targeted surgery planning, registration mapping and control of the mechanical arm to execute the surgery planning; a sports medicine planning and execution module for targeted surgery planning, registration mapping and control of the mechanical arm to execute the surgery planning; and an auxiliary rehabilitation module for generating a rehabilitation plan. In this way, the cold region surgery robot intelligent navigation positioning system can perform targeted auxiliary evaluation, surgery and auxiliary rehabilitation for different orthopedic diseases.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and more specifically, to an intelligent navigation and positioning system and method for a cold-region surgical robot used in orthopedics. Background Technology

[0002] People living in cold / high-latitude environments for extended periods are prone to bone abnormalities such as localized bone loss and relatively thickened but more brittle cortex due to prolonged low temperatures and insufficient sunlight. Specifically, this manifests as harder hard areas and more porous areas within the same bone, leading to significant fluctuations in resistance during drilling or osteotomy. Furthermore, in cold regions, low temperatures induce vasoconstriction, resulting in poor bone microcirculation. Combined with the increased bone fragility associated with low temperatures, this makes occult bone fractures or drill bit edge breakage more likely.

[0003] With the rapid development of the ice and snow economy, orthopedic diseases such as fractures in cold regions and sports injuries are increasing, covering multiple subspecialties including joint replacement, spinal surgery, trauma fracture fixation, and sports medicine repair. However, existing orthopedic surgical robot systems are usually designed for a single surgical procedure, and different procedures require different robot systems, leading to problems such as redundant equipment investment, cumbersome intraoperative switching, and lack of data interoperability between systems. Summary of the Invention

[0004] To address the aforementioned problems, the first aspect of this application provides an intelligent navigation and positioning system for cold-weather surgical robots in orthopedics, comprising: The intelligent assessment module is used to conduct targeted disease assessments on patients based on their medical data and cold-region-specific data, generating personalized auxiliary assessment results for each patient. The joint replacement planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The spinal surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The trauma surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's post-traumatic medical data and cold-region-specific data. The sports medicine planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The auxiliary rehabilitation module is used to generate personalized preoperative rehabilitation plans for patients based on their medical data, surgical plans, and cold-region-specific data; and to generate personalized postoperative rehabilitation plans for patients based on their medical data, postoperative assessment results, and cold-region-specific data.

[0005] The second aspect of this application provides a control method for an intelligent navigation and positioning system for a surgical robot used in cold regions in orthopedics, comprising: Based on the patient's medical data and cold-region-specific data, a targeted disease assessment is conducted to generate personalized auxiliary assessment results for the patient. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's joint replacement surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's spinal surgery. Based on the patient's post-traumatic medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are performed for the patient's trauma surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's sports medicine surgery. Based on the patient's medical data, surgical plan, and cold-region-specific data, a personalized preoperative rehabilitation plan is generated for the patient; and based on the patient's medical data, postoperative assessment results, and cold-region-specific data, a personalized postoperative rehabilitation plan is generated for the patient.

[0006] A third aspect of this application provides an electronic device comprising: a memory and a processor; The memory is used to store programs; The processor, coupled to the memory, is used to execute the program to implement the above-described intelligent navigation and positioning system control method for cold-region surgical robots in orthopedics.

[0007] The fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the above-described control method for the intelligent navigation and positioning system of a cold-region surgical robot for orthopedic procedures.

[0008] In this way, the intelligent navigation and positioning system of the cold-weather surgical robot can provide targeted auxiliary assessment, surgery and rehabilitation for different orthopedic diseases. Attached Figure Description

[0009] Figure 1This is an overall architecture diagram of the intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics according to an embodiment of this application; Figure 2 This is a schematic diagram of the active-passive hybrid robotic arm architecture of the intelligent navigation and positioning system for cold-region surgical robots in orthopedics according to an embodiment of this application; Figure 3 This is a schematic diagram of the architecture of a handheld robot for an intelligent navigation and positioning system for cold-weather surgical robots in orthopedics according to an embodiment of this application. Figure 4 This is a schematic diagram of the dual-arm collaborative architecture of the intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics according to an embodiment of this application; Figure 5 This is a flowchart of a control method for an intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics, according to an embodiment of this application; Figure 6 This is an architectural diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0010] To make the above-mentioned objects, features, and advantages of this application more apparent and understandable, specific embodiments of this application will be described in detail below with reference to the accompanying drawings. Although exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0011] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application pertains.

[0012] It should be noted that the joint replacement surgery, spinal surgery, trauma surgery, and sports medicine surgery described in this application can all be performed using various methods, including handheld robotic arms, hybrid active-passive robotic arms, and dual robotic arms. Conventional six-axis or seven-axis robotic arms can also be used. This application only provides examples of different combinations for illustration and is not intended to limit the surgical procedures to a single method.

[0013] For ease of understanding, the following terms may be used and are explained below: This application provides an intelligent navigation and positioning system for a cold-region surgical robot in orthopedics. The specific solution of this system is as follows: Figures 1-4 As shown.

[0014] Combination Figure 1The diagram shown is an architectural representation of an intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics, according to an embodiment of this application; wherein the intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics includes: The intelligent assessment module is used to conduct targeted disease assessments on patients based on their medical data and cold-region-specific data, generating personalized auxiliary assessment results for each patient. The joint replacement planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The spinal surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The trauma surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's post-traumatic medical data and cold-region-specific data. The sports medicine planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The auxiliary rehabilitation module is used to generate personalized rehabilitation plans for patients based on their medical data, surgical plans, and cold-region-specific data.

[0015] In this way, the intelligent navigation and positioning system of the cold-weather surgical robot can provide targeted auxiliary assessment, surgery and rehabilitation for different orthopedic diseases.

[0016] In this application, a cold-region surgical robot intelligent navigation and positioning system is used to achieve full coverage of orthopedic scenarios by utilizing cold-region-specific data and providing personalized surgical planning and execution.

[0017] This application provides an intelligent navigation and positioning system for surgical robots operating in cold regions. The system is centered on a unified intelligent navigation and positioning platform, integrating six functional modules to cover the entire process from preoperative assessment to postoperative rehabilitation. The core design concept of the system is that different orthopedic surgical procedures have fundamentally different requirements for the robotic arm's shape. Therefore, on a unified navigation and positioning platform and a cold-region-specific data platform, the most suitable hardware execution unit is configured for each surgical procedure.

[0018] In one implementation, the intelligent assessment module is used to conduct targeted disease assessments on patients based on their medical data and cold-region-specific data, generating personalized auxiliary assessment results for the patients.

[0019] The implementation of the personalized assessment plan includes the following steps: Data Acquisition and Preprocessing. The system receives the patient's CT / MRI image data, DXA bone mineral density data, HR-pQCT microstructural data, and laboratory test data. The image data undergoes standardized preprocessing, including noise reduction, contrast enhancement, and isotropic resampling, to meet the input requirements of subsequent deep learning models.

[0020] Automatic segmentation and reconstruction of 3D skeletal models. An automatic segmentation model based on a 3D convolutional neural network (3D-CNN) was used to segment the skeletal structure of patient CT images. This 3D-CNN model employs a 3D U-Net architecture. Its encoder path includes four downsampling stages, each consisting of two 3×3×3 convolutional layers, a batch normalization layer, and a ReLU activation function, followed by a 2×2×2 max-pooling layer for downsampling. The decoder path is symmetrical to the encoder, achieving upsampling through transposed convolutions and fusing features from the same layer of the encoder via skip connections. The final output layer uses 1×1×1 convolutions and a Softmax activation function, outputting probability maps of each voxel belonging to the cortical bone, cancellous bone, articular cartilage, and background.

