ReCET-based diabetes electric pulse ablation method and system

Abstract

The invention provides a ReCET-based diabetes electric pulse ablation method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining a to-be-ablated point image; performing up-sampling on the image to obtain an up-sampling image; performing loop iteration on the blurred kernel of the up-sampling image and the blurred kernel of the to-be-ablated point image, and performing reconstruction to obtain an ablation point high-definition image; a to-be-ablated area is selected from the ablation point high-definition image; collecting image sets of the to-be-ablated area under different focal lengths; a to-be-ablated area image with the highest definition is screened out, and the shooting focal length is recorded; performing depth estimation on the to-be-ablated area image with the highest definition based on imaging focusing characteristics by combining a shooting focal length, and outputting an estimated distance between an ablation electrode and a to-be-ablated point; and under the condition that the estimated distance is equal to the preset distance, sending out a release electric pulse signal for reminding. The real-time distance between the ablation electrode and the to-be-ablated point is automatically detected, the accuracy is high, the interpretability is high, and medical negligence caused by errors of manual subjective judgment is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a ReCET-based diabetes electrical pulse ablation method and system. Background Art

[0002] ReCET (Reversible Electroporation for Cytosolic Therapy) is an emerging approach to treat type 2 diabetes through electrical pulse ablation technology, which uses an ablation catheter connected to an electrical pulse generator that can precisely transmit pulsed electric fields to the duodenal area of ​​the human body. Signals from the small intestine (particularly the duodenum) are key regulators of metabolic functions, including blood sugar management in the body. When signals in the duodenum become abnormal, the body lacks the ability to properly regulate blood sugar, resulting in continued high blood sugar in patients with type 2 diabetes.

[0003] Type 2 diabetes is the most common type, accounting for more than 90% of all diabetes cases. It is usually associated with insulin resistance and relative insulin deficiency. The ablation catheter used by ReCET passes through the mouth, through the stomach and into the duodenum, the first part of the small intestine. The electrical pulse generator generates precisely controlled electrical pulses, which are delivered through the catheter. The delivery of electrical pulses is intended to initiate the body's natural cell regeneration process, improve cell signaling by promoting the directed regeneration of stem cells in healthy cells, and thus better control blood sugar levels. ReCET technology intervenes in abnormal cell signaling in the duodenum of patients with type 2 diabetes, and by eliminating these cells with poor signals and promoting intestinal cell regeneration, it aims to restore normal metabolic signals, improve the body's ability to control blood sugar, and thus slow the progression of the disease. Regulating the patient's blood sugar level through electrical pulse ablation can regulate blood sugar from the patient's body itself, and can maintain normal blood sugar levels for a long time, thereby avoiding the inconvenience and pain caused by daily insulin injections and blood sugar testing for patients, and reducing medical pressure.

[0004] During the electric pulse ablation procedure, the distance between the ablation electrode and the ablation point directly affects the success of the ablation procedure. If the distance is not accurately determined and the electric pulse signal is sent out, it may even damage the body's normal cells, causing irreversible damage and increasing medical risks. However, the existing electric pulse ablation procedure is mainly based on the doctor's judgment of the distance between the ablation electrode and the ablation point, which is highly subjective. The distance deviation of the pulse signal is large, which is prone to human errors leading to medical accidents and is not conducive to patient recovery. Summary of the invention

[0005] In order to solve the technical problems in the prior art that the distance between the ablation electrode and the ablation point is mainly determined by the physician during the electric pulse ablation procedure, which is highly subjective, has a large deviation in the distance of the pulse signal, is prone to human errors leading to medical accidents, and is not conducive to the patient's recovery, the present invention provides a diabetes electric pulse ablation method and system based on ReCET.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect

[0008] An embodiment of the present invention provides a ReCET-based diabetes electric pulse ablation method, which is applied to an ablation device, wherein the ablation device includes an electric pulse generator and a multi-focus endoscope both connected to an ablation catheter, and the method includes:

[0009] S1: Acquire the image of the point to be ablated;

[0010] S2: performing upsampling on the ablation point image based on bicubic interpolation to obtain an upsampled image;

[0011] S3: iterating the blur kernels of the upsampled image and the image of the ablation point in a loop until the structural similarity index of the adjacent images is greater than the preset structural similarity index, and reconstructing the high-definition image of the ablation point, wherein the blur kernel is a normalized dome kernel;

[0012] S4: Select the area to be ablated in the high-definition image of the ablation point;

[0013] S5: collecting an image set of the area to be ablated at different focal lengths, wherein the image set includes a plurality of images of the area to be ablated;

[0014] S6: Filter out the image of the area to be ablated with the highest definition from the image set, and record the shooting focal length;

[0015] S7: combining the shooting focal length, and based on the imaging focus characteristic, performing depth estimation on the image of the area to be ablated with the highest definition, and outputting the estimated distance between the ablation electrode and the point to be ablated, wherein the imaging focus characteristic is that the image clarity is highest when the shooting object, i.e., the point to be ablated, is at the best focus position;

[0016] S8: When the estimated distance is equal to the preset distance, a release electric pulse signal reminder is issued.

