Urologic surgical navigation data visualization system and method

Abstract

The invention discloses a urinary surgery navigation data visualization system and method, and relates to the technical field of data visualization, MRI and CT image data of different organs are subjected to different preprocessing, rendering is carried out by distinguishing the importance degree in a rendering area, and navigation data required by the surgery is visualized during the urinary surgery; different preprocessing methods are selected to make modeling finer, and during rendering, regional importance degree distinguishing is performed, so that computing resources and computing time can be saved, the important structure of a diseased part of a patient can be displayed more clearly, and a wider visual field and more accurate assistance are provided for doctors.

Description

Technical Field

[0001] The present invention relates to the technical field of data visualization, and more particularly to a urology surgery navigation data visualization system. Background Art

[0002] With the continuous advancement of modern medical technology, the application of surgical navigation systems in complex surgeries has gradually become an important part of the medical field. The surgical navigation data visualization system accurately connects medical imaging, real-time physiological data and surgical operations, providing surgeons with intuitive and real-time visualization support to improve the accuracy, safety and efficiency of surgery.

[0003] The prior art has the following deficiencies:

[0004] The existing system models the patient's organs as a whole, without distinguishing the importance of each area of ​​the patient's surgical site. This not only disrupts the doctor's vision, but also wastes computing power and increases computing time. Summary of the invention

[0005] In order to overcome the above defects of the existing system, the embodiment of the present invention provides a urology surgery navigation data visualization system and method, which performs different preprocessing on the MRI and CT image data of different organs and renders them according to the importance of the rendering area, so as to solve the problems raised in the above background.

[0006] To achieve the above objectives, the present invention provides the following technical solutions.

[0007] A method for visualizing navigation data of urological surgery comprises the following steps:

[0008] Step S1: Perform MRI (magnetic resonance imaging) and CT (computed tomography) on the patient's diseased part to collect image data. During the collection, the unclear images are excluded and re-collected. The patient's body position data is collected, which needs to be matched with the three-dimensional modeling later. The spatial position data of the surgical instruments is collected.

[0009] Step S2: The obtained MRI and CT image data are processed separately according to different organs, and denoised, registered, and standardized. This facilitates the subsequent generation of three-dimensional models for each organ. The data clarity evaluation parameter K is calculated by analyzing the spectrum and local contrast.

[0010] Step S3: 3D modeling and reconstruction. According to the importance of organs and regions, the growth algorithm is used to generate the scope and key areas required for surgery. The standard data obtained in S2 is used for surface reconstruction and volume reconstruction to obtain the 3D models of each organ.

[0011] Step S4: Data fusion and matching: matching the obtained patient body position data with the target three-dimensional model point cloud, and fusing it with the surgical instrument position to display the surgical instrument position in the three-dimensional model.

[0012] Step S5: Real-time navigation and assistance: During the operation, real-time assistance is provided on the display to help the doctor expand the visual range, optimize the surgical path, and reduce surgical risks.

[0013] In a preferred embodiment, in step S1, firstly, DICOM format tomographic images are acquired from MRI and CT machines; after successful acquisition, the system determines whether the image is clear by analyzing the spectrum and local contrast of MRI and CT image data, and if it is not clear, it needs to be re-acquired; the patient's body position data is acquired, and a depth camera and structured light are used to acquire the patient's body position data. The spatial position data of surgical instruments are acquired, and electromagnetic tracking is used to install a small sensor on the surgical instrument to receive the signal emitted by the transmitter, and the position and posture of the surgical instrument in three-dimensional space are calculated; the two sets of data, the patient's body position data and the surgical instrument spatial position data, are retained for backup; the acquired MRI and CT image data are stored in the file system in DICOM format.