[0021] The training process of this 3D U-Net segmentation model is as follows: The training dataset consists of CT images of over 3000 orthopedic patients from cold regions and their corresponding expert-annotated segmentation masks. To ensure the model's adaptability to the skeletal features of patients from cold regions, the training set includes samples with different bone density levels (normal, reduced bone mass, and osteoporosis), with each type of sample sampled according to the actual distribution ratio of the cold-region population. A combined loss function is used during training, which is a weighted sum of the Dice loss and cross-entropy loss, with the Dice loss weighted at 0.7 and the cross-entropy loss weighted at 0.3. The Adam optimizer is used for training, with an initial learning rate of 1×10⁻⁻⁻⁶. 4 The learning rate was adjusted using a cosine annealing strategy, with a training period of 300 epochs and a batch size of 4. Data augmentation strategies included random affine transformation (rotation range ±15°, scaling ratio 0.85~1.15), random elastic deformation, random contrast adjustment, and random noise injection. The model achieved a segmentation accuracy with an average Dice coefficient greater than 0.95 in 5-fold cross-validation.

[0022] Refined mapping of bone mineral density distribution. Based on the patient's DXA and CT data, a convolutional neural network regression model is used to construct a refined mapping from CT grayscale values ​​to volumetric bone mineral density (vBMD). This regression model takes local image patches from the patient's CT images as input and outputs the vBMD value of the corresponding region. The model employs a ResNet-18 residual network architecture, removing the final fully connected classification layer and replacing it with a fully connected regression layer, outputting a continuous vBMD prediction value.

[0023] The training process of this bone mineral density mapping regression model is as follows: The training set is composed of data from cold-region patients who simultaneously possess CT images and standard DXA / QCT measurements of the same location. Image patches of 32×32×32 voxels are extracted from the CT images using a sliding window method. Each image patch is labeled with the standard vBMD measurement value corresponding to that region. Training employs the mean squared error (MSE) loss function, the Adam optimizer, and an initial learning rate of 5×10⁻⁻⁶. 4 The model was trained for 300 epochs. Data augmentation included random cropping offset (±4 voxels), random flipping, and random brightness adjustment. The trained model can convert the grayscale value of each voxel in the patient's preoperative CT images into the corresponding bone mineral density estimate, generating a three-dimensional bone mineral density distribution map of the whole bone, providing voxel-level bone information for subsequent surgical planning.

[0024] Comprehensive risk assessment for cold regions. Individual bone mineral density distribution and bone microstructure parameters of patients were compared and matched with a cold-region osteoporosis epidemiology database to calculate the Cold-Region Fracture Risk Index (CR-FRI). CR-FRI comprehensively considers bone mineral density T-scores, bone microstructure degradation indicators (such as the percentage decrease in BV / TV and the percentage increase in Ct.Po), vitamin D deficiency, and patient age and gender factors, and quantifies risk using a multivariate logistic regression model. The CR-FRI calculation results will serve as important input parameters for cold-region-specific adjustments in subsequent surgical planning modules.

[0025] Personalized assessment report generation. The system integrates the above analysis results into a personalized auxiliary assessment report, including a 3D bone model, bone density distribution heatmap, summary of bone microstructure parameters, CR-FRI score and its risk level, as well as surgical risk warnings and suggestions based on the above data, providing comprehensive auxiliary reference for the surgeon's surgical decision-making.

[0026] In one implementation, the assistive rehabilitation module is used to generate a personalized rehabilitation plan based on the patient's medical data, surgical planning scheme, and cold-region-specific data.

[0027] The process of generating a personalized rehabilitation plan is as follows: the system comprehensively considers factors such as the patient's surgical type and plan details (osteotomy amount, screw placement, fixation method, etc.), preoperative bone mineral density level and bone metabolism status, degree of vitamin D deficiency in cold regions, patient age and body mass index, etc., and generates a phased rehabilitation plan through a rehabilitation plan recommendation convolutional network.

[0028] The rehabilitation program recommendation network takes as input preoperative CT image features (bottleneck feature vectors extracted by a shared feature extractor), surgical program parameter vectors, and cold-region-specific parameter vectors. It outputs rehabilitation program parameter vectors through a multi-layer fully connected network, including time nodes for each rehabilitation stage, exercise intensity level, weight restriction ratio, rehabilitation training type and frequency, vitamin D and calcium supplementation plan, and regular follow-up schedule.

[0029] The training data for this network consists of preoperative data, surgical plans, and postoperative rehabilitation records (including functional scores and radiographic healing assessments at various time points) of patients who underwent orthopedic surgery in cold regions. The training objective is to maximize the consistency between the predicted rehabilitation plan and the historical plan that actually achieved the best functional recovery. The loss function is the sum of the weighted mean square errors of each rehabilitation parameter, with the weights of different parameters determined based on their contribution to the final functional recovery score. Training uses the Adam optimizer with a learning rate of 1×10⁻⁻⁶. 4 Train for 200 epochs.

[0030] The rehabilitation program takes into account the influence of cold environments: (a) considering the insufficient vitamin D synthesis among residents of cold regions, the dosage and frequency of vitamin D and calcium supplementation are strengthened in the rehabilitation program; (b) considering the limited outdoor sports conditions in winter, an indoor rehabilitation training program is recommended, and the exercise content is dynamically adjusted according to the season; (c) considering the impact of low temperature on joint mobility and soft tissue healing, the time and intensity of the warm-up preparation phase are increased in the rehabilitation program.

[0031] In one implementation, the joint replacement planning and execution module is specifically used for: This is used to develop targeted surgical plans for patients based on their medical images and cold-region-specific data, generating personalized osteotomy plans; the cold-region-specific data is used to correct the osteotomy area in the osteotomy plan. Based on the navigator and tracer, register the physical skeleton during the operation and the three-dimensional skeleton model before the operation, and map the osteotomy plan and cold-region specific data; Based on the mapped osteotomy scheme and cold-region specific data, the osteotomy path of the robotic arm is planned, and the robotic arm is controlled to execute the osteotomy path; the cold-region specific data is used to correct the osteotomy path; the robotic arm is any one of the following: a hybrid active-passive robotic arm, a handheld robotic arm, a six-axis / seven-axis robotic arm, or a multi-arm structure.

[0032] The process of generating a personalized osteotomy plan is as follows: Automatic extraction of joint morphological parameters. A three-dimensional skeletal model is constructed based on the patient's medical images, and anatomical landmarks are identified as joint morphological parameters.

[0033] Personalized osteotomy parameter optimization. Based on extracted joint morphological parameters, combined with the patient's bone mineral density map and cold-region-specific data, a personalized osteotomy plan is generated through an osteotomy parameter optimization convolutional network. This network takes as input three-dimensional CT images of the patient's joint region, overlaid bone mineral density maps, and cold-region-specific parameters encoded as multi-channel inputs (including CR-FRI scores and bone microstructure parameter vectors), and adopts a multi-task 3D convolutional network architecture.