[0017] Second aspect

[0018] An embodiment of the present invention provides a ReCET-based diabetes electrical pulse ablation system, comprising:

[0019] processor;

[0020] A memory having computer-readable instructions stored therein, wherein when the computer-readable instructions are executed by the processor, the ReCET-based diabetes electrical pulse ablation method as in the first aspect is implemented.

[0021] The third aspect

[0022] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the ReCET-based diabetes electrical pulse ablation method according to the first aspect is implemented.

[0023] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0024] In the present invention, a normalized dome core is used to perform high-definition reconstruction of the image of the point to be ablated collected by the endoscope. This allows for a softer transition when processing the image and reduces edge effects. This is suitable for biomedical imaging that requires uniform and smooth processing but does not want to introduce sharp boundaries. The area to be ablated can be accurately framed based on the reconstructed high-definition image of the ablation point. After that, images of the area to be ablated at different focal lengths are collected, and the image with the highest clarity is screened. The estimated distance between the ablation electrode and the area to be ablated is reversely deduced based on the principle that the image clarity is highest when the object, i.e., the point to be ablated, is at the best focusing position. Furthermore, the depth information necessary for ablation is accurately inferred without a known reference object size, thereby compensating for the large deviation and easy human error caused by subjective judgment of depth information, i.e., distance information. This method has high accuracy and strong interpretability, thereby improving the success rate of ablation surgery and patient experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic diagram of a flow chart of a diabetic electrical pulse ablation method based on ReCET provided in an embodiment of the present invention;

[0027] Figure 2 A schematic diagram of the structure of an ablation device provided by an embodiment of the present invention;

[0028] Figure 3 A schematic structural diagram of a ReCET-based diabetes electrical pulse ablation system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0030] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0031] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0032] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0033] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0034] Reference Manual Attached Figure 1 , showing a flow chart of a ReCET-based diabetes electrical pulse ablation method provided in an embodiment of the present invention.

[0035] Reference Manual Attached Figure 2 , showing a schematic structural diagram of an ablation device provided in an embodiment of the present invention.

[0036] An embodiment of the present invention provides a ReCET-based diabetes electric pulse ablation method, which is applied to an ablation device, wherein the ablation device includes an electric pulse generator and a multifocal endoscope both connected to an ablation catheter, and an ablation electrode is connected to the electric pulse generator through the ablation catheter.

[0037] Among them, the electric pulse generator is the core part of the ablation device, which is responsible for generating and controlling electric pulses. These electric pulses are transmitted to the ablation electrodes through the connected ablation catheter, and are used to directly act on specific tissues in the body, such as diseased tissues or abnormal cells, to achieve their ablation (i.e., destruction or removal) in the treatment of diabetes. The intensity, frequency, and duration of the electric pulses can be precisely controlled to meet different treatment needs.

[0038] A multifocal endoscope, also known as a variable-focus endoscope or an endoscope with high dynamic imaging capabilities, is an imaging device, such as the Olympus EVIS EXERA III Series, Pentax Medical EPK Series, and Fujifilm ELUXEOSeries endoscopes, which are used to observe and precisely locate the area to be treated during surgery. Compared with traditional single-focus endoscopes, multifocal endoscopes provide a wider field of view and more focus options, which enables doctors to more clearly observe tissue structures at different depths and ensure the accuracy and safety of ablation operations. Multifocal endoscopes are usually connected to a monitor to display images inside the body in real time.

[0039] The ablation catheter is a flexible tubular structure with one end connected to an electrical pulse generator and the other end equipped with an ablation electrode. Its main function is to safely transmit electrical pulses from the generator to the target location in the body. Ablation catheters usually need to have high flexibility and sufficient strength to be smoothly introduced into the lesion site, and also need to have good biocompatibility to reduce side effects on the human body.

[0040] The ablation electrode is the actuator of the ablation device, which directly contacts the area to be ablated in the body. When the electric pulse is transmitted to the ablation electrode through the ablation catheter, the electrode will generate enough heat or other forms of energy (such as electric field, magnetic field) to destroy the cell structure of the target tissue and achieve the therapeutic effect. The design of the ablation electrode requires precise control of the release of energy to ensure the safety and effectiveness of the operation.

[0041] The processing flow of the ReCET-based diabetes electrical pulse ablation method may include the following steps:

[0042] S1: Acquire the image of the point to be ablated.

[0043] The image of the point to be ablated is an image of the target area determined by the physician to be subjected to electric pulse ablation treatment.

[0044] In a possible implementation, images of the points to be ablated are collected by a multi-focus endoscope.