[0014] In a preferred embodiment, in step S2, the high signal processing result and the low signal processing result of the image are analyzed by using the spectrum analysis and local contrast judgment in step S1 to obtain the high frequency energy domain value E and the local contrast map domain value L of the center frequency division of each layer of the image; the clarity parameter K is defined as K=E i +L i Among them, E i ,L i are the average values ​​of high frequency energy and contrast of all MRI and CT images, respectively;

[0015] In a preferred embodiment, in step S3, the system first needs to define the surgical center as a seed volume element, and then use a growth algorithm to grow and spread from the seed volume element to the periphery until the conditions are no longer met and the growth stops. Complete the rendering area selection. Specifically as follows: First, the doctor assigns the surgical center to the system, and the system generates a seed volume element at the surgical center, and generates a spatial rectangular coordinate system with each vertex of the seed volume element. There is a new volume element in each quadrant of the coordinate system, and an octree data structure is used to store all volume elements. Whether a new volume element becomes a rendering area is determined by two points. The first point is the spatial importance. The spatial importance is centered on the seed volume element and decreases in importance toward the surrounding area. The second point is the importance of organs. The organs here mainly refer to nerves and blood vessels; the above two points combined with the clarity evaluation parameter K in step S2 can determine the importance of a volume element. Specify a domain value to determine whether a volume element is used as a rendering area, and confirm the rendering area;

[0016] In a preferred embodiment, in step S3, after obtaining the rendering area, the MRI and CT image data obtained in step S2 are used to perform surface reconstruction and volume reconstruction in the rendering area to generate a target three-dimensional model. Surface reconstruction is specifically as follows: using the MRI and CT images of different organs such as blood vessels and nerves in S2, mesh reconstruction is used after confirming the rendering area, and each volume element is classified to determine its outer and inner boundaries, and then the boundaries of the volume element are converted into triangular meshes according to predefined triangular surface generation rules. The generated triangular mesh represents the reconstructed surface. Volume reconstruction is as follows: using the three-dimensional surface generated by surface reconstruction, a volume element mesh is generated inside the three-dimensional surface, and using the ray projection method, rays are emitted from the center of each volume element to the surface model to check the intersection of the ray and the surface. If the ray intersects the surface, the volume element is inside the surface and volume filling is performed; after the filling is completed, the target model is successfully generated;

[0017] In a preferred embodiment, in step S4, the patient body position data collected in step S1 is used to perform point cloud matching with the target model, specifically: grid sampling is used on the target model to obtain the point cloud data to be matched, the target model point cloud data and the patient body position structured light point cloud data are matched so that the two have matching spatial positions, and the system then loads the surgical instrument spatial position data into the entire model data. The matching of model data and spatial position is completed;

[0018] Technical effects and advantages of a urology surgery navigation data visualization system and method of the present invention:

[0019] The present invention performs different preprocessing on MRI and CT image data of different organs and renders by distinguishing the importance of the rendering area. Different preprocessing methods are used to make the modeling more refined. When rendering, area selection can solve the problem of computing resources, save computing time, and more clearly display the important structures of the patient's diseased parts, providing doctors with a wider field of view and more accurate assistance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This invention is a schematic diagram of a method for visualizing navigation data in urological surgery.

[0021] Figure 2 This is the present invention: a schematic diagram of a urological surgery navigation data visualization system. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of 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.

[0023] The present invention performs different preprocessing on MRI and CT image data of different organs and renders by distinguishing the importance of the rendering area.

[0024] Embodiment 1, a method for visualizing navigation data of urological surgery, such as Figure 1 As shown, the following steps are included:

[0025] Step S1: Data collection and screening. Perform MRI (magnetic resonance imaging) and CT (computed tomography) on the patient's diseased part to collect image data. During the collection, the unclear images are excluded and re-collected. The patient's body position data is collected, which needs to be matched with the three-dimensional modeling later. The spatial position data of the surgical instruments is collected.

[0026] Step S2: Data preprocessing: the image data is processed according to different organs, and denoised, registered, and standardized. This facilitates the subsequent generation of three-dimensional models for each organ. The data clarity evaluation parameter K is calculated by analyzing the spectrum and local contrast.

[0027] Step S3: 3D modeling and reconstruction. According to the importance of organs and regions, the growth algorithm is used to generate the scope and key areas required for surgery. The standard data obtained in S2 is used for surface reconstruction and volume reconstruction to obtain the 3D models of each organ.

[0028] Step S4: Data fusion and matching: matching the obtained patient body position data with the target three-dimensional model point cloud, and fusing it with the surgical instrument position to display the surgical instrument position in the three-dimensional model.