[0034] Specifically, this multi-task 3D convolutional network includes a shared feature extraction backbone and multiple task-specific branches. The shared feature extraction backbone adopts a 3D ResNet-50 architecture to extract high-level feature representations from the input multi-channel 3D data. The task-specific branches include an osteotomy plane prediction branch (outputting the normal vector and plane equation parameters of the osteotomy plane), an osteotomy depth prediction branch (outputting the osteotomy depth values ​​of each osteotomy plane), and a prosthesis model recommendation branch (outputting a matching score with each model in the prosthesis database).

[0035] The role of cold-region-specific data in correcting osteotomy procedures is reflected in the following aspects: when a patient's CR-FRI score is in the medium-to-high risk range, the system automatically corrects the osteotomy area, specifically including: (a) increasing the osteotomy safety margin in areas where bone density is below a preset threshold (e.g., vBMD < 150 mg / cm³) to avoid osteotomy in areas with extremely thin bone; (b) adjusting the osteotomy plane angle based on bone microstructure parameters (especially cortical thickness Ct.Th and cortical porosity Ct.Po) to ensure that the osteotomy surface passes through areas with thicker cortical bone as much as possible, thus guaranteeing the mechanical support strength of the prosthesis fixation surface; and (c) marking and avoiding vulnerable bone zones around the osteotomy area by integrating bone density distribution and microcrack density data.

[0036] The training process of the osteotomy parameter optimization convolutional network is as follows: The training data consists of more than 1500 cases of joint replacement surgery in cold regions. Each case includes preoperative CT images, bone density distribution data, cold-region-specific parameters, and the optimal osteotomy plan verified by senior orthopedic experts (as the gold standard label). The loss function of the osteotomy plane prediction branch is a weighted sum of the normal vector cosine similarity loss and the plane distance loss; the osteotomy depth prediction branch uses Smooth L1 loss; and the prosthesis recommendation branch uses cross-entropy loss. The total loss function is the weighted sum of the losses of the three branches, with weights of 0.4, 0.3, and 0.3, respectively. Training uses the SGD optimizer with a momentum of 0.9, an initial learning rate of 1×10⁻², and a cosine annealing strategy to adjust the learning rate. The training period is 250 epochs, and the batch size is 2 (due to the memory limitation of the 3D data). During training, cold-region-specific parameters are encoded by channel splicing. Scalar parameters such as CR-FRI score, BV / TV, Ct.Th, and Ct.Po are expanded into constant channels with the same spatial size as the images. These channels are then spliced ​​with CT images and bone densitograms in the channel dimension and input into the network.

[0037] In one implementation, the spinal surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data.

[0038] The process of generating a personalized spinal screw placement plan includes: Automatic vertebral segmentation and identification. A vertebral segmentation model based on a 3D convolutional network is used to segment and number each vertebra in patient spinal CT images. The model employs a cascaded network architecture: the first-level network (based on 3D U-Net) performs overall spinal region detection and coarse segmentation; the second-level network (based on nnU-Net) performs fine segmentation and classification of each vertebra.

[0039] Personalized planning of pedicle screw access. Based on vertebral body segmentation results and bone mineral density distribution maps, a pedicle parameter prediction convolutional network is used to automatically plan the optimal screw entry point, entry angle (sagittal and transverse tilt angles), screw length, and diameter for each vertebra. This network takes a 3D CT image block of a single vertebra and its corresponding bone mineral density distribution as input, extracts features using the 3DResNet-34 architecture, and outputs a screw parameter vector through a fully connected regression layer.

[0040] The training process of this pedicle parameter prediction network is as follows: The training data consists of preoperative CT images from over 2500 spinal surgery cases in cold regions and screw parameters planned by senior spinal surgeons and verified as excellent postoperatively. For each data pair, the input is a 64×64×64 voxel 3D image block (including bone density channel) cropped centered on the target vertebral body, and the label is a six-dimensional parameter vector of the corresponding pedicle screw (entry point 3D coordinate offset, sagittal tilt angle, transverse tilt angle, screw length). The loss function used is Smooth L1 loss. Training uses the Adam optimizer with an initial learning rate of 5×10⁻⁻⁶. 4 The model was trained for 200 epochs. During training, samples with different bone density levels were oversampled to ensure the model's accuracy in planning osteoporotic vertebrae.

[0041] The correction effect of cold-region specific data is reflected in the following: when the local bone density of the vertebral body is lower than the safe threshold, the system automatically adjusts the trajectory direction of the screw channel so that the screw passes through the cortical bone area with higher bone density as much as possible. If necessary, it recommends the use of bone cement reinforcement or the selection of larger diameter screws to enhance the holding force.

[0042] In one implementation, the trauma surgery planning and execution module is specifically used for: This is used to plan a targeted surgery for the patient based on medical images after repositioning and cold-region-specific data, generating a personalized fixation plan; the cold-region-specific data is used to correct the fixation position. Based on the navigator and tracer, register the physical skeleton during the operation and the preoperative 3D skeleton model to map the fixation scheme and cold-region-specific data; Based on the fixed mapping scheme and cold-region specific data, a fixed path for the robotic arm is planned, and the robotic arm is controlled to execute the fixed path; the cold-region specific data is used to correct the fixed path; the robotic arm is any one of the following: a hybrid active-passive robotic arm, a handheld robotic arm, a six-axis / seven-axis robotic arm, or a multi-arm structure.

[0043] The implementation of personalized fixation plans includes: Automatic fracture type identification and reduction assessment. A fracture classification model based on a convolutional neural network was used to identify the fracture type (e.g., AO / OTA classification) in images of patients after reduction, and the reduction quality was assessed by comparison with the contralateral image or a standard template. The classification model adopted a 3D DenseNet-121 architecture, and the training dataset consisted of images of over 3000 cases of fractures in cold regions and their corresponding AO / OTA classification labels. Training employed the cross-entropy loss function, the Adam optimizer, a learning rate of 1×10⁻³, and 150 epochs. To address the issue of uneven fracture type distribution, a focal loss strategy was used during training, assigning higher loss weights to rare fracture types.

[0044] Personalized planning of fixation methods and locations. Based on the fracture type, post-reduction bone morphology, and bone mineral density distribution, a convolutional network for fixation planning is used to generate personalized fixation plans. The inputs to this network are post-reduction 3D CT images, bone mineral density maps, and coded cold-region-specific parameters; the outputs include recommended fixation methods (plate fixation, intramedullary nail fixation, external fixation, etc.), placement and angle of fixation devices, and number and location of screws.

[0045] The role of cold-region-specific data in fixation protocols is as follows: based on local bone density and microstructural parameters at the fracture site, the fixation position of screws can be adjusted to avoid placing screws in areas of extreme osteoporosis; when bone density is generally low, it is recommended to increase the number of screws or use locking plates to enhance fixation stability; at the same time, the selection parameters of fixation devices can be adjusted according to the CR-FRI score.

[0046] In one implementation, the sports medicine planning and execution module is specifically used for: This is used to plan targeted surgeries for patients based on their medical images and cold-region-specific data, generating personalized bone tunnel screw placement plans; the cold-region-specific data is used to verify whether the bone quality of the tunnel wall meets the preset mechanical stability threshold. Based on the navigator and tracer, the intraoperative solid skeleton and the preoperative 3D skeleton model are registered to map the bone tunnel nail placement scheme and cold-region specific data. Based on the mapped bone tunnel pin placement scheme and cold-region specific data, the pin placement path of the robotic arm is planned, and the robotic arm is controlled to execute the pin placement path; the cold-region specific data is used to correct the pin placement path; the robotic arm is any one of the following: active-passive hybrid robotic arm, handheld robotic arm, six-axis / seven-axis robotic arm, and multi-arm structure.

