Lumbar decompression operation real-time tracking and force feedback tactile detection method and system

Abstract

The invention provides a real-time tracking and force feedback tactile detection method and system for a lumbar decompression operation, and relates to the technical field of medical and industrial crossing. The method comprises the following steps: acquiring preoperative original imaging data of a patient; according to the original iconography data, constructing a patient operation area three-dimensional model; based on the patient operation area three-dimensional model, an operation simulation path is formulated; matching the real-time coordinates and the 3D virtual coordinates of the surgical instrument according to the surgical simulation path; tracking and positioning the surgical instrument based on a matching result; according to tracking and positioning results, the touch of the nerve root is detected in real time by holding a probe hook handle with a piezoelectric ceramic sensing element; based on the real-time detection result, a stress signal of the probe hook handle in the nerve root transverse pulling process is collected; carrying out quantization processing on the stress signal; establishing a nerve root elasticity quantitative model according to the stress signal after the quantitative processing; and quantizing the state of the nerve root touch sense through the nerve root elasticity quantification model to generate a touch sense level.

Description

TECHNICAL FIELD

[0001] The present application relates to the medical and engineering cross technology field, especially to a lumbar decompression surgery real-time tracking and force feedback tactile detection method and system. BACKGROUND

[0002] Based on artificial intelligence technology is observing, talking and calculating ability, for example, through the progress of computer vision and object recognition, artificial intelligence has vision, but, has not developed similar human touch, artificial intelligence is not good at measuring or generating "touch" on the surface of the object, artificial intelligence technology has not reached the level of "touch" intelligence of human beings.

[0003] Touch has an important role in surgical operation, the touch formed by the proprioception and the fine touch of the fingertips becomes the super ability obtained by the surgeon after long-term training, therefore, through real-time detection and quantitative analysis of the touch signal in the surgical process, the touch perception in the subjective experience of the surgeon can be converted into objective measurable mechanical indicators, the accurate determination of the tissue elasticity, tension and structure state is realized, so that the nerve damage is avoided, the safety and stability of the operation are improved, and data support is provided for the intelligent surgical robot and the intraoperative navigation system.

[0004] However, the existing spine surgery and surgical navigation platform is usually open or more mature pedicle screw placement and other processing methods, which may cause a larger wound and slower recovery. Secondly, with the various degenerative forms of lumbar degenerative diseases, the position is variable, and the patients are often multiple, so that there is a spatial registration error between the traditional image navigation and the intraoperative operation, it is difficult to finely identify and safely avoid the complex nerve roots and surrounding soft tissues, the surgeon can only rely on experience to judge the decompression operation, and lacks quantitative and real-time monitoring of the touch feedback, thereby increasing the risk of nerve damage and postoperative complications. SUMMARY

[0005] In order to solve the technical problems that the existing spine surgery and surgical navigation platform is usually open or more mature pedicle screw placement and other processing methods, which may cause a larger wound and slower recovery. Secondly, with the various degenerative forms of lumbar degenerative diseases, the position is variable, and the patients are often multiple, so that there is a spatial registration error between the traditional image navigation and the intraoperative operation, it is difficult to finely identify and safely avoid the complex nerve roots and surrounding soft tissues, the surgeon can only rely on experience to judge the decompression operation, and lacks quantitative and real-time monitoring of the touch feedback, thereby increasing the risk of nerve damage and postoperative complications, the present application provides a lumbar decompression surgery real-time tracking and force feedback tactile detection method and system.

[0006] The technical scheme provided by the embodiments of the present application is as follows: The first aspect of the embodiment of the present application provides a lumbar decompression surgery real-time tracking and force feedback tactile detection method, which comprises the following steps: S1: acquiring original imaging data of a patient before surgery; S2: constructing a three-dimensional model of a surgical area of the patient according to the original imaging data by a three-dimensional reconstruction algorithm; S3: formulating a surgical simulation path based on the three-dimensional model of the surgical area of the patient; S4: matching real-time coordinates of a surgical instrument and 3D virtual coordinates according to the surgical simulation path; S5: tracking and positioning the surgical instrument based on the matching result; S6: detecting nerve root touch feeling in real time by a probe handle with a piezoceramic sensing element based on the tracking and positioning result; S7: collecting force signals of the probe handle in the process of horizontally moving the nerve root based on the real-time detection result; S8: quantitatively processing the force signals; S9: establishing a nerve root elasticity quantitative model according to the quantitatively processed force signals; S10: quantifying the state of the nerve root touch feeling by the nerve root elasticity quantitative model to generate a touch feeling grade.

[0007] The second aspect of the embodiment of the present application provides a lumbar decompression surgery real-time tracking and force feedback tactile detection system, which comprises: a processor; a memory, wherein the memory stores computer readable instructions, and the computer readable instructions are executed by the processor to realize the lumbar decompression surgery real-time tracking and force feedback tactile detection method of the first aspect.

[0008] The third aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by the processor to realize the lumbar decompression surgery real-time tracking and force feedback tactile detection method of the first aspect.

