An intelligent identification method for determining the trajectory of anal fistula nail implantation

Abstract

The present invention relates to the field of medical imaging and computer-aided diagnosis and treatment, and in particular to a method for determining the trajectory of anal fistula nail implantation by intelligent identification. The content includes: performing multi-scale pyramid decomposition processing on the acquired image data of the anal fistula area, capturing the changes in spatiotemporal features, and generating multi-level dynamic sequence features; fusing the multi-level dynamic sequence features, performing dimensionality reduction processing on the generated shared space representation, and then performing time filtering and smoothing processing to obtain smoothed space-time features; extracting the target trajectory from the smoothed space-time features, and performing path optimization to obtain the optimal implantation path of the anal fistula nail; and performing path correction based on the optimal implantation path of the anal fistula nail. The method solves the problems that the traditional path planning method fails to accurately reflect the complex anatomical structure, affecting the accuracy and safety of the operation; fails to effectively capture the dynamic changes in the anal fistula area, resulting in inflexible path planning; lacks real-time feedback and correction mechanism, and cannot perform dynamic adjustment and correction.

Description

Technical Field

[0001] The present invention relates to the fields of medical imaging and computer-aided diagnosis and treatment, and in particular to a method for determining anal fistula nail implantation trajectory by intelligent identification. Background Art

[0002] Anal fistula is a common anorectal disease, usually accompanied by abnormal channels or fistulas between the anal canal and the tissues around the anus, which need to be treated surgically. During the operation, the precise implantation of anal fistula nails is crucial, which directly affects the success rate of the operation, the patient's recovery and the incidence of postoperative complications.

[0003] Currently, the treatment of anal fistula mainly relies on surgical resection and nail implantation. However, the anatomical structure of anal fistula is complex, the individual differences among patients are large, and the incomplete or inaccurate preoperative imaging information often leads to suboptimal positioning of the nail implantation during surgery, thereby increasing the risk of surgery and the probability of subsequent complications. Traditional path planning methods mostly rely on experience and two-dimensional images, and cannot fully consider the three-dimensional structure of anal fistula and its relationship with surrounding tissues, which limits the accuracy and personalization of path planning.

[0004] Traditional path planning methods have the following technical problems: failure to fully utilize three-dimensional spatial information, resulting in the implantation path of anal fistula nails may not accurately reflect the complex anatomical structure, thus affecting the accuracy and safety of the operation; failure to effectively capture the dynamic changes of the anal fistula area during the operation, resulting in inflexible path planning and inability to respond promptly to anatomical changes or instrument positioning deviations that may occur during the operation; lack of real-time feedback and correction mechanism, and inability to dynamically adjust and correct the implantation path during the operation, so errors may occur in actual operations, affecting the implantation effect and patient recovery. Summary of the invention

[0005] The present invention provides an intelligently identified anal fistula nail implantation trajectory determination method to solve the problems that traditional path planning methods fail to fully utilize three-dimensional spatial information, resulting in the anal fistula nail implantation path failing to accurately reflect the complex anatomical structure, thereby affecting the accuracy and safety of the operation; failing to effectively capture the dynamic changes of the anal fistula area during the operation, resulting in inflexible path planning and failure to promptly respond to anatomical changes or instrument positioning deviations that may occur during the operation; lacking a real-time feedback and correction mechanism, and failing to dynamically adjust and correct the implantation path during the operation, so errors may occur in actual operations, affecting the implantation effect and patient recovery.

[0006] The present invention provides an intelligently identified anal fistula nail implantation trajectory determination method, which specifically includes the following technical solutions:

[0007] A method for determining anal fistula nail implantation trajectory by intelligent identification, comprising the following steps:

[0008] S1. Obtain image data of the anal fistula area and perform multi-scale pyramid decomposition processing to extract detail features and structural features at different levels; based on the detail features and structural features at different levels, use a dynamic time convolution module to capture the spatiotemporal feature changes of the image data of the anal fistula area and generate multi-level dynamic sequence features; through adaptive multi-modal transformation, fuse the multi-level dynamic sequence features to generate a shared space representation;

[0009] S2. Perform spatial feature dimensionality reduction processing on the shared space representation, and then perform time filtering and smoothing processing to obtain smoothed space-time features; extract the target trajectory from the smoothed space-time features through a dynamic programming algorithm; based on the target trajectory, perform path optimization to obtain the optimal implantation path of the anal fistula nail, and perform path correction based on the optimal implantation path of the anal fistula nail.