[0047] The implementation of a personalized bone tunnel screw placement plan includes: Automatic ligament attachment point localization. An anatomical landmark detection model based on a convolutional neural network is used to automatically locate ligament attachment points (such as the femoral and tibial insertions of the ACL) on the patient's MRI or CT images. This model uses an HRNet architecture similar to the keypoint detection network in the joint replacement module, but the training data consists of ligament attachment point annotation data related to sports medicine surgery.

[0048] Bone tunnel parameter optimization. Based on the located ligament attachment points, combined with the patient's bone density distribution and bone microstructure parameters, a bone tunnel planning convolutional network is used to optimize the tunnel's entrance point, exit point, tunnel angle, and tunnel diameter. This network takes 3D images of the joint region and bone density distribution maps as input, employs a 3D convolutional architecture, and outputs a tunnel parameter vector.

[0049] The role of cold-region-specific data in verifying bone tunneling protocols is particularly crucial: the system verifies whether the tunnel wall bone quality meets a preset mechanical stability threshold based on the bone mineral density and bone microstructure parameters (especially BV / TV and Tb.Th) of the bone wall surrounding the planned tunnel. This threshold is determined based on mechanical test data of the corresponding population in a cold-region osteoporosis epidemiology database. If the tunnel wall bone quality does not meet the threshold, the system automatically adjusts the tunnel angle or position, or recommends using a larger diameter tunnel to increase the bone-graft contact area, or suggests using bone cement or artificial bone to reinforce the tunnel wall during surgery.

[0050] This application supports various types of robotic arms to meet the needs of different surgical scenarios.

[0051] In one implementation, combined with Figure 2 As shown, the robotic arm is a hybrid active-passive robotic arm, comprising a passive mechanical segment and an active mechanical segment: The passive mechanical segment is used to work in conjunction with the operator. It is manually pushed to the target area and locked by the operator to provide coarse positioning of the end effector. The active mechanical segment consists of multiple servo-driven active joints, which are used to plan and execute the motion path of the end effector based on the registration mapping results within the locked positioning range.

[0052] Hybrid active-passive robotic arms consist of a passive mechanical segment and an active mechanical segment. The passive mechanical segment is typically a multi-degree-of-freedom, unpowered articulated arm. The surgeon can manually push it to the vicinity of the target surgical area and then lock it using a pneumatic or electromagnetic locking device to achieve coarse positioning of the end effector. The active mechanical segment, connected to the end of the passive mechanical segment, consists of multiple servo-driven active joints (usually 3-6 degrees of freedom). Within the limited workspace after locking, it precisely plans and executes the motion path of the end effector based on the registration mapping results. This hybrid design combines the advantages of large-range flexible adjustment (passive segment) and small-range precise execution (active segment), while reducing the system's size and cost.

[0053] In one implementation, combined with Figure 3 As shown, the robotic arm is a handheld robotic arm, including a motion recognition module and a trajectory correction module; The motion recognition module is used to collect the surgeon's hand motion signals in real time and extract the intentional operation motion components as effective surgical motion commands. The trajectory correction module is used to compare the effective surgical movement command with the registered and mapped surgical path in real time, calculate the deviation vector and generate a correction torque to guide the surgeon's hand to move along the surgical path.

[0054] Handheld robotic arm: An intelligent handheld tool providing real-time force feedback guidance for the surgeon. It includes a motion recognition module and a trajectory correction module. The motion recognition module acquires the surgeon's hand motion signals in real time through an inertial measurement unit (IMU) and force / torque sensors integrated into the handheld device. Using a signal processing method based on an adaptive filtering algorithm, the surgeon's hand movements are decomposed into intentional operational motion components (low-frequency, large-amplitude cutting / drilling movements) and unintentional physiological tremor components (high-frequency, small-amplitude hand tremors), extracting only the intentional operational motion components as valid surgical movement commands. The trajectory correction module compares the valid surgical movement commands with the registered and mapped surgical path in real time, calculates the deviation vector, and generates a correction torque through a force / torque actuator within the device, guiding the surgeon's hand along the planned path in a tactile manner. When the surgeon deviates from the path beyond a safety threshold, the system automatically increases the damping force to limit the deviation.

[0055] In one embodiment, the robotic arm is a six-axis / seven-axis robotic arm.

[0056] Six-axis / seven-axis robotic arms: Fully active industrial-grade robotic arms with 6-7 rotational degrees of freedom, offering a large workspace and flexible posture adjustment capabilities. The additional degree of freedom in a seven-axis robotic arm compared to a six-axis arm provides redundancy, allowing it to adjust the arm's posture while maintaining the end-effector's position, thus avoiding obstacles in the surgical space.

[0057] In one implementation, combined with Figure 4 As shown, the robotic arm has a multi-arm structure, including a main operating arm and at least one auxiliary arm; The main operating arm performs predetermined surgical operations along the planned surgical path based on the registration mapping results; The auxiliary arm provides support during the surgical procedure performed by the main operating arm.

[0058] Multi-arm structure: Includes a main manipulator and at least one auxiliary arm. The main manipulator performs predetermined surgical procedures (such as osteotomy, drilling, and screw placement) along a planned surgical path based on the registration mapping results. The auxiliary arm provides support during the main manipulator's surgical procedures, including but not limited to: traction or stabilization of bone / soft tissue in the surgical area, holding the visual navigation camera for optimal field of view, clearing debris or fluid from the surgical area, and providing additional retractor traction when necessary. Collision avoidance and task coordination are achieved between the multiple arms through a cooperative control algorithm.

[0059] In one embodiment, the intelligent navigation and positioning system for cold-region surgical robots in the whole orthopedics field includes at least two of the following cold-region-specific data: epidemiological data on osteoporosis in cold regions, data on vitamin D deficiency and bone metabolism in cold regions, data on bone microstructure characteristics, and data on tissue characteristics during surgery in low-temperature environments.

[0060] The cold-region-specific data is the core data foundation that distinguishes this system from general surgical robot systems, and includes at least two of the following data: Epidemiological data on osteoporosis in cold regions include age-sex-bone mineral density distribution models based on large-scale population surveys in cold regions and the incidence of fragility fractures at various sites in cold regions. These data provide a population-level reference baseline for the intelligent assessment module, used to assess the relative level of bone risk in individual patients.

[0061] Data on vitamin D deficiency and bone metabolism in cold regions, including the distribution characteristics of serum 25(OH)D levels and seasonal bone loss patterns in high-latitude areas. Shorter daylight hours in winter in high-latitude regions lead to widespread vitamin D deficiency, resulting in disordered bone metabolism and accelerated bone loss.

[0062] Bone microstructural characteristics data include thinning of trabeculae, increased trabecular spacing, and cumulative density parameters of bone microcracks in patients from cold-climate regions. These microstructural parameters directly affect the mechanical load-bearing capacity of bone tissue and are core input parameters for microcrack-protective movement strategies.

[0063] Intraoperative tissue characteristic data in a cryogenic environment, including temperature correction factors for the elastic modulus and stiffness of the patient's bone tissue under cryogenic conditions. Cryogenic environments may affect the mechanical properties of bone tissue, necessitating temperature correction of relevant parameters during surgery.