[0029] Step S5: Real-time navigation and assistance. During the operation, real-time assistance is provided on the display to help the doctor expand the visual range, optimize the surgical path, and reduce surgical risks.

[0030] The specific implementation is as follows:

[0031] In step S1, first, a DICOM format tomographic image is obtained from an MRI machine and a CT machine. According to the doctor's diagnosis plan, the diseased part that needs to be collected by the patient is confirmed, and then the image data of the diseased part is collected using MRI and CT. After the collection is completed, the image data is input into the system. The system determines whether the image is clear by analyzing the spectrum and local contrast of the image data. If it is not clear, it needs to be re-collected. The specific method of spectrum analysis is: using fast Fourier transform (FFT) to calculate the image spectrum, calculate the energy of the high-frequency part of the spectrum, set a frequency domain value to distinguish high frequency from low frequency, and the higher the energy of the high frequency, the clearer the picture. Set a high-frequency energy domain value to control the clarity. The specific method of local contrast judgment is: select a fixed-size window in the image and slide the window in the image. In each window, calculate the local contrast. The calculation method is to calculate the standard deviation of the pixel values ​​of all pixels in the window, and attach the local contrast value of each window to the central pixel to generate a local contrast map. Analyze the local contrast map, calculate the mean of the local contrast map, the higher the mean, the clearer the image, and set a local contrast domain value to control the clarity. If both domain values ​​pass, there is no need to re-collect, otherwise, if they fail, re-collect. Collect the patient's body position data, use a depth camera, and use structured light to collect the patient's body position data. Collect the surgical instrument's spatial position data, use electromagnetic tracking, install a small sensor on the surgical instrument to receive the signal from the transmitter, and calculate the position and posture of the surgical instrument in three-dimensional space. Keep the two sets of data, the patient's body position data and the surgical instrument's spatial position data, for backup. Store the collected image data in the file system in DICOM format for use in subsequent steps.

[0032] In step S2, the image slices in DICOM format obtained in step S1 are processed by organ as follows:

[0033] Blood vessels: Blood vessels, especially those with high signal intensity when contrast agents are used, are denoised using non-local means (NLM). For each pixel p in the image, its similarity to the surrounding pixels is calculated using a local window such as 7×7 or 9×9, i.e., a region Ω centered on p. p, use texture analysis methods (such as Gabor filtering, local binary pattern (LBP) etc.) to extract texture features in the local window. The feature value is I(x, y), where x and y are pixel position parameters. After obtaining the feature values ​​of different windows, calculate the similarity metric, such as Euclidean distance. For window W x and W y , the calculation formula is: Where i, j are the offsets within the window.

[0034] According to the result of similarity measurement, a weight is assigned to each pixel. The more similar the pixel is, the greater its weight is. The weight is calculated using Gaussian function: where h is a parameter that controls the influence of the similarity measure.

[0035] Finally, the calculated weights are used to perform a weighted average on the values ​​of all similar pixels to update the value of the current pixel: Where Z(x) is a normalization factor to ensure that the weights sum to 1.

[0036] After denoising, the vascular MRI and CT images are stored for subsequent use. The h value is obtained through experimental and empirical values.

[0037] Nerves: Nerves present low signals in images (compared to surrounding tissue fat), and contrast enhancement is needed to improve nerve identifiability. Image contrast is enhanced by training a convolutional neural network (CNN) model, training data, and training using a set of low-contrast and high-contrast image pairs. The mean squared error (MSE) or perceptual loss function is used to evaluate the difference between the generated image and the target image. After training, the model is used on low-contrast images to generate high-contrast images.

[0038] After generating high-contrast images, slices of all neural layers are stored for later use.

[0039] Fat: Fat is a high signal area and the same result as blood vessels is used, so it is not processed repeatedly.

[0040] The remaining tissues are low signal areas and have the same results as the nerves and are not processed repeatedly.

[0041] By using the spectrum analysis and local contrast judgment in step S1, the high signal processing results and the low signal processing results of the image are analyzed to obtain the high frequency energy domain value E and the local contrast map domain value L of the center frequency division of each layer of the image.