[0009] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: In the embodiment of the present application, by establishing a three-dimensional model of the patient's operation area based on preoperative image reconstruction, and realizing real-time coordinate tracking of surgical instruments and virtual space registration during operation, and combining the probe handle with piezoelectric ceramic sensing elements to detect and quantitatively analyze the real-time touch of the nerve root, the subjective touch perception of the surgeon can be converted into objective stress and elasticity parameters, and real-time evaluation and touch level determination of the nerve root state can be realized. Therefore, not only the accuracy and safety of intraoperative operation are significantly improved, and the risk of nerve damage is reduced, but also a quantifiable and visualized auxiliary decision basis is provided for minimally invasive decompression surgery, thereby effectively improving the intelligent and individualized level of lumbar decompression surgery. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0011] Figure 1 A flowchart of a lumbar decompression surgery real-time tracking and force feedback tactile detection method provided by the embodiment of the present application.

[0012] Figure 2 A structure diagram of a lumbar decompression surgery real-time tracking and force feedback tactile detection system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0013] The technical solutions in the present application will be described below with reference to the drawings.

[0014] In the embodiments of the present application, the words such as "example", "for example" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0015] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0016] In the embodiments of the present application, sometimes the subscript such as W1 may be written in the form of non-subscript such as W1, and when the difference is not emphasized, the meanings expressed are consistent.

[0017] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0018] Reference is made to the accompanying drawings Figure 1 , which shows a flowchart of a lumbar decompression surgery real-time tracking and force feedback haptic detection method provided by an embodiment of the present application.

[0019] The present application provides a lumbar decompression surgery real-time tracking and force feedback haptic detection method, which can be realized by a lumbar decompression surgery real-time tracking and force feedback haptic detection device. The lumbar decompression surgery real-time tracking and force feedback haptic detection device can be a terminal or a server. The processing flow of the lumbar decompression surgery real-time tracking and force feedback haptic detection method can include the following steps:

[0020] S1: Obtain the original imaging data of the patient before surgery.

[0021] In one possible implementation, the original imaging data includes high-resolution CT images, MRI images, DICOM meta information, X-ray positioning films, and marker points and transformation matrices for spatial registration.

[0022] It should be noted that by systematically obtaining the preoperative multi-modal imaging data of the patient, the comprehensive information integration of the surgical area structure is realized. It not only provides accurate bone anatomical features represented by high-resolution CT images, but also fuses the details of soft tissues and nerve roots contained in MRI images, thereby constructing a basic data set with rich tissue levels. With the help of DICOM meta information and spatial marker points, the geometric consistency between different imaging modalities can be ensured, providing accurate spatial reference for subsequent three-dimensional reconstruction and path planning. Compared with the traditional method of relying only on single CT data, this step significantly improves the completeness of the data and the restoration accuracy of the model, lays a reliable digital foundation for intraoperative navigation and haptic detection, ensures the high correspondence between surgical simulation and actual operation, and improves the safety and individualized precision of the surgery.

[0023] S2: According to the original imaging data, a three-dimensional model of the surgical area of the patient is constructed by a three-dimensional reconstruction algorithm.

[0024] The three-dimensional reconstruction algorithm refers to an algorithm for converting a two-dimensional medical image sequence (such as a CT or MRI slice) into model data that can be displayed and calculated in three-dimensional space. The three-dimensional model of the patient's surgical area refers to a digital model generated according to the reconstruction algorithm, reflecting the individual anatomical features of the patient, including the geometric and topological structure information of the vertebral body, intervertebral disc, nerve root, spinal canal and other regions. The model is usually represented in the form of a triangular mesh or a point cloud, and is the core data structure for preoperative simulation and intraoperative navigation.

[0025] It should be noted that the two-dimensional image data of the patient is converted into a high-fidelity three-dimensional surgical area model by the three-dimensional reconstruction algorithm, which realizes the key transition from planar image to three-dimensional visual structure, can accurately restore the spatial relationship of bony structures, nerve root running and surrounding soft tissues, and provides a true anatomical basis for surgical planning. Compared with the traditional two-dimensional image comparison method, this step greatly improves the spatial recognition and operational safety, and provides a solid data foundation for individualized minimally invasive lumbar surgery.

[0026] In one possible implementation, S2 specifically includes: S201: Extracting a set of cortical bone layer voxels of the high-resolution CT image by threshold segmentation:

[0027] wherein, represents the set of cortical bone layer voxels, i.e., the set of target region voxels, x , y , z represents the image voxel coordinate point, represents the original three-dimensional image voxel space, I represents the gray intensity function, T c represents the threshold value.

[0028] The threshold segmentation is a gray value-based image segmentation method, which separates the pixels (usually corresponding to bone tissue) higher than the threshold value from the background by setting the gray threshold value, extracts the set of cortical bone layer voxels, and realizes the preliminary identification of the target tissue.

[0029] S202: Processing the set of cortical bone layer voxels by the Marching Cubes algorithm to generate a triangular mesh.

[0030] The Marching Cubes algorithm is a classic three-dimensional surface reconstruction algorithm that detects isosurface in voxel data and generates a triangular mesh by detecting isosurface in voxel data, which can convert discrete CT voxel points into a continuous polygonal surface model.

[0031] S203: Extracting the curvature extreme lines as the set of lesion boundary curves based on the lesion region on the triangular mesh.

