Image recognition-based operation early warning method, device, and endoscope

By using image recognition technology to determine the point of contact and impact angle between the fiber tip and the target stone, and providing early warning information, this method solves the problem of tissue damage caused by the randomness of stone location in holmium laser lithotripsy, and enables precise stone movement and safe operation.

CN120814900BActive Publication Date: 2025-12-02HUNAN VATHIN MEDICAL INSTR CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202511319220.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-09-16
Publication Date
2025-12-02
Estimated Expiration
2045-09-16

AI Technical Summary

Technical Problem

During holmium laser lithotripsy, the randomness of the stone's location leads to disordered movement, increasing the risk of damaging the inner wall of the tissue. Existing technologies make it difficult to precisely adjust the stone's position.

Method used

Image recognition technology is used to determine the point of impact between the fiber optic end and the target stone, calculate the impact angle, and provide early warning information to help users move the stone to the target area.

Benefits of technology

It reduces the difficulty of using endoscopes, decreases the risk of tissue damage caused by disordered stone movement, and improves the accuracy and safety of lithotripsy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120814900B_ABST
    Figure CN120814900B_ABST
Patent Text Reader

Abstract

This application provides an operation early warning method, device, and endoscope based on image recognition, relating to the field of equipment control technology. The method determines the point of impact between the fiber optic end and the target stone during endoscope operation. It then determines the impact angle using the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint. By predicting the impact of different points of impact of the fiber optic on the target stone, reliable predictive information is provided to the user. This assists the user in adjusting the point of impact between the fiber optic end and the target stone to move the target stone to the target area. This reduces the difficulty of using the endoscope and also reduces the risk of damage to the inner wall of the tissue caused by disordered movement of the target stone.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of equipment control technology, specifically to an operation early warning method, device, and endoscope based on image recognition. Background Technology

[0002] The working principle of holmium laser lithotripsy is mainly based on its unique photothermal and photoacoustic effects. By exciting holmium atoms in the crystal with a light source, a pulsed laser with a wavelength of 2100nm is released. This wavelength is located in the near-infrared spectral region and can be efficiently absorbed by water and water in biological tissues. Regarding the photothermal effect: the laser is absorbed by water molecules on and around the stone surface, instantly generating high temperatures. The water on the stone surface vaporizes, forming microbubbles, i.e., cavitation. The expansion and collapse of these vapor bubbles generate shock waves, achieving mechanical fragmentation of the stone. Regarding the photoacoustic effect: when the high-energy pulsed laser impacts the stone, it generates high-frequency vibration waves, directly causing the internal structure of the stone to disintegrate.

[0003] Regarding the photothermal effect of holmium lasers, with effective control of the excitation source, the cavitation effect in the photothermal effect can be amplified, thereby repelling the stones without breaking them. For example, a surgical laser system with controllable recoil adjustment function disclosed in European patent WO2025147452A1 uses holmium lasers to change the position of stones.

[0004] Due to the spatial structure of the kidney, the position of kidney stones is random when using lasers to adjust their location. After being repelled, the disorderly movement of the stones poses a risk of damaging the inner wall of the tissue. Summary of the Invention

[0005] This application provides an operation warning method, device and endoscope based on image recognition, which can reduce the difficulty of using the endoscope and reduce the risk of damage to the inner wall of the tissue caused by the disorderly movement of the target stone.

[0006] This application provides an operation early warning method based on image recognition, including:

[0007] Acquire a target image, the target image including the target stone and the end of an optical fiber;

[0008] The target adjustment region is determined based on the target image; the target adjustment region represents the desired adjustment area of ​​the target stone;

[0009] Determine the point of impact between the fiber optic end and the target stone, and determine the impact angle with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint;

[0010] Early warning information for circumferential rotation is determined based on the impact angle.

[0011] To achieve the above and other related objectives, this application provides an operation early warning device based on image recognition, comprising:

[0012] The data acquisition module is used to acquire a target image, which includes a target stone and an optical fiber end;

[0013] A region determination module is used to determine a target adjustment region based on a target image; the target adjustment region represents the desired adjustment region of the target stone.