[0045] It should be noted that even if the images of the ablation points are obtained through a multifocal endoscope, these images may still have some slight blur or distortion at different focal lengths, so further high-definition reconstruction is required. High-definition reconstruction technology can process the images in detail, reduce blur and improve detail clarity, so as to accurately locate the ablation points. This step is crucial because accurate positioning can avoid errors in the operation and prevent irreversible damage to surrounding tissues. In this way, we can significantly improve the accuracy and safety of the operation, thereby ensuring the treatment effect of patients and reducing risks.

[0046] S2: Upsampling the image of the ablation point based on bicubic interpolation to obtain an upsampled image.

[0047] Among them, bicubic interpolation is a commonly used interpolation method in image processing. It estimates the value of the new pixel by using the values ​​of the neighboring 16 pixels, and calculates it using a cubic polynomial. The advantages of this method when upsampling images include the ability to generate smoother image edges and transitions, avoiding the jagged or blurred boundaries that are common when using simpler interpolation methods. Therefore, bicubic interpolation is particularly suitable for applications that need to maintain the visual quality of images, such as medical imaging and high-quality image enlargement. It can maximize the preservation of details and clarity when enlarging images. The upsampled image is obtained by enlarging the low-resolution image, which does not add any new information, but only enlarges the pixels, making the image size larger and more blurred.

[0048] In a possible implementation, S2 specifically includes:

[0049] S201: Determine an upsampling factor, wherein the upsampling factor is a ratio of a resolution of an upsampled image to a resolution of a position corresponding to an image of a point to be ablated.

[0050] S202: interpolating each pixel position in the upsampled image according to the upsample factor:

[0051]

[0052] Where (i, j) represents the pixel index in the image of the point to be ablated, s w and h They represent the width upsampling factor and height upsampling factor respectively, and (i`,j`) represents the pixel index in the upsampled image.

[0053] S203: Calculate the lateral offset dx and the longitudinal offset dy:

[0054]

[0055] in, Indicates rounding down.

[0056] S204: Calculate the contribution weights of adjacent pixels:

[0057]

[0058] Wherein, w() represents the contribution weight, α represents the relative distance of the same pixel before and after upsampling, m and n both represent integers in the range of [-1,2] also known as index offsets, and a=0.5 represents the adjustment factor.

[0059] Where m and n represent the index offset of the 4x4 neighborhood grid around the target pixel (i`, j`) in the original image. Specifically, when |α|<1, the weight function gives the greatest weight to the original pixel closest to the target pixel, and the weight gradually decreases as the distance increases. When 1≤|α|<2, the weight decreases rapidly, reflecting that pixels farther away from the target pixel have less influence on it. When |α|≥2, the weight is 0, which means that original pixels two pixels or more away from the target pixel have no contribution to it. The horizontal weight w(m-dx) reflects the weight from arrive The vertical weight w(n-dy) reflects the effect of horizontal offset. arrive The effect of horizontal offset.

[0060] S205: Based on the contribution weight and the index offset, the adjacent pixels of the image of the ablation point to be ablated are weighted pixel by pixel to obtain an upsampled image:

[0061]

[0062] Wherein, P′(i′,j′) represents the pixel value of the upsampled image at the pixel index (i`,j`), and P() represents the pixel value in the image to be ablated.

[0063] Specifically, when implementing step S2, the upsampling factor is first determined as the proportional relationship between the image of the point to be ablated and the resolution of the upsampled image. This step determines how many times interpolation needs to be performed at each pixel position to enlarge the image to the required resolution. Next, the value of each pixel in the upsampled image is calculated by an interpolation algorithm, which involves accurately positioning the pixels according to the sampling factors in width and height. When calculating the horizontal and vertical offsets, ensure that the actual position offset of the original image pixels is taken into account during the interpolation process to improve the accuracy of the interpolation. Then, by calculating the contribution weights of adjacent pixels, the weights can be adjusted according to the relative distance of the pixels to ensure that the original image pixels closest to the target pixel have the greatest impact on the result, and gradually reduce the impact of pixels far away from the target. This weighting strategy helps preserve image details and reduce interpolation errors.

[0064] Finally, by applying these weights to the neighboring pixels of the image of the ablation point, the upsampled image is weighted pixel by pixel. This process ensures that the quality and details of the enlarged image are as close to the real image as possible, avoiding information loss or blurring during the enlargement process, thus providing a clear and accurate image basis for subsequent image processing and analysis. This meticulous processing method significantly improves the visual quality and practicality of the image, and provides reliable data support for precise positioning and treatment.

[0065] S3: The blur kernels of the upsampled image and the image of the ablation point are iterated cyclically until the structural similarity index of the adjacent images is greater than the preset structural similarity index, and a high-definition image of the ablation point is reconstructed.

[0066] The blur kernel is a normalized dome kernel.