[0032] where the curvature extreme line is the position where the geometry of the surface changes, and is used to define the geometric features of the lesion or bony boundary. Extracting these curves can obtain the spatial representation of the lesion profile.

[0033] S204: Calculating the mean curvature and Gaussian curvature of the set of lesion boundary curves, and taking the region where the mean curvature exceeds the threshold as the candidate region of the bony boundary.

[0034] S205: Equidistantly sampling each lesion boundary curve in the set of lesion boundary curves, and performing weighted least squares B-spline fitting on each sampling point in the local window:

[0035] where min denotes minimization, p i denotes the center point of the current curve to be fitted (i.e. the lesion boundary curve), w s denotes the weighting coefficient, p i+s denotes the neighborhood sampling point, i.e. the center point p i left and right d sampling points, respectively, constitute the local fitting window, denotes the value of the B-spline fitting function at point p i+s , i.e. the predicted value of the neighborhood sampling point by the spline function p i+s , s denotes the neighborhood index, d denotes the fitting window radius, denotes the local cubic B-spline curve established with the points in the p i window as control points near the center point i - d , i + d .

[0036] where the weighted least squares B-spline fitting is a local smoothing fitting method that controls the contribution of different neighborhood sampling points through a weight function, so that the fitted curve can smooth the noise and retain the boundary details, and is an important guarantee for shape accuracy in the reconstruction process.

[0037] S206: Determining the concave-convex sign based on the weighted least squares B-spline fitting result:

[0038] wherein, k s p i denotes the signed curvature, i.e. the curvature value with sign at the center point p i denotes the curvature vector, i.e. the curvature vector computed on the local B-spline curve whose neighborhood range is [-1, 1] at the center point p i d d denotes the curvature vector computed on the local B-spline curve, sign() denotes the sign function, p i denotes the unit normal vector of the curve at the center point

[0039] wherein, the curvature value with sign (positive / negative) is used to distinguish the concave / convex direction.

[0040] S207: constructing the scale-curvature function according to the concave / convex sign:

[0041] wherein, denotes the boundary curve denotes the curvature scale function value of the boundary curve under the scale d , i.e. denotes the point on the m-th boundary curve under the scale j i p i denotes the signed curvature value of the boundary curve under the scale d , i.e. denotes the point on the m-th boundary curve under the scale D denotes the scale set, denotes the m-th boundary curve. j

[0042] wherein, the scale-curvature function represents the curvature variation law under different observation scales, and a multi-scale geometric feature model is established to prevent over-smoothing or feature loss.

[0043] S208: mapping the scale-curvature function and the CSS feature corresponding to the scale-curvature function to the triangular mesh surface, and reconstructing the surface with curvature continuity as the constraint to construct the patient operation area three-dimensional model:

[0044] wherein, denotes the patient operation area three-dimensional model, x i denotes the point coordinate vector on the surface of the patient operation area three-dimensional model, i ​​​​​​​​​​denotes Laplacian operator (second order gradient), i.e. surface point x i second order spatial derivative (i.e. surface curvature degree) of denotes weight factor, denotes curvature scale function value of patient operation area three-dimensional model point at scale d i.e. point on patient operation area three-dimensional model corresponding to boundary curve at scale x i signed curvature value at scale d denotes target curvature mean value, i.e. denotes average curvature of boundary curve at scale d

[0045] wherein, CSS feature is curvature scale space feature, which is a feature mapping describing the change of curve shape with scale, and is used to capture stable features of boundary shape.

[0046] It should be noted that, first, threshold segmentation combined with gray scale feature automatically extracts bony region, avoiding errors caused by manual contouring, and improving reconstruction efficiency and objectivity. Second, the triangular mesh generated by the Marching Cubes algorithm ensures the topological consistency of the model surface, providing a standardized structure for subsequent geometric analysis. Third, by extracting the extreme value line of the curvature of the lesion area and fitting the B-spline weighted least squares, not only the refined reconstruction of the boundary is realized, but also the influence of noise is suppressed and the continuity of anatomical features is maintained. After introducing the signed curvature and scale-curvature function, the model has multi-scale geometric expression ability and can maintain stable boundary recognition effect at different resolutions. Finally, the Laplace surface reconstruction constrained by curvature continuity effectively balances smoothness and structure fidelity, making the generated operation area model closer to the real anatomical structure in spatial form. In summary, this step has the significant advantages of high precision, good feature preservation, strong noise resistance, and good geometric continuity, providing a high-quality digital foundation for subsequent path planning, force feedback detection, and surgical navigation.

[0047] S3: Based on the patient operation area three-dimensional model, formulate the surgical simulation path.

[0048] wherein, the surgical simulation path is an ideal operation path determined based on the surgical target, approach method and anatomical constraints in the virtual three-dimensional model. It defines the spatial trajectory of the surgical instrument entering, moving and acting, and is the reference standard for navigation and real-time tracking in surgery.