[0010] Preferably, the S1 specifically includes:

[0011] The detail features and structural features of each layer are input into the dynamic time convolution module, and multi-level dynamic sequence features are generated through convolution operations with different time steps. The specific formula is:

[0012] ,

[0013] in, Indicates the generated Layer dynamic sequence characteristics; is the convolution kernel range in the time dimension; and Indicated in Time step The convolution kernel weights of the layers are used to adjust the attention of dynamic sequence features to detail features and structural features respectively; and Respectively in time Time Detail Layer and structural layer Located at pixel position Characteristics of the place; is the bias term; Represents spatial coordinates; Represents the time dimension.

[0014] Preferably, the S1 specifically includes:

[0015] Adaptive multimodal transformation technology is used to fuse multi-level dynamic sequence features, map dynamic sequence features at different levels to a unified shared space representation, and generate a shared space representation.

[0016] Preferably, the S2 specifically includes:

[0017] Principal component analysis is used on the shared spatial representation, and the generated principal component matrix is ​​spatially projected to obtain the denoised and dimensionally reduced spatial feature representation.

[0018] Preferably, the S2 specifically includes:

[0019] A smoothing filter is used to perform temporal filtering and smoothing processing on the denoised and dimensionally reduced spatial feature representation to obtain a smoothed spatial-temporal feature.

[0020] Preferably, the S2 specifically includes:

[0021] Based on the smoothed space-time features, a dynamic programming algorithm is introduced to perform path extraction operations to obtain the target trajectory. The formula for path extraction is:

[0022] ,

[0023] in, is the target trajectory; represents a shared spatial representation; Indicates at time Shared space representation when Indicates at time The smoothed spatial-temporal characteristics of Indicates at time The shared space representation when Used to control the fit between the target trajectory and the smoothed space-time characteristics; the second Used to control the smoothness of the target trajectory; It is the time cycle; is the adjustment parameter of the smoothness of the target trajectory.

[0024] Preferably, the S2 specifically includes:

[0025] Based on the target trajectory, the objective function is defined, and the objective function is optimized and solved by the gradient descent method to obtain the optimal implantation path of the anal fistula nail; the objective function is specifically defined as:

[0026] ,

[0027] in, It is a path expression in the form of Bezier curve, which indicates the path for anal fistula nail implantation; is the objective function value; Used to control the smoothness of the path speed of anal fistula nail implantation; is the adjustment parameter; Used to ensure the continuity of acceleration along the anal fistula nail implantation path; is the weight coefficient; It is the fitting term of the anal fistula nail implantation path and the target trajectory; is the weight coefficient; Parameters for controlling deviation suppression; Is the anal fistula nail implantation path at the time The position at the time; is the target trajectory in time The position at that time.

[0028] Preferably, the S2 specifically includes:

[0029] A real-time feedback mechanism is introduced to monitor the position, angle and depth of the anal fistula nail in real time and generate feedback data. The optimal implantation path of the anal fistula nail is used as the reference path for correction, and the implantation path of the anal fistula nail is corrected in real time by comparing the deviation between the current position of the anal fistula nail in the feedback data and the optimal implantation path of the anal fistula nail.

[0030] The beneficial effects of the technical solution of the present invention are:

[0031] 1. The multi-scale pyramid decomposition technology can extract different levels of features in the image data of the anal fistula area, from the detail layer to the structure layer, ensuring that the anatomical features of the anal fistula area are fully preserved at different levels, so that path planning can be carried out at a more detailed level, thereby improving the accuracy and reliability of path planning; through the dynamic time convolution module, the present invention can effectively capture the spatial-temporal feature changes of the anal fistula area, identify the dynamic changes of the anal fistula area at different time steps, thereby more comprehensively reflecting the real-time anatomical state of the anal fistula area, and improving the real-time adaptability of path planning.

