Ultrasonic imaging endoscopic surgery navigation system and method

Abstract

The invention discloses an endoscopic surgery navigation system and method for ultrasonic imaging, and the method comprises the steps: determining a to-be-detected target region in image data through obtaining the image data of the internal tissue of a human body, and determining an incision of the to-be-detected target region; entering the interior of the to-be-detected target area through the entrance, and compensating a target obstacle in the to-be-detected target area by adopting an elastic model; identifying points are marked for all the target obstacles, and the identifying points are arranged in the to-be-detected target area in a three-dimensional image mode; based on the grey wolf algorithm, the target path of the to-be-detected target area is calculated through the multiple recognition points, then endoscopic surgery is propelled through the target path, an optimal navigation path is provided in the surgery, it is guaranteed that the propelling process of an endoscopic probe is rapid and accurate, and the diagnosis result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic imaging, and particularly relates to an endoscopic surgical navigation system and method for ultrasonic imaging. Background Art

[0002] The combination of ultrasonic imaging and endoscopic surgery, commonly known as interventional therapy guided by endoscopic ultrasonography, is a minimally invasive surgical technique that uses an ultrasonic probe integrated at the front end of the endoscope to perform real-time imaging, assisting doctors in accurately locating lesions and completing treatment operations.

[0003] Currently, the existing endoscopic ultrasonic navigation has the following disadvantages: highly coordinated efforts are required for ultrasonic image interpretation and navigation operations, resulting in a long learning curve; the ability to resolve static images of organ displacement or respiratory movement is insufficient, and the optimal navigation path cannot be provided during the operation, leading to deviations during the advancement of the endoscopic probe and unable to provide accurate diagnostic results for patients. Summary of the Invention

[0004] The purpose of the present invention is to provide an endoscopic surgical navigation system and method for ultrasonic imaging to solve the problems described in the background art.

[0005] The technical solution of the present invention is implemented as follows:

[0006] On the one hand, the present invention provides an endoscopic surgical navigation method for ultrasonic imaging, including:

[0007] Obtain human internal tissue image data, determine the target area to be detected in the image data, and determine the incision for the target area to be detected;

[0008] Enter the interior of the target area to be detected through the incision, and use an elastic model to compensate for the target obstacles in the target area to be detected;

[0009] Mark recognition points for all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the target area to be detected;

[0010] Based on the grey wolf algorithm, use multiple recognition points to calculate the target path of the target area to be detected.

[0011] A further technical solution is that the step of obtaining human internal tissue image data, determining the target area to be detected in the image data, and determining the incision for the target area to be detected includes:

[0012] Scan the human internal tissue with a medical imaging device and generate the human internal tissue image data;

[0013] According to the image data, determine the lesion location and determine the lesion location as the target area to be detected;

[0014] The cut-in point is determined by randomly selecting within the target area to be detected.

[0015] A further technical solution is that the step of determining the cut-in point by randomly selecting within the target area to be detected includes:

[0016] The target area to be detected is divided into a first interval, a second interval, and a third interval. The organizational structure is calculated in the first interval, the second interval, and the third interval, and the simplest structure of the organizational structure is determined as the cut-in point.

[0017] A further technical solution is that the step of entering the interior of the target area to be detected through the cut-in point and compensating for the target obstacles in the target area to be detected using an elastic model includes:

[0018] Extract the elastic mechanical properties of the target obstacles in the target area to be detected;

[0019] Establish a layered elastic model based on the elastic mechanical properties, and define the constitutive equations for each layer of the layered elastic model;

[0020] Predict the stress distribution and displacement field of the target obstacles according to the constitutive equations;

[0021] Reverse map the displacement field to the target area to be detected to generate the spatial coordinates of the target obstacles.

[0022] A further technical solution is that the step of establishing a layered elastic model based on the elastic mechanical properties and defining the constitutive equations for each layer of the layered elastic model includes:

[0023] Measure the static elastic modulus of each layer in the layered elastic model by ultrasonic shear wave elastography, and construct the constitutive equations with the static elastic modulus. Among them, the constitutive equations are used to predict the deformation of the layered elastic model by combining the parameters of the static elastic modulus of each layer.

[0024] A further technical solution is that the layered elastic model includes a viscoelastic model, and the static elastic modulus associated with the viscoelastic model constructs the constitutive equations through the parameters of the static elastic modulus of each layer to predict the deformation of the viscoelastic model.