[0067] Among them, the blur kernel (or convolution kernel) is used to blur the image, usually by convolution operation with the image. The blur kernel defines how the pixels in the neighborhood around each pixel affect the final value of the pixel. Kernels of different shapes and sizes will produce different blur effects. The normalized dome kernel is a blur kernel of a specific shape. Its shape is similar to a dome or a hemisphere. The value in the center is the largest and gradually decreases toward the edge. It is similar to a Gaussian kernel, but the edge usually drops more gently. Normalization means that the sum of all values ​​in the kernel is 1. This is to maintain the consistency of the image brightness and ensure that the total brightness of the image remains unchanged after the convolution operation.

[0068] It should be noted that the method of using a normalized form of the dome kernel to iterate the upsampled image and the image of the ablation point greatly improves the image quality, which is crucial for precise medical operations. As a smooth but edge-preserving blur kernel, the dome kernel iteratively optimizes the structural similarity index of the image, making the reconstructed high-definition image closer to the actual situation in detail, reducing noise and artifacts while maintaining the overall brightness of the image. This method significantly enhances the clarity and contrast of the image, allowing doctors to more accurately identify and locate the area to be ablated, thereby improving the success rate and safety of the operation. In addition, this careful adjustment of image processing can also help reduce the risk of misdiagnosis due to image quality problems, providing patients with safer and more effective treatment options.

[0069] In actual applications, although multifocal endoscopes are used to capture clear images at different focal lengths, in complex clinical environments, single image capture often fails to provide all the necessary details. High-definition reconstruction of ablation images can further enhance the details and clarity of the images, ensuring that doctors can more accurately identify and locate lesions. In addition, the reconstruction process can optimize image quality, reduce noise, and enhance contrast through algorithms, making ablation operations safer and more effective. This is particularly important for the implementation of precise medical interventions, which can help doctors make better treatment decisions, thereby improving the success rate of surgery and patient safety.

[0070] In a possible implementation, the normalized dome kernel is specifically:

[0071]

[0072] Among them, r represents the dome kernel k, that is, the kernel radius of the blur kernel, k(xh ,y h ) represents the coordinate (x h ,y h ), π represents the value of pi.

[0073] Among them, the dome kernel is a filter kernel with uniform values ​​within a circular area of ​​fixed radius, and zero values ​​outside the circular area. The characteristic of the dome kernel is that the edges are smoother than the standard box kernel, which enables it to transition more softly when processing images and reduce edge effects. Due to its simplicity and smoothness, it is very useful in some specific application scenarios, especially in biomedical imaging and optical simulations where uniform smoothing is required but sharp boundaries are not desired. For example, it can be used to simulate blurring due to inherent limitations of optical systems (such as the aperture effect of lenses). The purpose of the normalized form of the dome kernel is to be used in image processing. The dome kernel is normalized to ensure that it does not change the overall brightness of the image when applied. Normalization is usually done by ensuring that the sum of all values ​​in the kernel is 1.

[0074] During the iterative process of image reconstruction, this preliminary high-resolution image estimate will be continuously updated and optimized. The algorithm will gradually reduce the introduced artifacts while trying to recover more details and textures from the low-resolution data. Through the estimation of the blur kernel and deconvolution processing, the reconstructed image quality gradually approaches the real visual effect of the original scene.

[0075] In a possible implementation, S3 specifically includes:

[0076] S301: Determine the up-sampling image update function in combination with total variation regularization:

[0077]

[0078] Φ(x)=||▽x||1

[0079] Among them, x (n+1) represents the upsampled image estimate obtained after n+1 iterations of the upsampled image x, argmin x Represents taking function value The minimum x, y represent the image of the point to be ablated, * represents the convolution operation, and D() represents the downsampling of the blur kernel k and the upsampled image x. represents the square of the Euclidean norm, λ represents the first regularization parameter, Φ(x) represents the total variation regularization function with respect to the upsampled image x, ▽x represents the gradient of the upsampled image x, that is, the rate of change of the pixel value of each point in the upsampled image, and || ||1 represents the L1 norm.

[0080] in, It is a data fidelity term, which is used to measure the difference between the image generated by the current high-resolution image estimate and the blur kernel through the convolution operation (simulating the blur produced by the image passing through the system) and the downsampling operation (simulating the image sampling process) and the actual observed low-resolution image, that is, the image of the ablation point. λΦ(x) is the regularization term in the objective function, where Φ(x) is the regularization function applied to the image, which is usually used to introduce additional prior knowledge or desired properties (such as image smoothness, sparsity, etc.), and λ is a regularization parameter that adjusts the relative importance between the data fidelity term and the regularization term. A larger λ value increases the regularization effect, helps prevent overfitting and improves the stability of the solution. The prediction quality of the image is gradually improved by iteratively adjusting the upsampled image, while dealing with the blur and sampling restrictions that the image may be subject to. This process not only attempts to make the simulated blurred image as close as possible to the actual observation, but also ensures that the solution follows certain smoothness or other image quality standards. This method is very effective in recovering more details by processing low-resolution images, especially when no direct high-resolution images are available.