[0064] In this application, the cold-region-specific data of the patient is generated by cross-machine correction of the patient's medical images and based on the corrected medical images.

[0065] The cross-machine correction in this application involves setting a third-party module with the same density during image capture. During medical image processing, this third-party module performs normalization or standardization corrections, thereby eliminating differences between different machines.

[0066] Specifically, the cross-machine correction model employs a convolutional neural network based on the U-Net architecture. This network consists of an encoder and a decoder: the encoder comprises four convolutional layers, each containing two 3x3 convolutional layers, a batch normalization layer, and a ReLU activation function, and downsamples using 2x2 max pooling, with feature channels numbering 64, 128, 256, and 512 respectively; the decoder upsamples using deconvolution and establishes skip connections with the feature maps of the corresponding layers in the encoder to progressively restore spatial resolution. The network input is original CT image slices, and the output is corrected, normalized CT image slices.

[0067] The training process for cross-machine calibration convolutional models is as follows: Training data acquisition: Collect skeletal imaging data from at least five different brands / models of CT scanners, with each scanner containing scan data from at least 200 patients. Simultaneously, select a precisely calibrated reference scanner (such as a high-end CT scanner calibrated using a standard phantom) as the "gold standard." Perform scans on both the reference scanner and the scanner to be calibrated on a subset of patients (at least 50 cases) to acquire paired imaging data.

[0068] Training Strategy: A hybrid strategy combining paired and unpaired training is employed. For paired data, the mean squared error (MSE) loss function is directly used to minimize the difference between the corrected output and the reference standard. For unpaired data, an adversarial training mechanism is introduced, adopting the CycleGAN architecture to achieve cross-domain transformation through Cycle Consistency Loss and Adversarial Loss. Perceptual Loss is calculated using intermediate layer features from a pre-trained VGG-16 network to maintain the structural consistency of the image.

[0069] Training parameters: The Adam optimizer was used, with an initial learning rate of 0.0002, β1=0.5, and β2=0.999; the batch size was 8; the total number of training epochs was 200, with the learning rate linearly decaying to zero after the 100th epoch; data augmentation strategies were employed, including random rotation (-15° to +15°), random flipping, and random pruning.

[0070] The cold-region-specific data includes at least a three-dimensional bone density map, and may also include bone microstructure parameter maps, bone cortical thickness distribution maps, joint space distribution maps, bone degeneration scores, etc.

[0071] The generation of the 3D bone mineral density map employs a bone mineral density prediction model based on a 3D convolutional neural network (3D-CNN). This model takes corrected CT volume data as input and outputs voxel-level bone mineral density values ​​(unit: mg / cm³), thereby constructing a 3D bone mineral density distribution map of the patient's entire skeleton.

[0072] The training process of the 3D-CNN model for bone density prediction is as follows: Training data preparation: Simultaneous QCT (quantitative CT) scan data and DXA (dual-energy X-ray absorptiometry) measurement data were collected from at least 500 patients in cold-region areas. QCT data provided voxel-level bone mineral density reference values ​​as training labels, and DXA data provided regional bone mineral density validation values. In addition, data from at least 300 non-cold-region control groups were collected for comparative analysis.

[0073] Network Architecture: A 3D V-Net architecture is adopted. The encoding path contains four resolution stages, each consisting of 1-3 3x3x3 residual convolutional blocks with feature channels of 16, 32, 64, and 128 respectively. Downsampling is performed between each stage using 2x2x2 convolutions (with a stride of 2). The decoding path is symmetrical to the encoding path, performing upsampling through deconvolutions and fusing encoder features via skip connections. The final output layer uses 1x1x1 convolutions to map features to single-channel bone density predictions and a sigmoid activation function to map the output to a reasonable bone density range.

[0074] Loss function design: A composite loss function is adopted, in which mean squared error loss ensures overall prediction accuracy; gradient loss maintains the spatial continuity and boundary clarity of bone density distribution; and regional consistency loss ensures the consistency between the predicted regional average bone density and the DXA measurement value.

[0075] Training Strategy: A phased training strategy was adopted. Phase 1 (Pre-training): 100 epochs were trained using a mixed dataset (cold and non-cold regions) to learn general bone density prediction capabilities. The initial learning rate was 0.001, and cosine annealing was used for learning rate scheduling. Phase 2 (Fine-tuning): The first two layers of the encoder were frozen, and 50 epochs were fine-tuned using the cold-region dataset to learn cold-region-specific bone density distribution patterns. The learning rate was reduced to 0.0001. Mixed precision (FP16) acceleration was used throughout the training process, with a batch size of 4 (limited by the memory usage of 3D volumetric data).

[0076] In one embodiment, the intelligent navigation and positioning system for the cold-region surgical robot in the whole orthopedics, while controlling the robotic arm to execute the corresponding surgical plan, also automatically embeds bone-protecting operations into the surgical execution trajectory based on the microcrack protection motion strategy parameters determined by the cold-region specific data. The microcrack protection motion strategy includes at least one of the following: segmented feed strategy, pulse propulsion strategy, and drill unloading strategy.

[0077] The four planning and execution modules of this system automatically embed bone-protecting operations into the surgical trajectory based on microcrack protection motion strategy parameters determined by cold-region-specific data. This is the core technological innovation of this system, specifically addressing the bone characteristics specific to cold regions.

[0078] The microcrack-protective motion strategy includes: a segmented feed strategy, which divides the surgical execution path into multiple execution segments based on the patient's cumulative bone microcrack density parameters, and actively pauses for a preset time after each segment is completed to release the accumulated stress in the bone tissue; a pulse propulsion strategy, which uses intermittent pulse force output to replace continuous constant force propulsion for the feed motion of the end effector, reducing continuous load damage to the trabecular bone structure; and a drill withdrawal and unloading strategy, which periodically withdraws the end effector at preset depth intervals to release the accumulated deformation and thermal stress in the bone tissue around the borehole, with the withdrawal depth and frequency automatically adjusted based on the cumulative bone microcrack density parameters and the risk stratification level of bone in cold regions.

[0079] In the joint replacement module, the microcrack-protected motion strategy is embedded in the osteotomy and grinding execution trajectory of the main manipulator; in the spinal surgery and trauma surgery modules, the microcrack-protected motion strategy is embedded in the trajectory correction module of the handheld robot handle, which is automatically executed by the correction torque generation unit; in the sports medicine module, the microcrack-protected motion strategy is embedded in the active manipulator execution trajectory of the active-passive hybrid manipulator.

[0080] In one embodiment, the intelligent navigation and positioning system for cold-region surgical robots in orthopedics also includes an intraoperative processing system for personalized designing and processing of surgical instruments or fixation devices in the surgical plan based on the patient's preoperative imaging data and cold-region-specific bone data, so as to improve the fit between the surgical instruments or fixation devices and the individual patient.

[0081] This system may also include an intraoperative processing system for personalized design and processing of surgical instruments or fixation devices based on the patient's preoperative imaging data and cold-region-specific bone data. The intraoperative processing system includes an instrument design unit, a processing execution unit, and a quality verification unit. The instrument design unit generates personalized design schemes and corresponding processing instructions for surgical instruments or fixation devices based on preoperative imaging data and cold-region-specific bone data. The personalized design scheme includes at least the target geometry of the instrument and a fixation component configuration scheme. The processing execution unit includes a processing platform and at least one processing tool, and processes the instrument to be processed according to the processing instructions. The quality verification unit performs morphological inspection on the processed instrument and compares the inspection results with the target geometry in the personalized design scheme. The intraoperative processing system can be a stand-alone tableside processing device or integrated into the end effector of a surgical robot arm via a quick-change tool interface.