[0049] ​​Specifically, first, by setting the entry point, target point and safety boundary in the virtual space, the system can automatically generate multiple candidate paths, and select the optimal path according to the path length, angle, obstacle avoidance degree and other indicators, so as to ensure that the operation path is completed in the shortest and safest range. Secondly, with the help of accurate anatomical information of the three-dimensional model, the path can avoid key structures such as nerve root, blood vessel and spinal canal, significantly reducing the risk of misentry or injury during operation. Thirdly, through virtual simulation, the feasibility of the path and the accessibility of the surgical instrument can be verified in advance, and quantitative comparison of different schemes can be realized. Finally, the implementation of this step changes the operation planning from experience dependence to data-driven, providing accurate reference path for subsequent real-time tracking and tactile feedback analysis.

[0050] S4: According to the surgical simulation path, the real-time coordinates of the surgical instrument and the 3D virtual coordinates are matched.

[0051] Among them, the real-time coordinates refer to the spatial coordinate data of the end point, axis and attitude of the surgical instrument obtained in real time through the intraoperative positioning system (such as optical or electromagnetic tracking equipment). These data are continuously updated during the operation to reflect the real-time position and motion state of the instrument.

[0052] Among them, the 3D virtual coordinates refer to the geometric positions of each point in the virtual space in the three-dimensional model coordinate system established before the operation, which are used to represent the spatial information of the patient's anatomical structure and the surgical path.

[0053] In one possible implementation, S4 specifically includes: S401: Obtain intraoperative three-dimensional observation coordinates.

[0054] Among them, the intraoperative three-dimensional observation coordinates refer to the three-dimensional position coordinates of the instrument or marker point in the physical space obtained in real time through the intraoperative optical or electromagnetic tracking system. This data reflects the dynamic pose information of the instrument during the actual operation.

[0055] S402: According to the intraoperative three-dimensional observation coordinates, combine the template marker coordinates of the surgical instrument, and perform rigid body transformation:

[0056] Among them, R represents the optimal rotation matrix, K represents the number of marker points, represents the coordinates of the first marker point under the optimal rotation matrix R acting on the instrument's own coordinate system T , represents the coordinates of the th marker point, represents the set of legal three-dimensional rotation matrices, represents t the moment in the tracking coordinate system CThe three-dimensional coordinates of the real-time detected marker points. k

[0057] The template marker coordinates refer to spatial marker points preset on the surgical instrument, which have fixed known positions in the instrument coordinate system, and are used to establish the spatial correspondence between the instrument coordinate system and the tracking coordinate system.

[0058] The rigid body transformation is a spatial coordinate mapping method that maintains the geometric shape unchanged, including a rotation matrix R and a translation vector t, which is used to convert points in one coordinate system (such as the instrument system) to another coordinate system (such as the tracking system).

[0059] S403: According to the rigid body transformation result, the calibration relationship between the intraoperative tracking coordinate system and the preoperative virtual coordinate system is determined by combining the consistency of the intraoperative coordinates and the three-dimensional model coordinates of the patient's surgical area:

[0060] wherein, J represents the number of corresponding point pairs, represents the three-dimensional coordinates of the i-th marker point in the tracking coordinate system, j C represents the three-dimensional coordinates of the i-th marker point in the three-dimensional model coordinate system of the patient's surgical area, j V represents the squared Euclidean distance.

[0061] The intraoperative tracking coordinate system refers to a real-time spatial coordinate system defined by the navigation system, which is a reference for intraoperative positioning, and is used to describe the position of the marker point or the instrument in the physical surgical space.

[0062] The preoperative virtual coordinate system is a coordinate system defined by the preoperative three-dimensional model, which is used to represent the position of the patient's anatomical structure in the virtual space. The coordinate system and the tracking coordinate system establish a correspondence through spatial calibration.

[0063] S404: According to the calibration relationship, the instrument end point is mapped from the instrument coordinate system to the three-dimensional model coordinate system of the patient's surgical area.

[0064] S405: Based on the mapping result, the real-time coordinates and 3D virtual coordinates of the surgical instrument are matched in combination with the surgical simulation path.

[0065] ​​​​​It should be noted that first, through rigid body transformation, the template markers of surgical instruments can be accurately aligned with the output data of the tracking system, ensuring the consistency of the spatial expression of the instrument posture. Second, the calibration relationship between the intraoperative tracking coordinate system and the preoperative virtual model coordinate system is established through least squares optimization, which strictly corresponds the virtual model to the actual patient position, laying the mathematical foundation for real-time navigation. Third, through coordinate mapping, the instrument endpoint is visualized in the virtual model, realizing the bidirectional mapping of "real operation-digital model", so that the surgeon can observe the position relationship of the instrument relative to the key structures such as nerve root and spinal canal in the navigation interface in real time. Finally, through the dynamic matching of path geometric indicators and error monitoring, the system can continuously detect the deviation between the instrument motion and the planned path during the operation, and timely issue a risk warning.

[0066] S5: Based on the matching result, the surgical instrument is tracked and positioned.

[0067] Wherein, the matching result refers to the position correspondence of the surgical instrument in the virtual three-dimensional model coordinate system after coordinate calibration and spatial mapping. The result includes rotation matrix, translation vector and real-time spatial coordinates of the instrument endpoint in the model of the operation area.