[0081] Among them, total variation regularization can maintain the edges of the image. In the image reconstruction process, the edges usually correspond to areas where the pixel values ​​change rapidly, and these areas have large values ​​in the gradient map. Using ||▽x||1 as a regularization term can preserve these edges because it tends to maintain larger gradient values. ||▽x||1 means summing the absolute value of the gradient vector at each pixel of the upsampled image. The L1 norm helps promote the sparsity of the image, especially in areas with small gradients, making the image smoother while maintaining the edges with large gradient changes.

[0082] S302: Determine the blur kernel update function in combination with the kernel regularization term, and determine the blur kernel constraint to ensure the blur kernel is effective:

[0083]

[0084] k≥0,∑k=1

[0085] Among them, k (n+1) The blur kernel estimate obtained after n+1 iterations of blur kernel k, Ψ(k) represents the kernel regularization term about blur kernel k, argmin k represents the k value that minimizes the function, and μ represents the second regularization parameter.

[0086] Among them, the kernel regularization term is the L2 norm, which is used to maintain the smoothness of the kernel. The blur kernel constraint ensures that the blur kernel is effective. In the actual physical system, the blur kernel represents the diffusion of light, and the light intensity cannot be negative, that is, it is non-negative and the sum is 1. By gradually optimizing the blur kernel, the reconstructed image is clearer while maintaining consistency with the observed data. In addition, by introducing regularization terms and constraints, excessive degrees of freedom of the solution can be avoided, thereby obtaining a more stable and reliable reconstruction result.

[0087] S303: Fix the upsampled image and update the blur kernel using the upsampled image update function.

[0088] S304: Fix the blur kernel and update the upsampled image with the blur kernel update function under the blur kernel constraint.

[0089] S305: Calculate the structural similarity index between the current up-sampled image and the up-sampled image before updating:

[0090]

[0091] Wherein, SSIM() represents the structural similarity index, x A ,x B Respectively represent the current upsampled image and the upsampled image before updating, Respectively represent the average brightness of the current upsampled image and the upsampled image before updating, Respectively represent the standard deviation of the current upsampled image and the upsampled image before updating, i.e., the contrast. It represents the covariance, i.e., structural similarity, between the current upsampled image and the upsampled image before update. Both c1 and c2 are non-zero constants.

[0092] The purpose of the non-zero constants is to avoid the denominator being zero. In the SSIM calculation, the average brightness or standard deviation may be zero, especially in some uniform areas or low-contrast areas of the image. To avoid division by zero during the calculation process, these non-zero constants are added to the numerator and denominator of the formula to ensure the stability of the calculation. Specifically, c1=(k1L) 2 , c2=(k2L) 2 , where L is the dynamic range of pixel values, k1 is generally taken as 0.01, and k2 is generally taken as 0.03.

[0093] S306: the number of iterations is increased by 1, and the process returns to step S303 until the structural similarity index is greater than the preset structural similarity index or the number of iterations is greater than the preset number of iterations, and a high-definition image of the ablation point is output.

[0094] It should be noted that those skilled in the art can set the preset structural similarity index and the preset number of iterations according to actual needs, and the present invention does not limit this herein.

[0095] It should be noted that in the process of realizing high-definition reconstruction of upsampled images, first, the update function of the upsampled image is determined by using total variation regularization (TV regularization), which helps to reduce noise and unnecessary blur in the image while maintaining the details of the image edge, thereby improving the overall quality of the image. Total variation regularization retains the clarity of the image edge by optimizing the L1 norm of the image gradient, while the smooth area appears more uniform. Subsequently, the focus is on updating the blur kernel, and the kernel regularization term is combined to ensure the smoothness and effectiveness of the blur kernel. By introducing the L2 norm regularization, it is ensured that the blur kernel is not too complex and its physical feasibility (such as non-negativity and normalization properties) is maintained. This approach makes the image more realistic and consistent when simulating the real blur process, thereby improving the quality of the reconstructed image. In S303 and S304, the upsampled image and blur kernel are fixed respectively, and each component is ensured to be finely optimized by alternating updates, further improving the accuracy and clarity of the reconstructed image. The similarity between the current image and the previous iteration image is evaluated by calculating the structural similarity index, ensuring that the image is significantly improved in each iteration. Finally, by setting the conditions of the structural similarity index and the number of iterations, the iterative update is performed until the preset requirements are met, thereby outputting a high-definition reconstructed ablation point image. The design of this step ensures that the reconstructed image is not only clearer visually, but also highly consistent with the actual observed image in terms of data, significantly improving the reliability and accuracy of image processing.

[0096] S4: The area to be ablated is selected in the high-definition image of the ablation point.

[0097] Among them, the area to be ablated in the high-definition image of the ablation point is displayed more clearly, which can assist medical personnel in accurately locating and selecting the exact ablation position. Each pixel point in the area to be ablated only enhances the clarity, and the pixel point index is one-to-one corresponding to the pixel point index of the original image to be ablated.