[0025] A further technical solution is that the step of marking identification points for all the target obstacles, where the identification points are arranged in a three-dimensional image in the target area to be detected, includes:

[0026] Obtain the three-dimensional point cloud data of the target obstacles in the area to be detected and mark the identification points;

[0027] Arrange the selected recognition points according to their actual coordinates in the three-dimensional space within the area to be detected.

[0028] A further technical solution is that, based on the Grey Wolf algorithm, the steps of calculating the target path of the area to be detected by using multiple said recognition points include:

[0029] Generate multiple paths through the curve according to the curve connected by multiple said recognition points;

[0030] Use the Grey Wolf algorithm to evaluate multiple said paths to obtain the target path of the optimal area to be detected.

[0031] A further technical solution is that the steps of using the Grey Wolf algorithm to evaluate multiple said paths to obtain the target path of the optimal area to be detected include:

[0032] Initialize the Grey Wolf population and the number of iterations;

[0033] Randomly select a path, and calculate the fitness value of the Grey Wolf based on the minimum number of target obstacles in the path as an index;

[0034] Select another path, calculate the fitness value, compare the calculated fitness value with the previously calculated fitness value, and retain the path with the largest fitness value;

[0035] Perform iterative calculation, and output the path with the largest fitness value as the optimal target path, and establish a navigation route through the optimal target path.

[0036] On the one hand, the present invention provides an endoscopic surgery navigation system for ultrasonic imaging, including:

[0037] An image acquisition module, configured to acquire human internal tissue image data, determine the area to be detected in the image data, and determine the incision of the area to be detected;

[0038] A target image confirmation module, configured to enter the interior of the area to be detected through the incision, and compensate for target obstacles in the area to be detected by using an elastic model;

[0039] A target image marking module, configured to mark recognition points for all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the area to be detected;

[0040] A target image path selection module, configured to calculate the target path of the area to be detected by using multiple said recognition points based on the Grey Wolf algorithm.

[0041] The beneficial effects of the present invention are as follows:

[0042] The present invention provides an endoscopic surgical navigation system and method for ultrasonic imaging. By acquiring image data of internal tissues of the human body, a target area to be detected in the image data is determined and an incision entrance of the target area to be detected is determined; entering the interior of the target area to be detected through the incision entrance, and compensating for target obstacles in the target area to be detected by using an elastic model; marking recognition points for all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the target area to be detected; based on the grey wolf algorithm, calculating a target path of the target area to be detected by using a plurality of recognition points, and then advancing an endoscopic surgery through the target path, providing an optimal navigation path during the surgery, ensuring that the endoscopic probe advances quickly and accurately during the advancing process, and improving the diagnosis result. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flowchart of the method of the present invention;

[0044] Figure 2 It is a specific flowchart of step S1 of the present invention;

[0045] Figure 3 It is a specific flowchart of step S2 of the present invention;

[0046] Figure 4 It is a specific flowchart of step S3 of the present invention;

[0047] Figure 5 It is a specific flowchart of step S4 of the present invention;

[0048] Figure 6 It is a specific flowchart of step 402;

[0049] Figure 7 It is a system block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] In order to better understand the technical content of the present invention, specific embodiments are provided below, and the present invention will be further described in conjunction with the accompanying drawings.

[0051] Refer to Figures 1 to 6 , the present invention provides an endoscopic surgical navigation method for ultrasonic imaging on the one hand, including the following steps:

[0052] Step S1: Acquire image data of internal tissues of the human body, determine a target area to be detected in the image data, and determine an incision entrance of the target area to be detected;

[0053] Step S2: Enter the interior of the target area to be detected through the incision entrance, and compensate for target obstacles in the target area to be detected by using an elastic model;

[0054] Step S3: Mark recognition points for all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the target area to be detected;

[0055] Step S4: Based on the grey wolf algorithm, use multiple said recognition points to calculate the target path of the target area to be detected.

[0056] It should be noted that the target obstacle refers to the human organ with displacement or respiratory movement. By determining the target obstacle, it is convenient for subsequent dynamic image judgment.

[0057] Specifically, by obtaining the human internal tissue image data, determine the target area to be detected in the image data and determine the incision of the target area to be detected; enter the interior of the target area to be detected through the incision, and use an elastic model to compensate for the target obstacles in the target area to be detected; mark recognition points for all target obstacles, where the recognition points are arranged in a three-dimensional image in the target area to be detected; based on the grey wolf algorithm, use multiple recognition points to calculate the target path of the target area to be detected, and then advance the endoscopic surgery through the target path, providing the optimal navigation path during the operation, ensuring the rapid and accurate advancement of the endoscopic probe, and improving the diagnostic result.