[0068] It should be noted that first, based on the coordinate mapping relationship obtained by matching, the system can track the position and attitude of the instrument in three-dimensional space in real time, enabling the surgeon to intuitively grasp the spatial relationship between the instrument and the key structures in the operation area (such as nerve root, spinal canal and lesion area) in the navigation interface, significantly improving the spatial accuracy of the operation. Second, the continuous coordinate tracking and time sequence updating mechanism enables the system to refresh the instrument motion trajectory at a high frequency, ensuring that the virtual model display is synchronized with the actual operation, and avoiding positioning errors caused by delay or drift. Third, by combining error compensation algorithm and attitude stability control, this step can effectively suppress the deviation caused by sensor noise or slight changes in patient position, ensuring the navigation signal.

[0069] S6: According to the tracking and positioning results, the real-time detection of nerve root touch is realized by holding the probe handle with piezoelectric ceramic sensing elements.

[0070] Wherein, the tracking and positioning results refer to the real-time position, direction and attitude information of the surgical instrument in three-dimensional space obtained in step S5, which is the basic data for realizing the spatial correspondence of mechanical detection.

[0071] Wherein, the probe handle is an operating tool used by the surgeon to touch or pull the nerve root, spinal canal wall and other soft tissues during the operation.

[0072] Wherein, the nerve root refers to the nerve root part of the spinal cord, which is the key structure most needed to be protected in lumbar decompression surgery, and its mechanical state (tension or relaxation) is directly related to the safety and efficacy of the operation.

[0073] It should be noted that first, the piezoelectric ceramic sensing element can respond to extremely small mechanical changes, achieving millinewton-level resolution of force detection, enabling doctors to obtain quantitative mechanical feedback information in addition to visual navigation. Second, the system can continuously monitor the mechanical changes of the probe during the process of lateral displacement of the nerve root, automatically identify abnormal tactile sensation (such as excessive tension or transient instability), and assist doctors in judging the tension of the tissue and the safety boundary. At the same time, the piezoelectric ceramic sensing element has the advantages of small size, fast response, and strong anti-electromagnetic interference capability, ensuring the real-time and stability of the detection data.

[0074] S7: Based on the real-time detection results, the force signal of the probe handle during the process of lateral displacement of the nerve root is collected.

[0075] Among them, the force signal refers to the voltage signal output by the piezoelectric sensing element after the probe is subjected to the counterforce of the tissue during the lateral displacement operation. After amplification, filtering and digitization, the signal forms time series data for analyzing the mechanical response characteristics of the nerve root.

[0076] It should be noted that this step can convert the subjective tactile sensation of the surgeon during the lateral displacement of the nerve root into measurable and analyzable objective physical quantities, accurately capturing the elastic change, tension fluctuation and transient response of the nerve root under the action of force in the form of time series signal, thereby breaking through the limitations of traditional surgery relying on experience judgment. At the same time, the piezoelectric ceramic sensing element embedded in the probe handle has a compact structure and does not change the original operation habit, ensuring the natural feedback of the doctor's hand feeling. Through this step, the system can record the force signal and the spatial coordinates of the instrument simultaneously, realizing one-to-one correspondence between the tactile signal and the anatomical position of the operation area.

[0077] S8: Quantitative processing of the force signal.

[0078] Quantitative processing refers to mathematical and signal processing operations on the original force signal, which converts it from the original voltage value or digital sampling value to quantized data with physical meaning (such as Newton, Pascal or dimensionless elastic coefficient). This process usually includes gain calibration, filtering and denoising, normalization and robust feature extraction.

[0079] It should be noted that compared with traditional subjective experience judgment, this quantitative process significantly improves the objectivity and repeatability of tactile perception. Through systematic quantitative processing of the force signal collected by the probe handle, the conversion from the original electrical signal to structured physical parameters is realized, providing high-quality input data for subsequent nerve root elasticity modeling and tactile grading.

[0080] In one possible implementation, the quantitative processing includes gain calibration processing, filtering processing, and feature extraction processing.

[0081] The specific calculation formula for gain calibration is as follows:

[0082]

[0083] in, This indicates that the piezoelectric sensor inside the probe handle is at the first... n The corrected force signal output at each sampling time. This indicates that the piezoelectric sensor inside the probe handle is at the first... n The force signal output at each sampling time, This represents the mean of zero drift. This represents the gain calibration factor, which is the proportionality coefficient between the sensor output and the actual force applied. It is used to convert voltage or digital signals into physical units. N 0 indicates the number of zero-point samples.

[0084] S9: Based on the quantized force signal, establish a quantized model of nerve root elasticity.

[0085] Among them, the nerve root elasticity quantification model refers to modeling the relationship between quantified force signals and corresponding tissue response characteristics (displacement, tactile level, etc.) to obtain a mathematical function or predictive model that can characterize the elastic properties of nerve roots. Its output is usually the elastic modulus, equivalent stiffness coefficient, or comprehensive elasticity index.

[0086] It should be noted that by modeling the quantified force signal and establishing a nerve root elasticity quantification model, the transformation from mechanical signal to tissue biological characteristic parameters is realized. This is the core link of the present invention to realize intelligent tactile recognition. The model can express the relationship between the force signal and the mechanical response of the nerve root in mathematical form, transforming the tactile judgment that originally relied on the doctor's experience into a calculable and analyzable physical index, thereby realizing the objectification and standardization of tactile perception.