[0098] S5: Acquire image sets of the area to be ablated at different focal lengths.

[0099] The image set includes multiple images of the area to be ablated.

[0100] It should be noted that the images of the same area, i.e. the area to be ablated, under different focal lengths of the multifocal endoscope are collected. These images can provide a relative clarity scale comparison to facilitate the subsequent depth reasoning of the ablation electrode and the point to be ablated based on the correlation between clarity and focal length.

[0101] S6: Filter out the image of the area to be ablated with the highest definition from the image set, and record the shooting focal length.

[0102] It should be noted that the purpose of selecting the clearest image of the area to be ablated from the image set collected at different focal lengths and recording the shooting focal length is to ensure that the best quality images are used during the electric pulse ablation operation. Selecting the clearest image can greatly improve the visualization quality of the diseased tissue, allowing doctors to accurately identify and locate the lesion. Recording the shooting focal length is crucial for subsequent treatment steps, because this data is directly related to the depth information of the focus, which helps to calculate and adjust the exact position of the ablation device, thereby ensuring the accuracy and safety of the ablation operation. This step significantly improves the effectiveness of the treatment and reduces surgical risks and patient recovery time.

[0103] In a possible implementation, the image of the area to be ablated with the highest definition is selected from the image set by using the Laplace operator:

[0104] S max =max|▽ 2 I i |

[0105] Among them, 2 represents the Laplace operator, S max Represents i images of the area to be ablated I i The image of the area to be ablated with the highest definition, || indicates the absolute value.

[0106] It should be noted that the Laplace operator is used to select the image with the highest definition, and the edge information of the image is enhanced using the operator, so that the image with high definition has more obvious details and edges, thereby effectively distinguishing and selecting the clearest image in the image of the area to be ablated. This method can accurately extract important features in the image, ensure that the selected image is superior to other images in terms of clarity, and provide the most conducive data basis for subsequent analysis and processing.

[0107] S7: combining the shooting focal length and based on the imaging focus characteristics, depth estimation is performed on the image of the area to be ablated with the highest definition, and the estimated distance between the ablation electrode and the point to be ablated is output.

[0108] Among them, the imaging focus characteristic is that the image clarity is highest when the photographed object, i.e., the point to be ablated, is at the best focus position.

[0109] It should be noted that the clearest image previously screened out can be used to determine the exact distance between the point to be ablated and the ablation electrode by using the position of its best focusing point. The advantage of this method is that it allows the inherent focusing characteristics of the imaging system to be used to provide very precise depth information, thereby ensuring that the ablation electrode can be accurately positioned to the predetermined treatment area during execution. Accurate depth estimation is crucial for the effective delivery of electrical pulses, as it is directly related to optimizing the ablation effect, reducing damage to surrounding healthy tissues and the safety of the entire operation. In this way, we not only improve the accuracy of the operation, but also significantly reduce the risk of complications, bringing patients a higher treatment success rate and better recovery expectations.

[0110] In a possible implementation manner, the estimated distance is specifically:

[0111]

[0112] Where Z represents the estimated distance, d best represents the shooting focal length, and f represents the focal length of the lens.

[0113] Among them, the estimated distance is the distance from the point to be ablated to the multifocal endoscope lens, and the shooting focal length is the distance from the lens to the imaging sensor. Here, the lens focal length and the shooting focal length are both known, and then the depth of the point to be ablated, that is, the distance from the point to be ablated to the multifocal endoscope lens, is accurately estimated. Since the clarity of the image changes when the object is at different positions from the camera, the clearest image corresponds to the position of the object on the best focus plane. Through this inverse reasoning method, we can accurately determine the depth of the point to be ablated, thereby optimizing the ablation process and improving the accuracy and safety of the operation.

[0114] S8: When the estimated distance is equal to the preset distance, a release electric pulse signal reminder is issued.

[0115] It is understandable that when the estimated distance between the ablation electrode and the point to be ablated calculated by the previous steps is equal to the preset distance, the system will issue a reminder to release the electric pulse signal. The importance of this step is to ensure that the release of the electric pulse occurs precisely at the optimal treatment position. The preset distance is set in advance according to the treatment plan and the specific requirements of the ablation point, with the aim of ensuring that the ablation operation can accurately target the diseased tissue while protecting the surrounding healthy tissue from being affected to the greatest extent possible. When the ablation electrode is located at the precise treatment position, the release of the electric pulse can effectively ablate the target tissue while avoiding damage to non-target areas. In addition, this reminder also helps the operating physician to monitor and adjust the entire treatment process in real time to ensure the accuracy and safety of the ablation operation. In this way, the entire ablation process becomes more controllable, reduces surgical risks, and improves treatment outcomes.

[0116] It should be noted that those skilled in the art can set the size of the preset distance according to actual needs, and the present invention is not limited here.