[0058] In this embodiment, step S1 specifically includes:

[0059] Step 101: Scan the human internal tissue with a medical imaging device and generate the human internal tissue image data;

[0060] Step 102: According to the image data, determine the lesion location and determine the lesion location as the target area to be detected;

[0061] Step 103: Determine the incision by randomly selecting in the target area to be detected.

[0062] Use CT scanning to take multi-angle photos of the human body through X-rays and obtain two-dimensional image views, and show the details of bones and soft tissues through the two-dimensional image views. Obtain the image data of the patient's internal structure through a medical imaging device, and then identify abnormal areas, such as tumors and inflammatory pathological changes. Determine the lesion location based on the location of the pathological changes in the image, and mark this area as the target area to be detected. Based on this, select the best surgical incision in the target area to be detected. Through the incision, not only the accuracy of the diagnostic entrance is improved, but also the safety of the endoscopic surgery is ensured.

[0063] Further, step 101 specifically includes: Divide the target area to be detected into a first interval, a second interval, and a third interval, calculate the tissue structure in the first interval, the second interval, and the third interval, and determine the simplest structure of the tissue structure as the incision.

[0064] The determined target area to be detected is divided into three equal parts and marked as the first interval, the second interval, and the third interval according to these three equal parts. Among these three intervals, calculations are performed according to the human tissue structures corresponding to each interval. If in the first interval, the bone tissue and blood vessel tissue are less, then the first interval is determined as the incision site. Similarly, if the bone tissue and blood vessel tissue in the second interval or the third interval are less, the second interval or the third interval is determined as the endoscopic incision site.

[0065] In this embodiment, step S2 specifically includes:

[0066] Step 201, extract the elastomechanical properties of the target obstacle in the target area to be detected;

[0067] Step 202, establish a layered elastic model based on the elastomechanical properties and define the constitutive equations for each layer of the layered elastic model;

[0068] Step 203, predict the stress distribution and displacement field of the target obstacle according to the constitutive equations;

[0069] Step 204, inversely map the displacement field to the target area to be detected to generate the spatial coordinates of the target obstacle.

[0070] It should be noted that the elastomechanical property of the target obstacle is the elastic index of the organ.

[0071] By extracting the elastomechanical properties of the target obstacle in the target area to be detected, the elastomechanical properties include Young's modulus parameters. Based on the Young's modulus parameter attributes, a layered elastic model is constructed, and corresponding constitutive equations are defined for each layer to describe the behavior of the organ under stress. Subsequently, the finite element analysis method is used to predict the stress distribution and displacement field inside the target obstacle according to the defined constitutive equations to ensure the accuracy and reliability of the calculation results. Finally, the obtained displacement field data is inversely mapped to the image data to generate the spatial coordinates of the target obstacle, so as to visually display its actual position and deformation conditions.

[0072] Further, step 202 specifically includes: measuring the static elastic modulus of each layer in the layered elastic model through ultrasonic shear wave elastography, and constructing the constitutive equation with the static elastic modulus, where the constitutive equation is used to predict the deformation amount of the layered elastic model by combining the parameters of the static elastic modulus of each layer. The layered elastic model includes a viscoelastic model, and the static elastic modulus associated with the viscoelastic model constructs the constitutive equation through the parameters of the static elastic modulus of each layer to predict the deformation amount of the viscoelastic model.

[0073] It should be noted that the layered elastic model is the thickness layer of the organ. By identifying the tissues of each layer of the organ, the accuracy of calculating the organ deformation is improved. The viscoelastic model represents an organ with softer tissues and also represents linear elasticity.

[0074] According to the image of the target area to be detected, the velocity information of the shear wave propagating between different layers is extracted. The relationship between the shear wave velocity and the static elastic modulus of the medium is adopted, where the formula E = 3ρv is used. 2 Calculate the static elastic modulus associated with the viscoelastic model. Here, E represents the elastic modulus, ρ is the organ density, and v is the shear wave velocity. Through this formula, the static elastic modulus of each layer can be quickly calculated. Based on the static elastic moduli calculated for each layer, a suitable constitutive equation is defined for each layer of material. In the case of linear elasticity, Hooke's law σ = E∈ can be used to describe the relationship between stress and strain. Furthermore, the static elastic moduli of each layer of the layered structure are accurately measured through ultrasonic shear wave elastography technology, and a constitutive equation is constructed to predict its deformation amount, achieving an accurate assessment of the organ deformation behavior, which is convenient for subsequent path obstacle avoidance selection.