[0032] 2. The present invention adopts adaptive multimodal transformation technology to fuse dynamic sequence features from different levels to generate a high-dimensional shared space representation, which not only integrates the spatial and temporal characteristics of the anal fistula area, but also enhances the effectiveness of each layer of features through nonlinear transformation, making path planning more reliable and accurate; the dynamic programming algorithm is used to extract the optimal path from the smoothed space-time characteristics, and the smoothness and fitting of the path are optimized in combination with the objective function to ensure that the implantation path of the anal fistula nail not only meets the spatial structure, but also has the characteristics of smoothness and shortest distance, effectively reducing unnecessary deviations in the implantation process of the anal fistula nail and ensuring the accuracy and efficiency of the operation.

[0033] 3. The present invention combines principal component analysis to reduce the dimension of spatial features, uses a smoothing filter to smooth the time series, and applies a dynamic programming algorithm to extract the path. It can accurately extract the optimal implantation path of the anal fistula nail in the complex anatomical environment of the anal fistula area. At the same time, a real-time feedback mechanism is introduced to monitor the position, angle and depth of the implanter in real time through sensors, calculate and correct the angle deviation and depth deviation, and dynamically adjust the path to ensure that the anal fistula nail is implanted along the optimal implantation path of the anal fistula nail during the entire operation, thereby significantly improving the real-time adaptability and safety of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a method for determining anal fistula nail implantation trajectory through intelligent identification according to the present invention. DETAILED DESCRIPTION

[0035] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0037] The specific scheme of the method for determining the trajectory of anal fistula nail implantation by intelligent identification provided by the present invention is described in detail below with reference to the accompanying drawings.

[0038] Refer to the attached Figure 1 , which shows a flow chart of a method for determining anal fistula nail implantation trajectory by intelligent identification provided by an embodiment of the present invention, the method comprising the following steps:

[0039] S1. Obtain image data of the anal fistula area and perform multi-scale pyramid decomposition processing to extract detail features and structural features at different levels; based on the detail features and structural features at different levels, use a dynamic time convolution module to capture the spatiotemporal feature changes of the image data of the anal fistula area and generate multi-level dynamic sequence features; through adaptive multi-modal transformation, fuse the multi-level dynamic sequence features to generate a shared space representation;

[0040] Through medical imaging equipment, medical imaging technologies such as ultrasound, CT or MRI are selected to obtain image data of the patient's anal fistula area. The image data of the anal fistula area provides accurate anatomical structure information of the anal fistula area, including the specific location of the anal fistula opening, the direction of the fistula tract and the distribution of the surrounding tissue structures. The image data of the anal fistula area is saved in a three-dimensional form to ensure that sufficient spatial information and detail level support can be provided in subsequent image processing and path planning. The image data of the anal fistula area contains multi-dimensional information, has rich spatial resolution, and records the temporal dynamic changes of the anal fistula area in the form of continuous image frames.

[0041] The image data of the anal fistula area is processed by multi-scale pyramid decomposition to extract details and structural features at different levels, and the image data of the anal fistula area is decomposed into several levels of detail layers and structural layers to distinguish local and global features. The expression of multi-scale pyramid decomposition is:

[0042] ,

[0043] in, It is the image data of anal fistula area; and Respectively represent The detail layer and structure layer of the layer to ensure that the characteristics of each level are preserved; It is the total number of layers where the image data of the anal fistula area is decomposed. By utilizing the multi-scale characteristics of pyramid decomposition, the image data of the anal fistula area is decomposed in terms of local and global information to form a complete description of the anatomical structure characteristics of the anal fistula area at different scales.