[0117] In a possible implementation manner, after S8, the method further includes:

[0118] Visualize the estimated distances.

[0119] It is understandable that visualizing the estimated distance can help the operator intuitively understand the actual distance between the ablation point and the multifocal endoscope lens. This visualization not only provides clear depth information and improves the accuracy of the operation, but also helps to quickly confirm whether the distance meets the preset standard, thereby reducing the risk of operating errors and improving the safety and effectiveness of the entire ablation process.

[0120] Specifically, at the beginning, a multifocal endoscope is used to obtain preliminary images of the ablation point (S1). These images are upsampled by bicubic interpolation (S2) to increase their resolution. Although this does not increase the original information of the image, it provides more data points and higher image quality for the next step of image reconstruction. Subsequently, a normalized dome kernel is applied in S3 for image reconstruction. This kernel is particularly suitable for biomedical images because it can smooth the image while keeping the edges clear. The structural similarity index of the image is improved by iteration, and finally a high-definition ablation point image is obtained. This processing not only improves the image quality, but also lays the foundation for the precise operation of the subsequent steps. Based on the high-definition image, the area to be ablated is accurately selected in S4. The accuracy of this step is directly supported by the improvement of image quality in the previous steps. Next, a set of images of the area at different focal lengths is collected in S5. This step makes full use of the ability of the multifocal endoscope to capture images from multiple angles and focal lengths to ensure that all details of the area to be treated are covered, which is crucial for the subsequent selection of the best image for depth estimation. In S6, the image with the highest clarity is screened from the image set and its shooting focal length is recorded. The selection of the clearest image in this step is based on the comprehensiveness of the previous multi-focal length image set, ensuring that the selected image has the highest detail visibility. Then in S7, depth estimation is performed by combining the shooting focal length and imaging focus characteristics (i.e., the principle that the image clarity is highest at the optimal focus position). This step accurately calculates the distance between the ablation electrode and the point to be ablated, reducing human estimation errors and improving the accuracy of the operation. Finally, when the estimated distance reaches the preset value, the system issues an electric pulse release reminder (S8). This reminder ensures that the electric pulse is released at the precise moment in the clinical operation, increasing the success rate of the treatment.

[0121] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0122] In the present invention, a normalized dome core is used to perform high-definition reconstruction of the image of the point to be ablated collected by the endoscope. This allows for a softer transition when processing the image and reduces edge effects. This is suitable for biomedical imaging that requires uniform and smooth processing but does not want to introduce sharp boundaries. The area to be ablated can be accurately framed based on the reconstructed high-definition image of the ablation point. After that, images of the area to be ablated at different focal lengths are collected, and the image with the highest clarity is screened. The estimated distance between the ablation electrode and the area to be ablated is reversely deduced based on the principle that the image clarity is highest when the object, i.e., the point to be ablated, is at the best focusing position. Furthermore, the depth information necessary for ablation is accurately inferred without a known reference object size, thereby compensating for the large deviation and easy human error caused by subjective judgment of depth information, i.e., distance information. This method has high accuracy and strong interpretability, thereby improving the success rate of ablation surgery and patient experience.

[0123] Reference Manual Attached Figure 3 , showing a structural schematic diagram of a ReCET-based diabetes electric pulse ablation system provided by the present invention.

[0124] The present invention further provides a ReCET-based diabetes electrical pulse ablation system 20, which is applied to the above-mentioned ReCET-based diabetes electrical pulse ablation method, comprising:

[0125] Processor 201.

[0126] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201 , the ReCET-based diabetes electrical pulse ablation method as in the method embodiment is implemented.

[0127] The ReCET-based diabetes electric pulse ablation system 20 provided by the present invention can execute the above-mentioned ReCET-based diabetes electric pulse ablation method and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0128] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0129] In the present invention, a normalized dome core is used to perform high-definition reconstruction of the image of the point to be ablated collected by the endoscope. This allows for a softer transition when processing the image and reduces edge effects. This is suitable for biomedical imaging that requires uniform and smooth processing but does not want to introduce sharp boundaries. The area to be ablated can be accurately framed based on the reconstructed high-definition image of the ablation point. After that, images of the area to be ablated at different focal lengths are collected, and the image with the highest clarity is screened. The estimated distance between the ablation electrode and the area to be ablated is reversely deduced based on the principle that the image clarity is highest when the object, i.e., the point to be ablated, is at the best focusing position. Furthermore, the depth information necessary for ablation is accurately inferred without a known reference object size, thereby compensating for the large deviation and easy human error caused by subjective judgment of depth information, i.e., distance information. This method has high accuracy and strong interpretability, thereby improving the success rate of ablation surgery and patient experience.

[0130] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0131] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0132] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0133] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0134] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0135] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0136] Those skilled in the art will appreciate that the units 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 and design constraints of the technical solution. 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 the present invention.