[0075] In this embodiment, step S3 specifically includes:

[0076] Step 301: Obtain the three-dimensional point cloud data of the target obstacle in the area to be detected and mark the recognition points;

[0077] Step 302: Arrange the selected recognition points according to their actual coordinates in the three-dimensional space within the area to be detected.

[0078] First, the three-dimensional point cloud data of human tissues in the area to be detected is obtained, and key feature points are identified and marked as recognition points through image processing algorithms. These recognition points represent the blood vessels and tumor boundaries of human tissues. Then, the selected recognition points are arranged according to their actual coordinates in the three-dimensional space within the area to be detected, and coordinate system conversion and registration operations are performed to ensure that the spatial position relationships of the recognition points accurately reflect the true tissue structure of human tissues.

[0079] In this embodiment, step S4 specifically includes:

[0080] Step 401: Generate multiple paths through the curves connected by multiple recognition points;

[0081] Step 402: Use the gray wolf algorithm to evaluate multiple paths and obtain the optimal target path of the target area to be detected.

[0082] Connect the recognition points marked on the curved surface of the target obstacle into a curve, and generate multiple paths to be screened based on the curve. Then, use the grey wolf algorithm to evaluate the multiple paths to obtain the target path of the optimal target area to be detected, thereby realizing the rapid and stable advancement of the endoscope and improving the transmission rate for intraoperative image acquisition.

[0083] Further, step 401 specifically includes:

[0084] Step 412: Initialize the grey wolf population and the iteration times of the grey wolf population;

[0085] Step 422: Randomly select a path, and calculate the fitness value of the grey wolf based on the minimum number of the target obstacles in the path as an index;

[0086] Step 432: Select another path, calculate the fitness value, compare the calculated fitness value with the previously calculated fitness value, and retain the path with the largest fitness value;

[0087] Step 442: Perform iterative calculation, and output the path with the largest fitness value as the optimal target path, and establish a navigation route through the optimal target path.

[0088] When the grey wolf optimization algorithm screens based on the minimum number of the target obstacles in the path, it will calculate the fitness value of each path, and after each calculation of the fitness value is completed, it will compare with the previously calculated fitness value, and then retain the path with the larger fitness value. After the iterative calculation is completed, the path with the largest fitness value can be obtained, and this path is the optimal target path. The doctor can establish a navigation route through the optimal target path and realize the rapid and stable advancement of the endoscope. Since the number of nodes used in the optimal target path is the average number of nodes, the transmission rate of the endoscope for intraoperative image acquisition can be improved on the premise of ensuring the safety of the endoscope advancement.

[0089] Reference Figure 7 According to, on the one hand, the present invention provides an endoscopic surgical navigation system for ultrasonic imaging, including: an image acquisition module for acquiring human internal tissue image data, determining the target area to be detected in the image data, and determining the incision of the target area to be detected; a target image confirmation module for entering the interior of the target area to be detected through the incision and compensating for the target obstacles in the target area to be detected by using an elastic model; a target image marking module for marking recognition points on all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the target area to be detected; a target image path selection module for calculating the target path of the target area to be detected by using a plurality of the recognition points based on the grey wolf algorithm.

[0090] Specifically, by acquiring the internal tissue image data of the human body, determining the target area to be detected in the image data and determining the incision of the target area to be detected; entering the interior of the target area to be detected through the incision, and compensating for the target obstacles in the target area to be detected by using an elastic model; marking the recognition points for all the target obstacles, wherein the recognition points are arranged in a three-dimensional image in the target area to be detected; based on the grey wolf algorithm, calculating the target path of the target area to be detected by using multiple recognition points, and then advancing the endoscopic surgery through the target path, providing the optimal navigation path during the operation, ensuring the rapid and accurate advancement of the endoscopic probe, and improving the diagnostic result.

[0091] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An endoscopic surgical navigation method for ultrasonic imaging, characterized in that, Including: Obtain human internal tissue image data, determine a target area to be detected in the image data, and determine an incision for the target area to be detected; Enter the interior of the target area to be detected through the incision, and use an elastic model to compensate for target obstacles in the target area to be detected; Mark recognition points for all the target obstacles, where the recognition points are arranged in a three-dimensional image in the target area to be detected; Based on the Grey Wolf Algorithm, calculate the target path of the target area to be detected using multiple of the recognition points.