[0044] After completing the multi-scale pyramid decomposition process, the detail features and structural features of each layer are input into the dynamic time convolution module, and the multi-level dynamic sequence features are generated through the convolution operation of the multi-dimensional convolution kernel to capture the spatial-temporal feature changes of the image data of the anal fistula area; the purpose of dynamic time convolution is to continuously capture the changes of the image data of the anal fistula area in the time dimension through convolution operations of different time steps, while retaining the spatial distribution characteristics of each level. The specific formula is:

[0045] ,

[0046] in, Indicates the generated The layer dynamic sequence feature represents the The spatiotemporal characteristics of the layer are used to describe the dynamic changes of the image data in the anal fistula area; is the convolution kernel range in the time dimension, which indicates the window size for convolution in time, that is, the upper limit of the number of steps sampled in the time series; and Indicated in Time step The convolution kernel weights of the layers are used to adjust the attention of dynamic sequence features to detail features and structural features respectively; and Respectively in time Time Detail Layer and structural layer Located at pixel position Characteristics of the place; is a bias term used to balance the output value; Represents spatial coordinates; Represents the time dimension. Through multi-time step convolution superposition, the spatial and temporal information of the image data of the anal fistula area is integrated to form a set of dynamic sequence features, providing rich multi-dimensional feature support for subsequent adaptive multimodal transformation and path planning.

[0047] Furthermore, through adaptive multimodal transformation, multi-level dynamic sequence features are fused to map dynamic sequence features of different scales to a unified shared space representation, so as to achieve organic integration and feature enhancement of dynamic sequence features at each layer, thereby constructing a path planning feature set containing all relevant information; the specific formula is:

[0048] ,

[0049] in, for shared space representation; Represents an adaptive multimodal transformation function, which is used to map dynamic sequence features at different levels to a shared spatial representation middle; It is The weight coefficient of the layer is used to control the Layer dynamic sequence features Shared space Contribution of It is the adjustment parameter in nonlinear transformation, which is used to adjust the input Layer dynamic sequence features The amplitude of ,to adapt to the feature scale of different layers; is the nonlinear suppression parameter used to prevent the input Layer dynamic sequence features The amplitude is too large to ensure the balance in the fusion process. Through the adaptive multimodal transformation function, the dynamic sequence features at different levels are adaptively fused to form a unified shared space representation, providing reliable high-quality data support for path planning.

[0050] S2. Perform spatial feature dimensionality reduction processing on the shared space representation, and then perform time filtering and smoothing processing to obtain smoothed space-time features; extract the target trajectory from the smoothed space-time features through a dynamic programming algorithm; based on the target trajectory, perform path optimization to obtain the optimal implantation path of the anal fistula nail, and perform path correction based on the optimal implantation path of the anal fistula nail.

[0051] After obtaining the shared space representation, the target trajectory required for path planning is generated through multi-step processing to ensure that the target trajectory can fully reflect the spatial-temporal structural information of the image data in the anal fistula area to meet the precise needs of path planning. The shared space representation contains spatial and temporal features fused from multi-level dynamic sequence features. In order to obtain the optimal path for anal fistula nail implantation, it is necessary to extract key features with physical location and time information from the shared space representation and process the key features into a target trajectory suitable for path planning.

[0052] The shared space representation is a high-dimensional feature space that contains the spatial information of the anal fistula area. First, principal component analysis is performed on the shared space representation to obtain the spatial feature representation after denoising and dimensionality reduction. The main components reflecting the anatomical structure characteristics of the anal fistula area in the shared space representation are retained to reduce noise and redundant information. Assume that the generated principal component matrix is , then the formula for spatial projection is:

[0053] ,

[0054] in, It is the spatial feature representation after denoising and dimensionality reduction.

[0055] Furthermore, the temporal features in the spatial feature representation after denoising and dimensionality reduction are smoothed to ensure the continuity of temporal information. By using a smoothing filter (such as a Gaussian filter) to perform temporal filtering and smoothing, the irregular fluctuations in the temporal features in the spatial feature representation after denoising and dimensionality reduction are eliminated, so that the generated trajectory is smooth in the temporal dimension, and a feature sequence after temporal smoothing is obtained. The specific formula is:

[0056] ,

[0057] in, Indicates at time is the feature sequence after time smoothing, that is, the smoothed space-time feature; Indicates at time Spatial feature representation after denoising and dimensionality reduction; is the time variable in the integral, representing each convolved time point; is the standard deviation of the smoothing filter. The temporal filtering operation ensures the continuity of the target trajectory in the temporal dimension and avoids the instability of the target trajectory due to excessive noise.