[0014] The angle determination module is used to determine the point of impact between the end of the optical fiber and the target stone, and to determine the impact angle with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint.

[0015] The rotation warning module is used to determine circumferential rotation warning information based on the impact angle.

[0016] To achieve the above and other related objectives, this application also provides an endoscope, including an endoscope host, the endoscope host comprising:

[0017] One or more processors;

[0018] Memory used to store the executable program code of the processor;

[0019] The processor is configured to execute the program code to implement the above-described image recognition-based operation warning method.

[0020] As described above, the operation warning method, device, and endoscope based on image recognition provided in this application have the following beneficial effects:

[0021] This application discloses an image recognition-based operation warning method. During endoscopy operation, the method determines the point of impact between the fiber optic end and the target stone. Using the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint, it determines the impact angle. By predicting the impact of different points of impact of the fiber optic on the target stone, it provides reliable predictive information to the user, assisting the user in adjusting the point of impact between the fiber optic end and the target stone to move the target stone to the target area. This reduces the difficulty of using the endoscope and also reduces the risk of damage to the tissue wall caused by disordered movement of the target stone.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0024] Figure 1 This is a schematic diagram illustrating the implementation environment of an image recognition-based operation early warning method, as shown in an exemplary embodiment of this application.

[0025] Figure 2 This is a flowchart illustrating an image recognition-based operation warning method as an exemplary embodiment of this application;

[0026] Figure 3 This is a structural block diagram of an image recognition-based operation warning device, as illustrated in an exemplary embodiment of this application. Detailed Implementation

[0027] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0028] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0029] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0030] Please see Figure 1 This is a schematic diagram illustrating the implementation environment of an image recognition-based operation warning method, which is an exemplary embodiment of this application.

[0031] Endoscope 110 can be inserted into renal calyx 120. A camera 111 can be installed at the tip of the endoscope. An optical fiber 112 extends from the endoscope's working channel to the tip for laser lithotripsy. During laser lithotripsy, the target stone is not fixedly connected to the renal cavity wall. The target stone changes position under the influence of the laser shock wave, resulting in unsatisfactory lithotripsy. A shock wave generated by a holmium laser is needed to reposition the target stone, pushing it into a target adjustment area for better lithotripsy. The target adjustment area can be the upper calyx 121, middle calyx 122, lower calyx 123, or renal pelvis. Two dashed lines adjacent to camera 111 represent the field of view of camera 111.

[0032] However, the change in the position of the stone under the action of holmium laser is random. It can only rely on the user's experience to change the point of contact between the holmium laser and the stone so that the target stone can move to the target adjustment area. This adjustment method is relatively complicated. If the user is not experienced enough, the stone may move randomly under the action of laser, which may damage the inner wall of the tissue.

[0033] The image recognition-based operation warning method provided in this application embodiment can determine the impact angle by taking the center of the target stone as the vertex 150, the center of the target adjustment area 160 as the endpoint, and the point of action between the end of the optical fiber and the target stone 140 as the endpoint. It can then predict whether the stone may move to the target adjustment area under the action of the laser based on the impact angle. If it does, the operation can be performed directly. If not, a circumferential rotation warning message can be determined and displayed to prompt the user to change the circumferential position of the optical fiber, so that the user can move the target stone to the target adjustment area by controlling the position of the optical fiber.

[0034] Please see Figure 2 , Figure 2 This is a flowchart illustrating an image recognition-based operation warning method in an exemplary embodiment of this application. This image recognition-based operation warning method can be applied to... Figure 1 The implementation environment is shown. (Reference) Figure 2 It can be seen that this image recognition-based operation early warning method may include:

[0035] Step S210: Obtain the target image.

[0036] The target image includes the target stone and the end of the optical fiber.