2. The endoscopic surgical navigation method for ultrasonic imaging according to claim 1, wherein, The steps of obtaining human internal tissue image data, determining a target area to be detected in the image data, and determining an incision for the target area to be detected include: Scan human internal tissues with a medical imaging device and generate the human internal tissue image data; According to the image data, determine the lesion location and determine the lesion location as the target area to be detected; Determine the incision by randomly selecting within the target area to be detected.

3. An endoscopic surgery navigation method for ultrasonic imaging according to claim 2, characterized in that, The step of determining the incision by randomly selecting within the target area to be detected includes: Divide the target area to be detected into a first interval, a second interval, and a third interval, calculate the tissue structure in the first interval, the second interval, and the third interval, and determine the simplest structure of the tissue structure as the incision.

4. The endoscopic surgical navigation method for ultrasonic imaging according to claim 1, wherein The steps of entering the interior of the target area to be detected through the incision and using an elastic model to compensate for target obstacles in the target area to be detected include: Extract the elastic mechanical properties of the target obstacles in the target area to be detected; Establish a layered elastic model based on the elastic mechanical properties, and define constitutive equations for each layer of the layered elastic model; Predict the stress distribution and displacement field of the target obstacles according to the constitutive equations; Inverse map the displacement field to the target area to be detected to generate the spatial coordinates of the target obstacles.

5. An endoscopic surgery navigation method for ultrasonic imaging according to claim 4, characterized in that, The step of establishing a layered elastic model based on the elastic mechanical properties and defining constitutive equations for each layer of the layered elastic model includes: Measure the static elastic modulus of each layer in the layered elastic model by ultrasonic shear wave elastography, and construct the constitutive equation with the static elastic modulus, where the constitutive equation is used to predict the deformation of the layered elastic model by combining the parameters of the static elastic modulus of each layer.

6. The endoscopic surgical navigation method for ultrasonic imaging according to claim 5, characterized in that, The layered elastic model includes a viscoelastic model, and the static elastic modulus associated with the viscoelastic model constructs the constitutive equation through the parameters of the static elastic modulus of each layer to predict the deformation of the viscoelastic model.

7. An endoscopic surgical navigation method for ultrasonic imaging according to claim 1, characterized in that, The step of marking recognition points for all the target obstacles, where the recognition points are arranged in a three-dimensional image in the target area to be detected includes: Obtain the three-dimensional point cloud data of the target obstacles in the area to be detected and mark the recognition points; Select the recognition points and arrange them according to their actual coordinates in the three-dimensional space within the area to be detected.

8. An endoscopic surgery navigation method for ultrasonic imaging according to claim 1, characterized in that, The steps of calculating the target path of the target area to be detected based on the Grey Wolf Algorithm using multiple of the recognition points include: According to the curves connected by multiple of the recognition points, generate multiple paths through the curves; Evaluate multiple paths using the described Grey Wolf Algorithm to obtain the optimal target path for the target area to be detected.

9. The endoscopic surgical navigation method for ultrasonic imaging according to claim 8, wherein The step of evaluating multiple paths using the described Grey Wolf Algorithm to obtain the optimal target path for the target area to be detected includes: Initialize the Grey Wolf population and the number of iterations; Randomly select a path and calculate the fitness value of the Grey Wolf based on the minimum number of target obstacles in the path as an indicator; Select another path, calculate the fitness value, compare the calculated fitness value with the previously calculated fitness value, and retain the path with the maximum fitness value; Perform iterative calculations, output the path with the maximum fitness value as the optimal target path, and establish a navigation route through the optimal target path.

10. A system for an endoscopic surgery navigation method of ultrasonic imaging according to any one of claims 1-9, characterized in that, Including: An image acquisition module for acquiring human internal tissue image data, determining the target area to be detected in the image data, and determining the incision of the target area to be detected; A target image confirmation module for entering the interior of the target area to be detected through the incision and compensating for target obstacles in the target area to be detected using an elastic model; A target image marking module for marking recognition points for all the target obstacles, where the recognition points are arranged in a three-dimensional image in the target area to be detected; A target image path selection module for calculating the target path of the target area to be detected using multiple recognition points based on the Grey Wolf Algorithm.

Images