[0087] In one possible implementation, S9 specifically includes: S901: Based on the quantized force signal, subjective tactile sensation is determined by constructing a window-level mechanical observation vector and a design matrix.

[0088] Among them, the window-level mechanical observation vector refers to selecting an observation window of a specific length (such as 0.1–0.5s) in the time series, and combining the force characteristics (such as average force, peak force, and rise slope) within the window into a multi-dimensional vector to describe the mechanical state of the nerve root within that time segment.

[0089] The design matrix is ​​a mathematical structure used to establish the relationship between input variables (mechanical features) and output variables (subjective tactile sensation). Each row represents the observation data for a time window, and each column corresponds to a mechanical feature.

[0090] S902: By using the weighted least squares method, the window-level mechanical observations are fitted to determine the objective mechanical strength index of the window.

[0091] S903: The elasticity quantification index is determined by weighting subjective tactile sensation and objective mechanical strength index through a Bayesian fusion algorithm with inverse error weighting.

[0092] S904: Perform time-series smoothing on the elasticity quantification index to obtain a smoothed elasticity index.

[0093] S905: Based on the smooth elasticity index, a quantitative model of nerve root elasticity is established by integrating mechanical parameters and tactile level.

[0094] in, Indicates the first r A smooth elasticity quantification index for each time window. This represents the mapping function of the nerve root elasticity quantization model. k (r) Indicates the first r The equivalent stiffness coefficient for each time window. c (r) Indicates the first r The equivalent viscosity coefficient for each time window. s (r) Indicates the first r The subjective touch level of each time window, Indicates the first r The residual terms for each time window.

[0095] It should be noted that, firstly, by using windowing and weighted least squares algorithms, the instantaneous fluctuations of the force signal are smoothed out, with each time window corresponding to a stable mechanical observation vector, effectively improving the robustness and temporal consistency of the data. Secondly, a Bayesian fusion algorithm is introduced, using an error-inverse weighting mechanism to adaptively integrate the doctor's subjective tactile judgment with the objective mechanical measurement results from the sensors, preserving the reliable judgment of clinical experience while ensuring the physical interpretability of the model. Thirdly, the temporal smoothing mechanism can suppress high-frequency noise such as operational jitter and respiratory interference, making the elasticity quantification index show a continuous and traceable trend, thus better reflecting the actual physiological state of the nerve root.

[0096] S10: The state of tactile sensation of nerve roots is quantified by using a nerve root elasticity quantification model to generate tactile levels.

[0097] wherein the tactile level is a kind of elastic response-based tissue tactile grading mechanism, which converts physiological characteristics (tension and tension of nerve root) into quantifiable and determinable level output, realizes the leap from "subjective touch" to "objective determination", and provides data-based basis for safe decision and fine operation in lumbar decompression surgery.

[0098] It should be noted that first, this step can convert the subjective touch experience of the doctor into measurable and comparable numerical indicators, standardize and quantify the traditional experience-dependent tactile judgment process, and reduce the risk of misjudgment caused by operator differences. Secondly, through the mapping relationship between the elastic index and the tactile level output by the model, the doctor can know the stress state and safety interval of the nerve root in real time, realize intraoperative risk visual warning, and avoid nerve damage caused by excessive traction or insufficient decompression. The method combines continuous index and grading determination, which can reflect the details of the elastic change of the nerve root, and has intuitive grading results, which is convenient for quick understanding during operation.

[0099] In one possible implementation, S10 specifically includes: S1001: mapping the smooth elastic index to the discrete elastic level by the nerve root elastic quantification model.

[0100] wherein the smooth elastic index refers to the elastic quantification result after time series smoothing processing. Its value represents the comprehensive elasticity degree of the nerve root within a certain time window, and is a stable index after eliminating operation noise and short-term fluctuations. The trend of the smooth elastic index directly reflects the physiological state transition of the nerve root from tight to relaxed.

[0101] wherein the discrete elastic level refers to the discrete interval result after dividing the continuous elastic index. For example, when the smooth elastic index range is [0, 1.5], multiple thresholds can be set to divide it into "high stiffness", "medium stiffness", "low-medium stiffness" and "low stiffness" four levels, which are used to represent the mechanical state of the tissue from tight to relaxed. The discretization result is the direct basis for subsequent tactile level determination.

[0102] S1002: determining the current touch state interval of the nerve root according to the discrete elastic level.

[0103] wherein the touch state interval refers to the physiological state range corresponding to the discrete elastic level, which is an intermediate layer expression from mechanical data to tactile perception. Each interval represents a touch experience area, such as "tight", "relatively tight", "relatively loose" and "relaxed". It plays a bridge role between "data-tactile semantics" in the model.

[0104] S1003: determining the tactile level according to the current touch state interval.

[0105] The tactile level refers to a standardized level result finally generated according to the tactile state interval, and is used to intuitively express the current tactile state of the nerve root.