[0137] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0138] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0139] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0140] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0141] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0142] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the ReCET-based diabetes electrical pulse ablation method as described in the method embodiment is implemented.

[0143] A computer-readable storage medium provided by the present invention can implement the steps and effects of the ReCET-based diabetes electrical pulse ablation method of the above method embodiment. To avoid repetition, the present invention will not go into details.

[0144] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0145] In the present invention, a normalized dome core is used to perform high-definition reconstruction of the image of the point to be ablated collected by the endoscope. This allows for a softer transition when processing the image and reduces edge effects. This is suitable for biomedical imaging that requires uniform and smooth processing but does not want to introduce sharp boundaries. The area to be ablated can be accurately framed based on the reconstructed high-definition image of the ablation point. After that, images of the area to be ablated at different focal lengths are collected, and the image with the highest clarity is screened. The estimated distance between the ablation electrode and the area to be ablated is reversely deduced based on the principle that the image clarity is highest when the object, i.e., the point to be ablated, is at the best focusing position. Furthermore, the depth information necessary for ablation is accurately inferred without a known reference object size, thereby compensating for the large deviation and easy human error caused by subjective judgment of depth information, i.e., distance information. This method has high accuracy and strong interpretability, thereby improving the success rate of ablation surgery and patient experience.

[0146] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0147] There are a few points to note:

[0148] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0149] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0150] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0151] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A ReCET-based diabetes electrical pulse ablation method, characterized in that: Applied to an ablation device, wherein the ablation device comprises an electric pulse generator and a multi-focus endoscope both connected to an ablation catheter, an ablation electrode is connected to the electric pulse generator via the ablation catheter, and the method comprises: S1: Acquire the image of the point to be ablated; S2: performing upsampling based on bicubic interpolation on the image of the to-be-ablated point to obtain an upsampled image; S3: performing a cyclic iteration on the blur kernels of the upsampled image and the image of the point to be ablated until the structural similarity index of the adjacent images is greater than a preset structural similarity index, and reconstructing a high-definition image of the ablation point, wherein the blur kernel is a normalized dome kernel; S4: Selecting a region to be ablated in the high-definition image of the ablation point; S5: Acquire an image set of the area to be ablated at different focal lengths, wherein the image set includes a plurality of images of the area to be ablated; S6: Filtering out the image of the area to be ablated with the highest definition from the image set, and recording the shooting focal length; S7: performing depth estimation on the image of the area to be ablated with the highest clarity based on the imaging focus characteristic in combination with the shooting focal length, and outputting an estimated distance between the ablation electrode and the point to be ablated, wherein the imaging focus characteristic is that the clarity of the image is highest when the object to be shot, i.e., the point to be ablated, is at the best focus position; S8: When the estimated distance is equal to the preset distance, a release electric pulse signal reminder is issued.

2. The ReCET-based diabetes electrical pulse ablation method according to claim 1, characterized in that: The images of the ablation points are collected through a multifocal endoscope.

3. The ReCET-based diabetes electrical pulse ablation method according to claim 1, characterized in that: The S2 specifically includes: S201: Determine an upsampling factor, wherein the upsampling factor is a ratio of the resolution of the upsampled image to the resolution of the position corresponding to the image of the point to be ablated; S202: interpolating each pixel position in the upsampled image according to the upsampled factor; S203: Calculate lateral offset and longitudinal offset; S204: Calculate contribution weights of adjacent pixels; S205: Based on the contribution weight and the index offset, weight the adjacent pixels of the image of the point to be ablated pixel by pixel to obtain the up-sampled image.

4. The ReCET-based diabetes electrical pulse ablation method according to claim 3, characterized in that: The S3 specifically includes: S301: Determine an up-sampling image update function in combination with total variation regularization; S302: Determine a fuzzy kernel update function in combination with a kernel regularization term, and determine a fuzzy kernel constraint that ensures the fuzzy kernel is effective; S303: Fix the up-sampled image, and update the blur kernel using the up-sampled image update function; S304: Fix the blur kernel, and update the up-sampled image with the blur kernel update function under the constraint of the blur kernel; S305: Calculating a structural similarity index between the current up-sampled image and the up-sampled image before updating; S306: the number of iterations is increased by 1, and the process returns to step S303, until the structural similarity index is greater than a preset structural similarity index or the number of iterations is greater than a preset number of iterations, and the high-definition image of the ablation point is output.

5. The ReCET-based diabetes electrical pulse ablation method according to claim 1, characterized in that: The image of the area to be ablated with the highest definition is selected from the image set by using the Laplace operator.

6. The ReCET-based diabetes electrical pulse ablation method according to claim 1, characterized in that: After S8, the method further includes: The estimated distance is visualized.

7. A diabetes electrical pulse ablation system based on ReCET, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the ReCET-based diabetes electrical pulse ablation method as described in any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the ReCET-based diabetes electrical pulse ablation method as described in any one of claims 1 to 6 is implemented.

Images