[0058] After obtaining the smoothed space-time features, a path extraction operation is further performed. By applying the dynamic programming algorithm, a target trajectory (i.e., an ideal path) that best meets the anatomical structure characteristics is selected from the smoothed space-time features; the ideal path has smoothness and the shortest distance while meeting the anatomical structure characteristics of the anal fistula area. The formula for path extraction is:

[0059] ,

[0060] in, is the target trajectory; Indicates at time Shared space representation when Indicates at time The shared space representation when Used to control the fit between the target trajectory and the smoothed space-time characteristics; the second Used to control the smoothness of the target trajectory; It is the time cycle; is the adjustment parameter of the smoothness of the target trajectory. The path extraction process can be solved by dynamic programming, and finally the target trajectory that conforms to the anatomical structure characteristics is obtained. .

[0061] Generated target trajectory The core spatial-temporal characteristics of the image data of the anal fistula area are retained, which can truly reflect the anatomical characteristics of the anal fistula area. Then, the path optimization step is entered. By setting the objective function , to control the smoothness of the anal fistula nail implantation path and the degree of fit with the target trajectory, optimize a smooth anal fistula nail implantation path that meets anatomical requirements, and generate the optimal anal fistula nail implantation path. The objective function formula is as follows:

[0062] ,

[0063] in, It represents the path expression in the form of Bezier curve, i.e. the anal fistula nail implantation path; is the objective function value, used to optimize the anal fistula nail implantation path The smoothness and fit with the target trajectory are used to determine the optimal implantation path for anal fistula nails; Is the anal fistula nail implantation path at the time The position at the time; is the target trajectory in time The position at that time; each item in the objective function formula is the constraint condition of the anal fistula nail implantation path: the first item Used to control the smoothness of the anal fistula nail implantation path speed. is the adjustment parameter; the second Used to ensure the continuity of acceleration along the anal fistula nail implantation path. is the weight coefficient; the third term is the fitting term of the anal fistula nail implantation path and the target trajectory, is the weight coefficient, which is used to control the deviation between the anal fistula nail implantation path and the target trajectory. is the parameter for controlling deviation suppression. By optimizing the objective function, the optimal implantation path of anal fistula nail is obtained. , ensuring that the anal fistula nail implantation path is smooth and precise in terms of anatomical characteristics.

[0064] Objective Function The optimization solution adopts the existing technology gradient descent method, and gradually approaches the optimal solution through multiple rounds of iterations. In each iteration process, the gradient of the anal fistula nail implantation path relative to the objective function is calculated, and the position of the anal fistula nail implantation path point is updated, so that the objective function value gradually tends to the minimum. When the change of the objective function value in the continuous iteration process is less than the preset minimum change threshold, or the number of iterations exceeds the preset maximum number of iterations, the iteration is stopped, and finally the optimal implantation path of the anal fistula nail is obtained. , that is, the implantation trajectory scheme under static conditions. However, due to the dynamic nature of the surgical process, in order to ensure the real-time and accuracy of the optimal implantation path of the anal fistula nail, a real-time feedback mechanism needs to be introduced to correct the implantation path of the anal fistula nail.

[0065] The optimal implantation path of the anal fistula nail is used as the reference path for correction. During the operation, the position, angle and depth of the anal fistula nail are monitored in real time through the sensor on the implanter to generate feedback data. In each feedback cycle, the deviation between the current position of the anal fistula nail and the optimal implantation path of the anal fistula nail is compared, and the angle deviation and depth deviation are calculated to correct the deviation in the implantation path of the anal fistula nail in real time. The correction formula is as follows:

[0066] ,

[0067] in, It's in time The implantation path of anal fistula nail was adjusted. is the angle deviation, which represents the deviation of the angle in the feedback data from the corrected reference path; is the depth deviation, which indicates the deviation of the depth in the feedback data from the corrected reference path; the periodic function and Used to ensure the smoothness of the correction process. The real-time correction operation gradually adjusts the implantation position and angle of the anal fistula nail according to the feedback data, ensuring that the anal fistula nail always moves along the optimal implantation path of the anal fistula nail, meeting the needs of real-time adjustment of the surgery.