[0082] The implementation method for personalized device design is as follows: Based on the patient's three-dimensional bone model and bone density distribution data, a personalized device design convolutional network is used to automatically generate a three-dimensional digital model of an osteotomy guide plate, fixation plate, or implant that precisely conforms to the patient's bone surface. This network takes the three-dimensional bone model (represented in voxels) and bone density distribution of the patient's surgical area as input, and uses a three-dimensional convolutional encoder-decoder architecture (similar to V-Net) to output a three-dimensional shape voxel representation of the device.

[0083] The training process of this personalized device design network is as follows: The training data consists of pairing over 1000 patient skeletal 3D models with corresponding optimal device 3D models validated by finite element analysis. Cold-region-specific data is introduced as conditional input during training, enabling the network to learn to adjust the device's thickness distribution and contact surface design based on bone density levels and microstructural characteristics—for example, increasing the contact area and wall thickness between the device and bone in areas of low bone density, and adding reinforcing ribs in areas of weak bone cortex. Training employs a combination of voxel-level cross-entropy loss and shape regularization loss, the Adam optimizer, and a learning rate of 1×10⁻. 4 Train for 300 epochs.

[0084] The generated 3D digital model of the instrument is processed in real time using additive manufacturing equipment (such as a metal 3D printer) or subtractive manufacturing equipment (such as a five-axis CNC milling machine) integrated into the intraoperative processing system. The processing material is selected based on the type of instrument, using biocompatible materials such as medical-grade titanium alloy, medical-grade stainless steel, or medical-grade PEEK. After processing, the instrument is sterilized before use in the current surgery.

[0085] In one implementation, in constructing a three-dimensional skeletal model based on a patient's medical images, the skeletal medical images are segmented using a segmentation model, and then a three-dimensional skeletal model is constructed based on the bone edges in the segmentation results.

[0086] In one implementation, the segmentation process of the segmentation model includes: The image to be segmented is downsampled sequentially to obtain downsampled images at multiple levels; Upsampling based on multi-scale extraction and multi-attention extraction is performed sequentially on downsampled images at multiple levels to obtain upsampled images at multiple levels. The highest-level upsampled image is processed to obtain the segmentation result.

[0087] Specifically, in the segmentation process: The image to be segmented is downsampled sequentially to obtain the first downsampled image, the second downsampled image, the third downsampled image, the fourth downsampled image, and the fifth downsampled image. Multi-scale extraction processing is performed on the fifth downsampled image to obtain the fifth upsampled image; Multi-attention extraction is performed on the first downsampled image, the second downsampled image, the third downsampled image, and the fourth downsampled image respectively to obtain the first attention map, the second attention map, the third attention map, and the fourth attention map; Upsampling is performed on the fifth upsampled image, the fourth attention image, the third attention image, the second attention image, and the first attention image to obtain the fourth upsampled image, the third upsampled image, the second upsampled image, and the first upsampled image in sequence. The segmentation result is obtained after convolution processing of the first upsampled image.

[0088] The technical details of upsampling are as follows: The fifth upsampled image and the fourth attention image are upsampled to obtain the fourth upsampled image.

[0089] The upsampling and downsampling processes can be referred to existing similar processes and will not be described in detail in this application.

[0090] The process of multi-attention extraction is as follows: The input feature map is divided into three branches, and convolutions of different sizes are performed to extract the feature maps. The feature maps of the three branches are then concatenated to obtain the concatenated feature map. The spliced ​​feature map is divided into two branches; Within the first branch, the concatenated feature map is convolved and positional information is added. Then, it is multiplied with the original concatenated feature map to obtain the multiplied feature map. The multiplied feature map is added to the original concatenated feature map and then convolved to obtain the branch feature map of the first branch. In the second branch, the concatenated feature map is convolved and channel features are added. Then it is multiplied with the original concatenated feature map to obtain the multiplied feature map. The multiplied feature map is added to the original concatenated feature map and then convolved to obtain the branch feature map of the second branch. By concatenating the branch feature maps of the first and second branches, the output feature map of multi-attention extraction is obtained.

[0091] In this way, by extracting positional attention features and channel attention features through branching, more contextual information can be captured using positional and channel attention at different scales, and the importance of each channel can be selectively weighted to produce the best output characteristics.

[0092] The multi-scale extraction process is as follows: The input feature map is subjected to convolution, normalization, and activation processing to obtain the first convolutional map; The first convolutional image is convolved, normalized, and activated to obtain the second convolutional image. The input feature map is subjected to feature extraction in multiple branches to obtain the corresponding feature maps; the convolution kernel of each branch is different. After concatenating and merging the feature maps of multiple branches, a concatenated convolutional map is obtained. After combining the concatenated convolutional map and the input feature map, multi-head attention, normalization, and multi-layer perception processing are performed to obtain a multi-layer perception map. By combining the second convolutional map and the multilayer perceptron map, we obtain the output feature map extracted at multiple scales.

[0093] In this way, the context extracted by larger convolutional kernels is integrated with deeper information flow, and multi-scale features are formed by integrating convolutional kernels of different depths and sizes. Multi-head attention is then used to fuse multi-scale features, thereby achieving further integration of features.

[0094] In this application, it should be noted that if inconsistent sizes occur during the specific feature extraction process, they can be unified by reshaping. The specific location for this reshaping can be determined based on the actual situation, and will not be elaborated upon in this application.

[0095] In this application, the training process of the segmentation model described above involves obtaining training samples, which include the input image to be segmented and the labeled segmentation results; inputting the image to be segmented from the training samples into the segmentation model to obtain a predicted segmentation result; calculating a loss function, namely the DiceLoss function, based on the predicted segmentation result and the sample segmentation result; and iterating the parameters of the entire segmentation model based on the loss function until the loss function converges. During training, the learning rate ranges from 1e-4 to 1e-3, the weight decay is 1e-4, and the total number of training epochs is 200-500.

[0096] It should be noted that, unless otherwise specified, the personalized processing model in this application can be obtained by targeted fine-tuning of an existing large model. The specific fine-tuning process may include: acquiring the patient's multimodal features and output information as sample data for the model based on its input and output requirements; modifying the input and output layers of the pre-trained large model to adapt it to the model's input and output; adding a low-rank adapter module next to the model's key layer (attention mechanism) so that only these few new parameters are trained during training; training the large model based on the sample data and updating the parameters within the low-rank adapter module and the modified input and output layer parameters until the loss converges. Further details can be found in the training requirements of existing models, and will not be elaborated upon in this application.

[0097] This application provides a navigation and positioning method for the intelligent navigation and positioning system of the orthopedic cold-region surgical robot described above. The specific scheme of this method is as follows: Figure 5 As shown below, the control method of the intelligent navigation and positioning system for the orthopedic cold-region surgical robot will be described in detail.