[0042] Define the clarity parameter K as K = E i +L i Among them, E i ,L iare the average values ​​of high frequency energy and contrast of all tomographic images, respectively.

[0043] In step S3, according to the importance of the region and the organ, the system first needs to define the center of the surgery as a seed volume element, and then use a growth algorithm to grow and spread from the seed volume element to the surrounding area until the conditions are no longer met and the growth stops. Complete the selection of the rendering area. Specifically as follows: First, the doctor assigns the center of the surgery to the system, and the system generates a seed volume element such as a 1×1×1 (unit: cm) volume element in the center of the surgery as the seed volume element. After the seed volume element is generated, a spatial rectangular coordinate system is generated with each vertex of the seed volume element. There is a new volume element in each quadrant of the coordinate system, and an octree data structure is used to store all volume elements. Whether a new volume element becomes a rendering area is determined by two points. The first point is the spatial importance. The spatial importance is centered on the seed volume element and decreases in importance toward the surrounding area. We specify the decrease as Gaussian decrease, and the formula is: Among them, σ is the dispersion, μ is the center position, the center position is the seed volume element position, and the σ dispersion can be adjusted according to the results based on the different needs of doctors. x is the distance from the target volume element to the seed volume element. A judgment basis parameter can be defined by f(x): J = f(x) J's data type is percentage. The second point is the importance of the organ. The organs here mainly refer to nerves and blood vessels. When the system determines a new volume element, two parameters are defined as follows: blood_den, volume element blood vessel density. nerve_den, volume element nerve density.

[0044] When the system determines a new volume element, it models and calculates the parameters blood_den and nerve_den within the volume element from the blood vessel and nerve image data obtained in step S2. The data types of both parameters are percentages.

[0045] After obtaining the two parameters, combined with the clarity parameter K in S2, the judgment basis is obtained: D = K (blood_den + nerve_den). Combining the two parameters J and D, the system can obtain a comprehensive parameter V: V = JD; V is used as a parameter to determine whether the newly added volume element is used as a rendering area. V is affected by the image data clarity and parameter K and the organ importance parameter. When the V value is greater than the domain value, the newly added volume element will be used as a rendering area. The V domain value is an empirical value based on the existing data experiment as a reference, and the doctor can also adjust the V domain value to control the system rendering area.

[0046] After obtaining the rendering area, use the image data obtained in step S2 to perform surface reconstruction and volume reconstruction in the rendering area to generate a target three-dimensional model. Surface reconstruction is specifically as follows: using the tomographic image in S2, use mesh reconstruction after confirming the rendering area, classify each volume element, determine its outer boundary and inner boundary, and then convert the boundary of the volume element into a triangular mesh according to the predefined triangular surface generation rules. The generated triangular mesh represents the reconstructed surface. Volume reconstruction is as follows: using the three-dimensional surface generated by surface reconstruction, generate a volume element mesh inside the three-dimensional surface, use the ray projection method, emit rays from the center of each volume element to the surface model to check the intersection of the ray and the surface, if the ray intersects with the surface, the volume element is inside the surface, and volume filling is performed.

[0047] After filling is completed, the target model is generated successfully.

[0048] In step S4, the patient body position data collected in step S1 is used to perform point cloud matching with the target model. Specifically, grid sampling is used on the target model to obtain the point cloud data to be matched, and the target model point cloud data and the patient body position structured light point cloud data are matched so that the two have matching spatial positions. The system then loads the surgical instrument spatial position data into the entire model data. The model data and spatial position matching is completed.

[0049] In step S5, during the operation, according to the method in step S1, the patient's body position data and the surgical instrument spatial position data are continuously input into the system. After the system fuses the data according to step S4, it can display the current surgical instrument position and surgical progress status in real time. Based on the images and surgical instrument positions provided by the system, the doctor can obtain a better field of view, observe the diseased position more intuitively, reduce surgical risks, and optimize the surgical path.