[0037] In one embodiment of this application, a target image captured by an endoscopic camera can be acquired. The target image may include the fiber optic end and the target stone. When the target image includes multiple stones, the target stone can be the stone closest to the fiber optic end in the target image; when the target image includes only one stone, that stone can be identified as the target stone.

[0038] Step S220: Determine the target adjustment area based on the target image.

[0039] The target adjustment region represents the desired adjustment area for the target stone.

[0040] In one embodiment of this application, the target adjustment area can be determined based on the target image.

[0041] In one possible implementation, the target adjustment area can be selected by the user. For example, the image captured by the endoscope camera can be displayed on the endoscope host's screen, and the user can select the target adjustment area from the image displayed on the endoscope host via buttons or touch screen controls.

[0042] In another possible implementation, a three-dimensional model of the renal calyces, including the target stone, can be obtained in advance. This model can include labels such as the renal pelvis, upper calyx, middle calyx, lower calyx, and the stone itself. The center point of the renal pelvis, the opening position of each calyx, the direction and length of the calyx neck, and the relative positions between the calyces can be calculated from the three-dimensional model. After the endoscope's camera is inserted, each frame of the image can be analyzed to determine the current position of the endoscope lens. Once the target stone is identified, the renal pelvis, upper calyx, middle calyx, or lower calyx can be selected as the target area in the region near the target stone. Since the renal pelvis, upper calyx, middle calyx, and lower calyx are not fully displayed in the image, a portion of the image display area of ​​the renal pelvis, upper calyx, middle calyx, or lower calyx can be used as the target area.

[0043] Step S230: Determine the point of impact between the end of the optical fiber and the target stone, and determine the impact angle with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint.

[0044] In one embodiment of this application, the point of impact between the fiber optic end and the target stone can be determined, and the impact angle can be determined with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint. Ideally, the impact angle is close to 180 degrees, which makes the point of impact between the fiber optic end and the target stone more likely to have a positive effect on the movement of the target stone toward the target area.

[0045] For example, the coordinates of the fiber optic end and the contour of the stone can be determined from the target image. If the endoscope is a dual-camera system, a stereo matching algorithm can be used to calculate the disparity, and the disparity can be converted into a depth map based on the camera calibration parameters. The depth values ​​of the fiber optic end and the stone region can then be extracted from the depth map to obtain the 3D coordinates of the fiber optic end and the 3D point cloud of the stone surface. If the endoscope is monocular, a monocular depth estimation model and the target image can be used to output a depth map. The 3D coordinates can be determined based on the coordinates of the fiber optic end in the depth map. A 3D point cloud of the stone surface can be generated based on the depth map and a segmentation mask. The target image can be input into a pre-trained semantic segmentation model to obtain the segmentation mask. Based on the 3D coordinates of the fiber optic end and the 3D point cloud of the stone surface, the laser beam is modeled as a ray. The ray-surface intersection algorithm is used to calculate the intersection point with the stone surface, thus obtaining the point of interaction between the fiber optic end and the target stone.

[0046] For example, the target image can be the original RGB image, which can be converted to other color spaces, such as HSV, to obtain an HSV image. A Gaussian filter or median filter is used to filter the HSV image to obtain a smoothed image. Histogram equalization is then used to process the smoothed image, increasing the gray-level difference between the stone region and the surrounding area, resulting in a contrast-enhanced image. Based on the contrast-enhanced image, Otsu's method is used to obtain a global threshold, and the pixels in the contrast-enhanced image are divided based on the global threshold to obtain a binary image. In the binary image, the pixel value of the stone region is 255, and the pixel value of the background region is 0. Alternatively, a pre-trained semantic segmentation model can be used. The target image is input into the pre-trained semantic segmentation model to obtain a segmentation mask output by the semantic segmentation model. Each pixel value in the segmentation mask represents the probability that it belongs to the stone or the background.

[0047] Based on binary images or segmentation masks, stone boundaries can be extracted, and edge detection algorithms can be used to find the sequence of all boundary points representing the stone region. The i-th contour It is a by The set of coordinate points represents the boundary of a closed or polygonal area. The stone closest to the end of the optical fiber can be identified as the target stone.