[0068] In summary, an intelligent identification method for determining the anal fistula nail implantation trajectory was completed.

[0069] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0071] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for determining the implantation trajectory of anal fistula nails by intelligent identification, characterized in that: The following steps are involved: S1. Obtain image data of anal fistula area and perform multi-scale pyramid decomposition processing to extract detail features and structural features at different levels; Based on the detail features and structural features at different levels, the dynamic time convolution module is used to capture the spatiotemporal feature changes of the image data of the anal fistula area and generate multi-level dynamic sequence features; The specific formula is: ， in, Indicates the generated Layer dynamic sequence characteristics; is the convolution kernel range in the time dimension; and Indicated in Time step The convolution kernel weights of the layers are used to adjust the attention of dynamic sequence features to detail features and structural features respectively; and Respectively in time Time Detail Layer and structural layer Located at pixel position Characteristics of the place; is the bias term; Represents spatial coordinates; Represents the time dimension; Through adaptive multimodal transformation, multi-level dynamic sequence features are fused to generate shared space representation; S2, perform spatial feature dimensionality reduction processing on the shared space representation, and then perform temporal filtering and smoothing processing to obtain smoothed space-time features; The target trajectory is extracted from the smoothed space-time features through a dynamic programming algorithm; based on the target trajectory, the path is optimized to obtain the optimal implantation path of the anal fistula nail, and the path is corrected based on the optimal implantation path of the anal fistula nail.

2. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 1, characterized in that: The S1 specifically includes: Adaptive multimodal transformation technology is used to fuse multi-level dynamic sequence features, map dynamic sequence features at different levels to a unified shared space representation, and generate a shared space representation.

3. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 1, characterized in that: The S2 specifically includes: Principal component analysis is used on the shared spatial representation, and the generated principal component matrix is ​​spatially projected to obtain the denoised and dimensionally reduced spatial feature representation.

4. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 3, characterized in that: The S2 specifically includes: A smoothing filter is used to perform temporal filtering and smoothing processing on the denoised and dimensionally reduced spatial feature representation to obtain a smoothed spatial-temporal feature.

5. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 4, characterized in that: The S2 specifically includes: Based on the smoothed space-time features, a dynamic programming algorithm is introduced to perform path extraction operations to obtain the target trajectory. The formula for path extraction is: ， in, is the target trajectory; represents a shared spatial representation; Indicates at time Shared space representation when Indicates at time The smoothed spatial-temporal characteristics of Indicates at time The shared space representation when Used to control the fit between the target trajectory and the smoothed space-time characteristics; the second Used to control the smoothness of the target trajectory; It is the time cycle; is the adjustment parameter of the smoothness of the target trajectory.

6. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 5, characterized in that: The S2 specifically includes: Based on the target trajectory, the objective function is defined, and the objective function is optimized and solved by the gradient descent method to obtain the optimal implantation path of the anal fistula nail; the objective function is specifically defined as: ， in, It is a path expression in the form of Bezier curve, which indicates the path for anal fistula nail implantation; is the objective function value; Used to control the smoothness of the anal fistula nail implantation path speed; is the adjustment parameter; Used to ensure the continuity of acceleration along the anal fistula nail implantation path; is the weight coefficient; It is the fitting term of the anal fistula nail implantation path and the target trajectory; is the weight coefficient; Parameters for controlling deviation suppression; Is the anal fistula nail implantation path at the time The position at the time; is the target trajectory in time The position at that time.

7. The method for determining the implantation trajectory of anal fistula nails by intelligent identification according to claim 6, characterized in that: The S2 specifically includes: A real-time feedback mechanism is introduced to monitor the position, angle and depth of the anal fistula nail in real time and generate feedback data. The optimal implantation path of the anal fistula nail is used as the reference path for correction, and the implantation path of the anal fistula nail is corrected in real time by comparing the deviation between the current position of the anal fistula nail in the feedback data and the optimal implantation path of the anal fistula nail.

Images