[0050] Embodiment 2, a system for visualizing navigation data of urological surgery. Figure 2 As shown, a system for realizing a urological surgery navigation data visualization system includes a data acquisition module, an algorithm processing module, a data preprocessing module, a modeling and data fusion module, and a user interaction module:

[0051] The data acquisition module is used to collect MRI and CT image data, use a depth camera to apply structured light to collect patient body position data, use an electromagnetic position sensor to collect surgical instrument position data, and transfer the collected data to the data preprocessing module;

[0052] The data preprocessing module is used to perform preprocessing such as denoising and standardization on the collected image data, and interact with the algorithm processing module to generate clarity evaluation parameters and pass the processed data to the modeling and data fusion module;

[0053] The modeling and data fusion module is used to interact with the algorithm module to build a three-dimensional model and match it with the patient's body position, then load the surgical instrument position and pass the result to the user interaction module;

[0054] The user interaction module is used to cooperate with the algorithm processing module to process the required data and user operations in real time and display them on the display, serving as a user's direct contact with the front-end module.

[0055] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0056] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0057] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0058] 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 who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0059] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for visualizing urological surgical navigation data, characterized in that: The following steps are involved: Step S1: collecting MRI and CT image data at the patient's diseased part, excluding unclear MRI and CT image data and re-collecting them when collecting MRI and CT image data; using a depth camera and structured light to collect patient body position data; and using an electromagnetic sensor to collect surgical instrument spatial position data; Step S2: The obtained MRI and CT image data of different organs including blood vessels, nerves, fat, and other tissues are processed separately, denoised, and registered to obtain standardized data; the standardized data clarity evaluation parameter K is calculated by analyzing the spectrum and local contrast of the standardized data; Step S3: using a growth algorithm to generate key areas required for surgery, using the standardized data obtained in step S2 to perform surface reconstruction and volume reconstruction to obtain a three-dimensional model of each organ; Step S4: Perform point cloud matching on the patient's body position data and the three-dimensional models of each organ, and add the spatial position of the surgical instrument to the three-dimensional model of each organ, so that the position of the surgical instrument can be displayed in the three-dimensional model of each organ.

2. The method for visualizing urological surgery navigation data according to claim 1, characterized in that: In step S1, first, DICOM format tomographic images must be obtained from the MRI and CT machines; after successful acquisition, the system determines whether the image is clear by analyzing the spectrum and local contrast of the MRI and CT image data, and if it is not clear, it needs to be re-acquired; the patient's body position data is collected by using a depth camera and structured light; the surgical instrument spatial position data is collected by using electromagnetic tracking, and a small sensor is installed on the surgical instrument to receive the signal sent by the transmitter, and the position and posture of the surgical instrument in three-dimensional space are calculated; the two sets of data, the patient's body position data and the surgical instrument spatial position data, are retained for backup; the collected MRI and CT image data are stored in the file system in DICOM format.

3. The method for visualizing urological surgery navigation data according to claim 2, characterized in that: In step S1, the specific method of spectrum analysis is: using fast Fourier transform to calculate the image spectrum, calculating the energy of the high-frequency part of the spectrum, setting a frequency domain value to distinguish high frequency from low frequency, and setting a high-frequency energy domain value to control clarity; The specific method for judging the local contrast is: select a window of a fixed size in the image and slide the window in the image. In each window, calculate the local contrast by finding the standard deviation of the pixel values ​​of all pixels in the window, attach the local contrast value of each window to the central pixel, and generate a local contrast map; analyze the local contrast map, calculate the mean of the local contrast map, and set a local contrast domain value to control the clarity; if both domain values ​​pass, there is no need to re-collect, otherwise, if it fails, re-collect.

4. The method for visualizing urological surgery navigation data according to claim 1, characterized in that: In step S2, using the DICOM format MRI and CT image data obtained in step S1, different organs including blood vessels, nerves, fat, and other tissues are processed as follows: Blood vessels, especially those that show high signal areas when contrast agents are used, are denoised using non-local means; the denoised standardized MRI and CT image data are stored under the blood vessel entry; Neural: Neural networks present low signals in images, and contrast enhancement is needed to improve neural recognizability; image contrast is enhanced by training a convolutional neural network model, training data, and a set of low-contrast and high-contrast image pairs for training; the mean square error or perceptual loss function is used to evaluate the difference between the generated image and the target image; After training, the model is used for low-contrast images to generate high-contrast images; the high-contrast standardized MRI and CT image data are stored under the neurological entry; Fat: Fat is a high signal area, and the same result as blood vessels is used, so no repeated processing is required; Other tissues: Other tissues are low signal areas and have the same results as nerves, so they are not processed repeatedly.