[0048] After identifying the target stone, its outline can be obtained. Based on this outline, the centroid coordinates of the target stone can be determined. The centroid x-coordinate is calculated as the sum of the x-coordinates of all points divided by the total number of points, and the centroid y-coordinate is calculated as the sum of the y-coordinates of all points divided by the total number of points. The centroid coordinates are then used as the center of the target stone. If the centroid of the target stone is located outside its outline, the minimum circumcircle algorithm can be used to obtain the center coordinates of the circle, which are then used as the center of the target stone.

[0049] Step S240: Determine circumferential rotation warning information based on the impact angle.

[0050] In one embodiment of this application, circumferential rotation warning information can be determined based on the impact angle. The circumferential rotation warning information can indicate to the user that the current impact angle is less than 180 degrees, and that the point of contact between the end of the optical fiber and the target stone needs to be adjusted to make the impact angle as close to 180 degrees as possible, so as to achieve a more reliable movement of the target stone.

[0051] It should be noted that the circumferential rotation warning information is used to prompt the user to adjust the point of action between the fiber end and the target stone, so that the target stone can move towards the target area under the action of the laser. In other words, the circumferential rotation warning information is used to prompt the user to adjust the fiber end to have a positive effect on the movement of the target stone towards the target area.

[0052] In one embodiment of this application, the process of determining circumferential rotation warning information based on the impact angle may include: obtaining the impact angles corresponding to two consecutive time points; if the impact angle at the later time point is less than the impact angle at the previous time point, then determining the circumferential rotation warning information and displaying the circumferential rotation warning information.

[0053] It should be noted that for the same operation targeting the same stone, if the user adjusts the point of action at the fiber optic end, the impact angles at two consecutive time points can be obtained, i.e., the impact angles after two consecutive fiber optic operations. If the impact angle at a later time is less than the impact angle at the previous time, it indicates that after the user adjusts the point of action between the fiber optic end and the target stone, the later impact angle is much less than 180 degrees compared to the previous impact angle. In this state, adjusting the fiber optic end cannot produce a positive operation to move the target stone to the target area. At this time, a circumferential rotation warning message can be determined. The circumferential rotation warning message can serve as a reverse prompt to the user for subsequent operations and can be displayed on the endoscope host's screen.

[0054] For example, if the next operation is to control the endoscope to the left, and the impact angle of the next operation is smaller than that of the previous operation, then the circumferential rotation warning information can be a prompt to the user to control the endoscope to the right. The specific circumferential rotation warning information can be set by the designer according to the actual situation.

[0055] In one embodiment of this application, the operation warning method based on image recognition may further include: determining a first line segment between the center of the target stone and the point of action; determining an extension line with the center of the target stone as the endpoint and the point of action as the extension direction, and determining the intersection point of the extension line and the boundary of the target stone; determining a second line segment based on the intersection point of the center of the target stone and the boundary; generating an over-limit warning message and displaying the over-limit warning message when the length of the first line segment is greater than the length of the second line segment.

[0056] When the length of the first segment is greater than the length of the second segment, the point of contact between the fiber end and the target stone is outside the target stone. If the fiber is operated to emit a laser at the target stone, it will be impossible to operate the target stone effectively. Therefore, an over-limit warning message can be generated to prompt the user to move the fiber end, thereby changing the point of contact between the fiber end and the target stone.

[0057] In one embodiment of this application, the operation warning method based on image recognition may further include: determining the ratio of a first line segment to a second line segment; if the ratio is not within a preset range, generating and displaying a prompt message, the prompt message being used to prompt the user to move the end of the optical fiber.

[0058] The preset range can be [0.3, 0.9]. If the ratio is within the preset range, the point of contact between the fiber optic end and the target stone can effectively impact the target stone, causing it to move to the target area under the impact of the laser. The prompt message can instruct the user to move the fiber optic cable closer to or further away from the center of the target stone.