[0106] S1004: Determine the confidence of each level in the tactile level in combination with the internal residual error of the nerve root elasticity quantification model and the quality of the stress signal.

[0107] The confidence is a kind of statistical quantity, which is used to measure the reliability of the current tactile level judgment result. The calculation of the confidence comprehensively considers factors such as model residual error, signal quality and algorithm stability. The higher the confidence is, the more reliable the tactile level determination is, and it is suitable for intraoperative real-time display and auxiliary decision-making.

[0108] It should be noted that the continuous smooth elasticity index is converted into discrete levels through the mapping mechanism, so that the complex mechanical parameter results are intuitively presented in the form of simple classification, greatly improving the readability of intraoperative information and decision-making efficiency. Secondly, according to the corresponding relationship between the discrete levels and the tactile state interval, the current tactile level of the nerve root is automatically determined, and the doctor does not need to rely on experience to quickly judge whether the decompression is sufficient and whether the nerve root is safe. Thirdly, the confidence correction mechanism of the model internal residual error and the stress signal quality can automatically reduce the weight of abnormal data or low signal-to-noise ratio signals, thereby enhancing the resistance of the system to interference and the stability of the output results. Through this multi-layer verification mechanism, the system can provide the tactile level and the corresponding confidence range at the same time, so that the doctor can intuitively judge the reliability of the results during the operation, and avoid misoperation.

[0109] In one possible implementation, the tactile level includes level I, level II, level III and level IV.

[0110] Level I is a tight state.

[0111] Level II is a relatively tight state.

[0112] Level III is a relatively relaxed state.

[0113] Level IV is a relaxed state.

[0114] It should be noted that the tactile level classification realizes the standardization and visual expression of tactile cognition by classifying the stress state and elastic response of the nerve root into four levels of I to IV, and has significant technical and clinical application advantages. First, this grading mechanism is based on the stress characteristics of the nerve root, and converts the continuous elastic quantization result into discrete level output, so that the doctor can intuitively judge the current tissue state without relying on complex numerical calculation, thereby greatly improving the intraoperative information recognition efficiency. Secondly, the classification of I to IV covers the complete physiological interval from "tight" to "relaxed", which can accurately reflect the dynamic process of the nerve root from compression to complete decompression, and provides a clear standard for judging the sufficiency of decompression.

[0115] In the embodiment of the present application, by establishing a three-dimensional model of the patient's operation area based on preoperative image reconstruction, and realizing real-time coordinate tracking and virtual space registration of surgical instruments during operation, and combining the real-time detection and quantitative analysis of the nerve root tactile sensation of the probe handle with piezoelectric ceramic sensing elements, the subjective tactile perception of the surgeon can be converted into objective stress and elastic parameters, and real-time evaluation and tactile level determination of the nerve root state can be realized. Therefore, not only the accuracy and safety of intraoperative operation are significantly improved, and the risk of nerve injury is reduced, but also a quantifiable and visualized auxiliary decision-making basis is provided for minimally invasive decompression surgery, thereby effectively improving the intelligent and individualized level of lumbar decompression surgery.

[0116] Referring to the accompanying drawings Figure 2 of the specification, a structure schematic diagram of a lumbar decompression surgery real-time tracking and force feedback tactile detection system provided by the present application is shown.

[0117] The present application also provides a lumbar decompression surgery real-time tracking and force feedback tactile detection system 20, which is applied to the lumbar decompression surgery real-time tracking and force feedback tactile detection method described above, and comprises: a processor 201.

[0118] a memory 202, the memory 202 stores computer readable instructions, and when the computer readable instructions are executed by the processor 201, the lumbar decompression surgery real-time tracking and force feedback tactile detection method of the method embodiment is realized.

[0119] The lumbar decompression surgery real-time tracking and force feedback tactile detection system 20 provided by the present application can execute the lumbar decompression surgery real-time tracking and force feedback tactile detection method described above, and realize the same or similar technical effects. To avoid repetition, the present application will not be described again.

[0120] It should be appreciated that a processor in the embodiments of the present application can be a central processing unit (CPU). The processor can also be other general purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general purpose processor can be a microprocessor or the processor can be any conventional processor.

[0121] It should also be appreciated that the memory in the embodiments of the present application can be volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. In non-volatile memory, a read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM) or flash memory can be used. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, and not limitation, many forms of random access memory (RAM) can be used, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM) and direct rambus RAM (DR RAM).

[0122] The above-described embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through a wired (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0123] It should be understood that the term "and / or" herein merely describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents that the associated objects before and after it are in an "or" relationship, but it can also represent an "and / or" relationship, which can be understood according to the context before and after it.

[0124] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0125] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined according to their functions and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0126] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0128] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0129] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0130] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0131] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0132] The embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the lumbar decompression surgery real-time tracking and force feedback tactile detection method.

[0133] The computer readable storage medium provided by the present application can realize the steps and effects of the lumbar decompression surgery real-time tracking and force feedback tactile detection method of the above-mentioned method embodiment. To avoid repetition, the present application will not be described again.

[0134] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0135] The following points need to be explained: (1) The drawings of the embodiments of the present application only involve the structures related to the embodiments of the present application, and other structures can be referred to the general design.