[0098] Combination Figure 5 As shown, the control method for the intelligent navigation and positioning system of the orthopedic cold-region surgical robot includes: S101, based on the patient's medical data and cold-region-specific data, conducts targeted disease assessments for the patient and generates personalized auxiliary assessment results for the patient; S102, based on the patient's medical data and cold-region-specific data, performs targeted surgical planning, registration mapping, and controls the robotic arm to execute the surgical plan for the patient's joint replacement surgery; S103, based on the patient's medical data and cold-region-specific data, performs targeted surgical planning, registration mapping, and controls the robotic arm to execute the surgical plan for the patient's spinal surgery; S104, based on the patient's post-traumatic medical data and cold-region-specific data, performs targeted surgical planning, registration mapping, and controls the robotic arm to execute the surgical plan for the patient's trauma surgery; S105, based on the patient's medical data and cold-region-specific data, performs targeted surgical planning, registration mapping, and controls the robotic arm to execute the surgical plan for the patient's sports medicine surgery; S106 generates a personalized preoperative rehabilitation plan for the patient based on the patient's medical data, surgical planning scheme, and cold-region-specific data; and generates a personalized postoperative rehabilitation plan for the patient based on the patient's medical data, postoperative assessment results, and cold-region-specific data.

[0099] In one implementation, S105, based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan for the patient's sports medicine surgery are performed; including: Based on the patient's sports medicine-related medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional model of the surgical site and a personalized three-dimensional surgical plan. The three-dimensional surgical plan is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer; The system controls a hybrid active-passive robotic arm to perform surgical operations according to a three-dimensional surgical plan. The hybrid active-passive robotic arm includes a passive robotic arm and an active robotic arm mounted at the distal end of the passive robotic arm. The passive robotic arm is used to work in conjunction with the surgeon, and is manually pushed to the target area by the surgeon and then locked to perform coarse positioning of the end effector. The active robotic arm consists of multiple servo-driven active joints, which are used to plan and execute the motion path of the end effector based on the registration mapping results within the locked positioning range.

[0100] In one implementation, S103, based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan for the patient's spinal surgery are performed; including: Based on the patient's spinal medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional spinal model and a personalized three-dimensional screw placement plan for spinal surgery. The three-dimensional pin placement scheme is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer. The surgeon performs the pin placement operation by holding a handheld robot handle along the pin placement path; wherein, the handheld robot handle has a built-in motion recognition module and a trajectory correction module; the motion recognition module is used to collect the surgeon's hand motion signals in real time and extract the intentional operation motion component as the effective surgical motion command; the trajectory correction module is used to compare the effective surgical motion command with the registered and mapped pin placement path in real time, calculate the deviation vector and generate a correction torque to guide the surgeon's hand to move along the pin placement path.

[0101] In one embodiment, S102, based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are performed for the patient's joint replacement surgery; including: Based on the patient's joint medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional joint model and a personalized three-dimensional surgical plan for joint replacement. The three-dimensional surgical plan is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer; The main control arm performs osteotomy and / or grinding operations along the osteotomy path based on the registration mapping results; wherein, the main control arm is fixedly mounted on the operating trolley and consists of multiple servo-driven active joints; The control arm provides auxiliary support during the surgical procedure performed by the main operating arm; wherein the auxiliary arm includes a passive arm segment consisting of at least one unpowered passive joint and an active arm segment consisting of multiple servo-driven active joints mounted at the distal end of the passive arm segment.

[0102] The control method of the intelligent navigation and positioning system for the cold-weather surgical robot in the whole orthopedics provided in the above embodiments of this application has a corresponding relationship with the intelligent navigation and positioning system for the cold-weather surgical robot in the whole orthopedics provided in the embodiments of this application. Therefore, the specific content of the method has a corresponding relationship with the intelligent navigation and positioning system for the cold-weather surgical robot in the whole orthopedics. The specific content can be referred to the records in the intelligent navigation and positioning system for the cold-weather surgical robot in the whole orthopedics. This application will not repeat it here.

[0103] The control method of the intelligent navigation and positioning system for cold-weather surgical robots in orthopedics provided in the above embodiments of this application is based on the same inventive concept as the intelligent navigation and positioning system for cold-weather surgical robots in orthopedics provided in the embodiments of this application, and has the same beneficial effects as the methods adopted, run or implemented by the application stored therein.

[0104] Based on the same inventive concept, another embodiment of the present invention provides an electronic device for implementing the intelligent navigation and positioning system control method for a cold-region surgical robot in orthopedics described in the above embodiments. Figure 6 As shown, the electronic device includes a memory 301 and a processor 303.

[0105] Memory 301 can be configured to store a program.

[0106] Additionally, memory 301 can also be configured to store various other data to support operation on the electronic device. Examples of this data include instructions for any application or method used to operate on the electronic device, contact data, phonebook data, messages, pictures, videos, etc.

[0107] Memory 301 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Processor 303, coupled to memory 301, is used to execute programs in memory 301 for: Based on the patient's medical data and cold-region-specific data, a targeted disease assessment is conducted to generate personalized auxiliary assessment results for the patient. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's joint replacement surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's spinal surgery. Based on the patient's post-traumatic medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are performed for the patient's trauma surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's sports medicine surgery. Based on the patient's medical data, surgical plan, and cold-region-specific data, a personalized preoperative rehabilitation plan is generated for the patient; and based on the patient's medical data, postoperative assessment results, and cold-region-specific data, a personalized postoperative rehabilitation plan is generated for the patient.

[0108] In one implementation, the processor 303 is further configured to: Based on the patient's sports medicine-related medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional model of the surgical site and a personalized three-dimensional surgical plan. The three-dimensional surgical plan is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer; The system controls a hybrid active-passive robotic arm to perform surgical operations according to a three-dimensional surgical plan. The hybrid active-passive robotic arm includes a passive robotic arm and an active robotic arm mounted at the distal end of the passive robotic arm. The passive robotic arm is used to work in conjunction with the surgeon, and is manually pushed to the target area by the surgeon and then locked to perform coarse positioning of the end effector. The active robotic arm consists of multiple servo-driven active joints, which are used to plan and execute the motion path of the end effector based on the registration mapping results within the locked positioning range.

[0109] In one implementation, the processor 303 is further configured to: Based on the patient's spinal medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional spinal model and a personalized three-dimensional screw placement plan for spinal surgery. The three-dimensional pin placement scheme is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer. The surgeon performs the pin placement operation by holding a handheld robot handle along the pin placement path; wherein, the handheld robot handle has a built-in motion recognition module and a trajectory correction module; the motion recognition module is used to collect the surgeon's hand motion signals in real time and extract the intentional operation motion component as the effective surgical motion command; the trajectory correction module is used to compare the effective surgical motion command with the registered and mapped pin placement path in real time, calculate the deviation vector and generate a correction torque to guide the surgeon's hand to move along the pin placement path.

[0110] In one implementation, the processor 303 is further configured to: Based on the patient's joint medical imaging data and cold-region-specific bone data, a three-dimensional reconstruction is performed to generate a three-dimensional joint model and a personalized three-dimensional surgical plan for joint replacement. The three-dimensional surgical plan is registered and mapped with the actual intraoperative anatomical space using a navigator and a tracer; The main control arm performs osteotomy and / or grinding operations along the osteotomy path based on the registration mapping results; wherein, the main control arm is fixedly mounted on the operating trolley and consists of multiple servo-driven active joints; The control arm provides auxiliary support during the surgical procedure performed by the main operating arm; wherein the auxiliary arm includes a passive arm segment consisting of at least one unpowered passive joint and an active arm segment consisting of multiple servo-driven active joints mounted at the distal end of the passive arm segment.