[0059] In one embodiment of this application, the operation warning method based on image recognition may further include: acquiring current input data; the current input data includes perfusion fluid concentration, intrarenal pressure, intrarenal three-dimensional model, target stone three-dimensional model, and impact angle; inputting the current input data into a pre-trained laser prediction model to obtain predicted laser energy and predicted frequency, and determining the predicted laser energy and predicted frequency as fiber optic control parameters.

[0060] The three-dimensional models of the kidney and the target stone can be created by segmenting two-dimensional tomographic images obtained from CT scans, extracting the contours of the kidney and stone respectively, and then stacking and converting these contours into a 3D mesh. For example, step 1 involves acquiring data by exporting the complete two-dimensional tomographic image from the CT scanner or copying it from the patient's CD. Step 2 involves importing the two-dimensional tomographic image into software such as 3D Slicer, which will automatically reconstruct it into a three-dimensional volume. Step 3 involves image segmentation, where pixel regions belonging to the kidney and those belonging to the stone can be manually segmented. Step 4 generates the three-dimensional model.

[0061] It should be noted that the perfusion fluid concentration is a preset amount. The 3D models of the kidney and the target stone can be obtained through ultrasound or other means, and the intrarenal pressure can be collected using a pressure sensor. The laser prediction model can be a conventional neural network model. A training sample set can be pre-acquired, consisting of multiple training sample pairs. Each training sample pair includes a training sample and a label. The training sample can include the perfusion fluid concentration, intrarenal pressure, 3D model of the kidney, 3D model of the target stone, and impact angle. The label can include the corresponding laser energy and frequency. A suitable loss function is selected, and the laser prediction model is trained with the goal of minimizing the loss function, resulting in a pre-trained laser prediction model. The laser prediction model can output predicted laser energy and frequency. Using the laser prediction model, relatively accurate laser energy and frequency can be obtained, reducing user adjustment time.

[0062] Figure 3 This is a block diagram illustrating an image recognition-based operation warning device, as shown in an exemplary embodiment of this application. Figure 3 As shown, the exemplary image recognition-based operation warning device 300 includes:

[0063] Data acquisition module 310 is used to acquire a target image, the target image including a target stone and an optical fiber end;

[0064] The region determination module 320 is used to determine the target adjustment region based on the target image; the target adjustment region represents the desired adjustment region of the target stone;

[0065] The angle determination module 330 is used to determine the point of action between the end of the optical fiber and the target stone, and to determine the impact angle with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of action as the endpoint.

[0066] The rotation warning module 340 is used to determine circumferential rotation warning information based on the impact angle.

[0067] In one embodiment of this application, the rotation warning module includes:

[0068] Angle acquisition unit is used to acquire the impact angle corresponding to two consecutive time points;

[0069] The early warning unit is used to determine and display circumferential rotation early warning information if the impact angle at a later moment is less than the impact angle at a previous moment.

[0070] In one embodiment of this application, the image recognition-based operation warning device includes:

[0071] The first determining module is used to determine the first line segment between the center of the target stone and the point of action;

[0072] The second determining module is used to determine the extension line with the center of the target stone as the endpoint and the point of action as the extension direction, and to determine the intersection point of the extension line and the boundary of the target stone, and to determine the second line segment based on the intersection point of the center of the target stone and the boundary.

[0073] The over-limit warning module is used to generate and display over-limit warning information when the length of the first line segment is greater than the length of the second line segment.

[0074] In one embodiment of this application, the image recognition-based operation warning device includes:

[0075] The ratio determination module is used to determine the ratio between the first line segment and the second line segment;

[0076] The prompt message determination module is used to generate and display a prompt message if the ratio is not within a preset range. The prompt message is used to prompt the user to move the end of the optical fiber.

[0077] In one embodiment of this application, the image recognition-based operation warning device includes:

[0078] The input data acquisition module is used to acquire the current input data, which includes the perfusion fluid concentration, intrarenal pressure, intrarenal 3D model, target stone 3D model, and impact angle.