[0136] (2) For the sake of clarity, the thickness of the layers or regions is exaggerated or reduced in the drawings used to describe the embodiments of the present application, that is, the drawings are not drawn according to the actual proportion. It can be understood that when an element such as a layer, a film, a region or a substrate is referred to as being located "on" or "under" another element, the element can be "directly" located on or under another element or there can be an intermediate element.

[0137] (3) In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other to obtain new embodiments.

[0138] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for real-time tracking and force feedback tactile detection during lumbar decompression surgery, characterized in that, include: S1: Obtain the patient's original preoperative imaging data; S2: Based on the original imaging data, a three-dimensional model of the patient's surgical area is constructed using a three-dimensional reconstruction algorithm; S3: Based on the three-dimensional model of the patient's surgical area, formulate the surgical simulation path; S4: Match the real-time coordinates and 3D virtual coordinates of the surgical instruments according to the surgical simulation path; S5: Based on the matching results, the surgical instruments are tracked and located; S6: Based on the tracking and positioning results, the nerve root tactile sensation is detected in real time by holding a probe handle with a piezoelectric ceramic sensing element. S7: Based on real-time detection results, collect the force signal of the probe handle during the transverse prying of the nerve root; S8: Quantize the force signal; S9: Based on the quantized force signal, establish a quantized model of nerve root elasticity; S10: The state of tactile sensation of the nerve root is quantified by the nerve root elasticity quantification model to generate a tactile level.

2. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 1, characterized in that, The original imaging data includes: high-resolution CT images, MRI images, DICOM metadata, X-ray localization films, and markers and transformation matrices for spatial registration.

3. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 2, characterized in that, S2 specifically includes: S201: Extract the cortical bone layer voxel set of the high-resolution CT image by threshold segmentation; S202: The cortical bone layer voxel set is processed using the Marching Cubes algorithm to generate a triangular mesh; S203: Based on the lesion region on the triangular mesh, extract the curvature extremum lines as a set of lesion boundary curves; S204: Calculate the average curvature and Gaussian curvature of the lesion boundary curve set, and take the region with the average curvature exceeding the threshold as the candidate region of bony boundary; S205: Sample each lesion boundary curve in the lesion boundary curve set at equal intervals, and perform weighted least squares B-spline fitting on each sampling point within a local window; S206: Determine the concavity / convexity sign based on the weighted least squares B-spline fitting results; S207: Construct a scale-curvature function based on the concave-convex symbols; S208: Map the scale-curvature function and the corresponding CSS features to the triangular mesh surface, and reconstruct the surface with curvature continuity as a constraint to build a three-dimensional model of the patient's surgical area.

4. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 1, characterized in that, S4 specifically includes: S401: Obtain intraoperative three-dimensional observation coordinates; S402: Based on the intraoperative three-dimensional observation coordinates and the template mark coordinates of the surgical instruments, perform rigid body transformation; S403: Based on the rigid body transformation results and the consistency between the intraoperative coordinates and the coordinates of the three-dimensional model of the patient's surgical area, determine the calibration relationship between the intraoperative tracking coordinate system and the preoperative virtual coordinate system. S404: Based on the calibration relationship, map the instrument endpoint from the instrument coordinate system to the patient surgical area three-dimensional model coordinate system; S405: Based on the mapping results and combined with the surgical simulation path, match the real-time coordinates and 3D virtual coordinates of the surgical instruments.

5. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 1, characterized in that, The quantization process includes: gain calibration, filtering, and feature extraction.

6. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 1, characterized in that, S9 specifically includes: S901: Based on the quantized force signal, subjective tactile sensation is determined by constructing a window-level mechanical observation vector and a design matrix; S902: The window-level mechanical observations are fitted using the weighted least squares method to determine the objective mechanical strength index of the window; S903: The subjective tactile sensation and the objective mechanical strength index are weighted and fused using a Bayesian fusion algorithm with inverse error weighting to determine the elasticity quantification index; S904: Perform time-series smoothing on the elasticity quantification index to obtain a smoothed elasticity index; S905: Based on the smooth elasticity index, and by integrating mechanical parameters and tactile level, establish the nerve root elasticity quantification model.

7. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 6, characterized in that, S10 specifically includes: S1001: The smooth elasticity index is mapped to discrete elasticity levels using the nerve root elasticity quantization model. S1002: Determine the current tactile state range of the nerve root based on the discrete elasticity level; S1003: Determine the tactile level based on the current tactile state range; S1004: Combine the internal residuals of the nerve root elasticity quantization model with the quality of the force signal to determine the confidence level of each level in the tactile level.

8. The method for real-time tracking and force feedback tactile detection in lumbar decompression surgery according to claim 7, characterized in that, The tactile levels include Level I, Level II, Level III, and Level IV; Level I is a state of tension; Level II is a relatively tight state; Level III is a relatively relaxed state; Level IV is a relaxed state.

9. A real-time tracking and force feedback tactile detection system for lumbar decompression surgery, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the method for real-time tracking and force feedback tactile detection of lumbar decompression surgery as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for real-time tracking and force feedback tactile detection of lumbar decompression surgery as described in any one of claims 1 to 8.

Images