5. The method for visualizing urological surgery navigation data according to claim 4, characterized in that: In step S2, the high signal processing result and the low signal processing result of the image are analyzed by using the spectrum analysis and local contrast judgment in step S1, and the high frequency energy domain value E and the local contrast map domain value L of the center frequency division of each layer of the image are obtained; the clarity parameter K is defined as K=E i +L i .

6. The method for visualizing urological surgery navigation data according to claim 1, characterized in that: In step S3, the system first needs to define the surgical center as a seed volume element, and then use a growth algorithm to grow and spread from the seed volume element to the periphery until the conditions are no longer met and the growth stops; the rendering area selection is completed; specifically as follows: first, the doctor assigns the surgical center to the system, and the system generates a seed volume element at the surgical center, and generates a spatial rectangular coordinate system with each vertex of the seed volume element. There is a new volume element in each quadrant of the coordinate system, and an octree data structure is used to store all volume elements; whether a new volume element becomes a rendering area is determined by two points, the first point is the spatial importance, the spatial importance is centered on the seed volume element and decreases in importance toward the surrounding area; the second point is the organ importance, where the organs mainly refer to nerves and blood vessels; the above two points combined with the clarity evaluation parameter K in step S2 can determine the importance of a volume element; specify a domain value to determine whether a volume element is used as a rendering area, and confirm the rendering area.

7. The method for visualizing urological surgery navigation data according to claim 5, characterized in that: In step S3, after obtaining the rendering area, the MRI and CT image data obtained in step S2 are used to perform surface reconstruction and volume reconstruction in the rendering area to generate a target three-dimensional model; the surface reconstruction is specifically: using the MRI and CT images of different organs in S2, including blood vessels, nerves, fat, and other tissues, and confirming the rendering area, using mesh reconstruction, classifying each volume element, determining its outer boundary and inner boundary, and then converting the boundary of the volume element into a triangular mesh according to the predefined triangular surface generation rules; the generated triangular mesh represents the reconstructed surface; the volume reconstruction is: using the three-dimensional surface generated by surface reconstruction, generating a volume element mesh inside the three-dimensional surface, using the ray projection method, emitting rays from the center of each volume element to the surface model to check the intersection of the ray and the surface, if the ray intersects with the surface, the volume element is inside the surface, and the volume is filled; After filling is completed, the target model is generated successfully.

8. The method for visualizing urological surgery navigation data according to claim 1, characterized in that: In step S4, the patient's body position data collected in step S1 is used to perform point cloud matching with the target model. Specifically, grid sampling is used on the target model to obtain the point cloud data to be matched, and the target model point cloud data and the patient's body position structured light point cloud data are matched so that the two have matching spatial positions. The system then loads the surgical instrument spatial position data into the entire model data to complete the matching of the model data and the spatial position.

9. A urology surgery navigation data visualization system, based on a urology surgery navigation data visualization method according to any one of claims 1 to 8, characterized in that: It includes data acquisition module, algorithm processing module, data preprocessing module, modeling and data fusion module, and user interaction module; The data acquisition module is used to collect MRI and CT image data, use a depth camera to apply structured light to collect patient body position data, use an electromagnetic position sensor to collect surgical instrument position data, and transfer the collected data to the data preprocessing module; The data preprocessing module is used to denoise and standardize the collected image data, and interact with the algorithm processing module to generate clarity evaluation parameters and pass the processed data to the modeling and data fusion module; The modeling and data fusion module is used to interact with the algorithm module to build a three-dimensional model and match it with the patient's body position, then load the surgical instrument position and pass the result to the user interaction module; The user interaction module is used to cooperate with the algorithm processing module to process the required data and user operations in real time and display them on the display, serving as a user's direct contact with the front-end module.

Images