[0079] The parameter prediction module is used to input the current input data into a pre-trained laser prediction model to obtain the predicted laser energy and frequency, and then determine the predicted laser energy and frequency as fiber control parameters.

[0080] It should be noted that the image recognition-based operation warning device and the image recognition-based operation warning method provided in the above embodiments belong to the same concept. The specific methods by which each module and unit performs operations have been described in detail in the method embodiments and will not be repeated here. In practical applications, the image recognition-based operation warning device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0081] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, enable the electronic device to implement the image recognition-based operation warning method provided in the above embodiments.

[0082] Another aspect of this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the image recognition-based operation warning method provided in the various embodiments described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not be assembled into the electronic device.

[0083] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the image recognition-based operation warning method provided in the various embodiments described above.

[0084] Another aspect of this application provides an endoscope, which includes an endoscope host, the endoscope host including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the image recognition-based operation warning method provided in the above embodiments.

[0085] In the embodiments of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The terms "comprising" and "including" as used throughout the specification and claims are open-ended terms and should therefore be interpreted as "comprising but not limited to".

[0086] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. An operation early warning device based on image recognition, characterized in that, include: The data acquisition module is used to acquire a target image, which includes a target stone and an optical fiber end; The region determination module is used to determine the target adjustment region based on the target image; The target adjustment region represents the desired adjustment region of the target stone; The angle determination module is used to determine the point of impact between the end of the optical fiber and the target stone, and to determine the impact angle with the center of the target stone as the vertex, the center of the target adjustment area as the endpoint, and the point of impact as the endpoint. The rotation warning module is used to determine circumferential rotation warning information based on the impact angle; The circumferential rotation warning information is used to prompt users to adjust the fiber optic end to positively influence the movement of the target stone toward the target area.

2. The operation early warning device based on image recognition according to claim 1, characterized in that, The rotation warning module includes: Angle acquisition unit is used to acquire the impact angle corresponding to two consecutive time points; The early warning unit is used to determine and display circumferential rotation early warning information if the impact angle at a later moment is less than the impact angle at a previous moment.

3. The operation early warning device based on image recognition according to claim 1, characterized in that, The image recognition-based operation early warning device includes: The first determining module is used to determine the first line segment between the center of the target stone and the point of action; The second determining module is used to determine the extension line with the center of the target stone as the endpoint and the point of action as the extension direction, and to determine the intersection point of the extension line and the boundary of the target stone, and to determine the second line segment based on the intersection point of the center of the target stone and the boundary. The over-limit warning module is used to generate and display over-limit warning information when the length of the first line segment is greater than the length of the second line segment.

4. The operation early warning device based on image recognition according to claim 3, characterized in that, The image recognition-based operation early warning device includes: The ratio determination module is used to determine the ratio between the first line segment and the second line segment; The prompt message determination module is used to generate and display a prompt message if the ratio is not within a preset range. The prompt message is used to prompt the user to move the end of the optical fiber.

5. The operation early warning device based on image recognition according to claim 1, characterized in that, The image recognition-based operation early warning device includes: The input data acquisition module is used to acquire the current input data, which includes the perfusion fluid concentration, intrarenal pressure, intrarenal 3D model, target stone 3D model, and impact angle. The parameter prediction module is used to input the current input data into a pre-trained laser prediction model to obtain the predicted laser energy and frequency, and then determine the predicted laser energy and frequency as fiber control parameters.

6. An endoscope, characterized in that, Includes an endoscope host, the endoscope host comprising: One or more processors; Memory used to store the executable program code of the processor; The processor is configured to execute the program code to implement the steps of the image recognition-based operation warning device as described in any one of claims 1 to 5.

Citation Information

Patent Citations

  • Surgical laser system with controlled retropulsion setting

    WO2025147452A1

  • Flexible bendable electronic nephroscope for percutaneous nephroseopy surgery and application method of flexible bendable electronic nephroscope

    CN104873167A

  • Shockwave plasty catheter

    CN119235411A