[0111] In this application, Figure 6 The diagram only shows some components and does not mean that the electronic device includes only these components. Figure 6 The components shown.

[0112] The electronic device provided in this embodiment is based on the same inventive concept as the force-optimized robotic arm osteotomy control method provided in the embodiments of this application, and has the same beneficial effects as the methods adopted, run or implemented by the application stored therein.

[0113] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0115] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0116] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory. Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0117] This application also provides a computer-readable storage medium corresponding to the force-optimized robotic arm osteotomy control method provided in the foregoing embodiments, wherein a computer program (i.e., a program product) is stored thereon. When the computer program is run by a processor, it executes the interactive image analysis assistance method for 3D aerial imaging provided in any of the foregoing embodiments.

[0118] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, optical disc read-only memory (CDROM), digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0119] The computer-readable storage medium provided in the above embodiments of this application and the intelligent navigation and positioning system control method for cold-region surgical robots in orthopedics provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application stored therein.

[0120] It should be noted that numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known structures and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0121] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, system, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, system, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, system, article, or apparatus that includes said element.

[0122] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A smart navigation and positioning system for a cold-weather surgical robot used in orthopedics, characterized in that, include: The intelligent assessment module is used to conduct targeted disease assessments on patients based on their medical data and cold-region-specific data, generating personalized auxiliary assessment results for each patient. The joint replacement planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The spinal surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The trauma surgery planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's post-traumatic medical data and cold-region-specific data. The sports medicine planning and execution module is used to perform targeted surgical planning, registration mapping, and control the robotic arm to execute the surgical plan based on the patient's medical data and cold-region-specific data. The auxiliary rehabilitation module is used to generate personalized rehabilitation plans for patients based on their medical data, surgical plans, and cold-region-specific data.

2. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedic fields according to claim 1, characterized in that, The joint replacement planning and execution module is specifically used for: This is used to develop targeted surgical plans for patients based on their medical images and cold-region-specific data, generating personalized osteotomy plans; the cold-region-specific data is used to correct the osteotomy area in the osteotomy plan. Based on the navigator and tracer, register the physical skeleton during the operation and the three-dimensional skeleton model before the operation to map the osteotomy plan and cold-region specific data; Based on the mapped osteotomy scheme and cold-region specific data, the osteotomy path of the robotic arm is planned, and the robotic arm is controlled to execute the osteotomy path; the cold-region specific data is used to correct the osteotomy path; the robotic arm is any one of the following: a hybrid active-passive robotic arm, a handheld robotic arm, a six-axis / seven-axis robotic arm, or a multi-arm structure.

3. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, The trauma surgery planning and execution module is specifically used for: This is used to plan a targeted surgery for the patient based on medical images after repositioning and cold-region-specific data, generating a personalized fixation plan; the cold-region-specific data is used to correct the fixation position. Based on the navigator and tracer, register the physical skeleton during the operation and the preoperative 3D skeleton model to map the fixation scheme and cold-region-specific data; Based on the fixed mapping scheme and cold-region specific data, a fixed path for the robotic arm is planned, and the robotic arm is controlled to execute the fixed path; the cold-region specific data is used to correct the fixed path; the robotic arm is any one of the following: a hybrid active-passive robotic arm, a handheld robotic arm, a six-axis / seven-axis robotic arm, or a multi-arm structure.

4. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, The sports medicine planning and execution module is specifically used for: This is used to plan targeted surgeries for patients based on their medical images and cold-region-specific data, generating personalized bone tunnel screw placement plans; the cold-region-specific data is used to verify whether the bone quality of the tunnel wall meets the preset mechanical stability threshold. Based on the navigator and tracer, the intraoperative solid skeleton and the preoperative 3D skeleton model are registered to map the bone tunnel nail placement scheme and cold-region specific data. Based on the mapped bone tunnel pin placement scheme and cold-region specific data, the pin placement path of the robotic arm is planned, and the robotic arm is controlled to execute the pin placement path; the cold-region specific data is used to correct the pin placement path; the robotic arm is any one of the following: active-passive hybrid robotic arm, handheld robotic arm, six-axis / seven-axis robotic arm, and multi-arm structure.

5. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, The robotic arm is a hybrid active-passive robotic arm, comprising a passive mechanical segment and an active mechanical segment: The passive mechanical segment is used to work in conjunction with the operator. It is manually pushed to the target area and locked by the operator to provide coarse positioning of the end effector. The active mechanical segment consists of multiple servo-driven active joints, which are used to plan and execute the motion path of the end effector based on the registration mapping results within the locked positioning range.

6. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedic fields according to claim 1, characterized in that, The robotic arm is a handheld robotic arm, including a motion recognition module and a trajectory correction module; The motion recognition module is used to collect the surgeon's hand motion signals in real time and extract the intentional operation motion components as effective surgical motion commands. The trajectory correction module is used to compare the effective surgical movement command with the registered and mapped surgical path in real time, calculate the deviation vector and generate a correction torque to guide the surgeon's hand to move along the surgical path.

7. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, The robotic arm has a multi-arm structure, including a main operating arm and at least one auxiliary arm; The main operating arm performs predetermined surgical operations along the planned surgical path based on the registration mapping results; The auxiliary arm provides support during the surgical procedure performed by the main operating arm.

8. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, The cold-region-specific data shall include at least two of the following: epidemiological data on osteoporosis in cold regions, data on vitamin D deficiency and bone metabolism in cold regions, data on bone microstructure characteristics, and data on intraoperative tissue characteristics in low-temperature environments.

9. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedic fields according to claim 1, characterized in that, When controlling the robotic arm to execute the corresponding surgical plan, bone protection operations are automatically embedded in the surgical execution trajectory based on the microcrack protection motion strategy parameters determined by the cold-region specific data. The microcrack protection motion strategy includes at least one of the segmented feed strategy, pulse propulsion strategy, and drill unloading strategy.

10. The intelligent navigation and positioning system for cold-weather surgical robots in all orthopedics according to claim 1, characterized in that, It also includes an intraoperative processing system, which is used to personalize the design and processing of surgical instruments or fixation devices in the surgical plan based on the patient's preoperative imaging data and cold-region-specific bone data, so as to improve the fit between the surgical instruments or fixation devices and the individual patient.

11. A control method for an intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics, applicable to the intelligent navigation and positioning system for a cold-weather surgical robot in orthopedics as described in any one of claims 1-10, characterized in that, include: Based on the patient's medical data and cold-region-specific data, a targeted disease assessment is conducted to generate personalized auxiliary assessment results for the patient. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's joint replacement surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's spinal surgery. Based on the patient's post-traumatic medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are performed for the patient's trauma surgery. Based on the patient's medical data and cold-region-specific data, targeted surgical planning, registration mapping, and control of the robotic arm to execute the surgical plan are carried out for the patient's sports medicine surgery. Based on the patient's medical data, surgical plan, and cold-region-specific data, a personalized preoperative rehabilitation plan is generated for the patient. Furthermore, based on the patient's medical data, postoperative assessment results, and cold-region-specific data, a personalized postoperative rehabilitation plan is generated for the patient.

12. An electronic device, characterized in that, It includes a processor and a memory; the memory stores a computer program, which, when executed by the processor, implements the control method of the intelligent navigation and positioning system for cold-region surgical robots in orthopedics as described in claim 11.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the control method of the intelligent navigation and positioning system for the cold-region surgical robot in orthopedics as described in claim 11.

Images