An ultrasonic knife position recognition method and device based on image processing

Abstract

The application provides an ultrasonic knife position recognition method and device based on image processing, which is suitable for the field of image processing technology, and the method comprises the following steps: performing correction transformation processing according to to-be-recognized intraoperative image information to generate corrected intraoperative image information; obtaining ultrasonic knife body pixel point information according to the corrected intraoperative image information; generating ultrasonic knife body position range information according to the ultrasonic knife body pixel point information, to-be-recognized image acquisition time information and a preset ultrasonic knife body recognition pixel position offset correction coefficient; and generating ultrasonic knife position recognition information according to the to-be-recognized intraoperative image information, the to-be-recognized image acquisition time information, the ultrasonic knife body pixel point information, the ultrasonic knife body position range information and an ultrasonic knife position recognition model. The application improves the accuracy and operation real-time performance of the intraoperative position recognition result of the ultrasonic knife, and meets the high-precision and high-real-time instrument positioning use requirements of the intraoperative image navigation system.

Description

Technical Field

[0001] This application belongs to the field of image processing technology, and in particular relates to an ultrasonic scalpel position recognition method and apparatus based on image processing. Background Technology

[0002] With the rapid iteration of minimally invasive surgery and intraoperative image navigation technology, spatial positioning of surgical instruments based on image processing has become a key supporting technology for precision surgery. As an energy-based surgical instrument frequently used in laparoscopic and open surgery, the ultrasonic scalpel has become a research hotspot in the field of precision medicine at home and abroad by relying on the technical route of automated position recognition based on intraoperative optical imaging and ultrasound imaging.

[0003] In existing technologies, ultrasonic scalpel feature tracking and localization schemes based on multiple consecutive intraoperative images typically output ultrasonic scalpel localization results by constructing a linear computation model with fixed empirical coefficients.

[0004] However, existing technologies cannot specifically remove irregular speckles and noise caused by bleeding in the surgical area and tissue obstruction, and invalid stray contours are easily left after contour extraction. Summary of the Invention

[0005] In view of this, the present application provides an ultrasonic scalpel position recognition method and apparatus based on image processing, which aims to solve the problems of weak anti-interference ability and difficulty in improving the position recognition accuracy of ultrasonic scalpels in the prior art.

[0006] The first aspect of this application provides an ultrasonic scalpel location recognition method based on image processing, including:

[0007] Acquire multiple intraoperative image information to be identified and multiple image acquisition time information to be identified; the intraoperative image information to be identified and the image acquisition time information to be identified correspond one-to-one.

[0008] Based on the multiple intraoperative image information to be identified, a correction and transformation process is performed to generate multiple corrected intraoperative image information;

[0009] Based on the preset ultrasonic scalpel blade pixel recognition model, multiple ultrasonic scalpel blade pixel information are calculated according to the multiple corrected intraoperative image information.

[0010] Based on the pixel information of the multiple ultrasonic scalpel bodies, the acquisition time information of multiple images to be identified, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient, multiple ultrasonic scalpel body position range information are generated.

[0011] Based on the multiple intraoperative image information to be identified, the multiple image acquisition time information to be identified, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model, multiple ultrasonic scalpel position recognition information are generated.

[0012] A second aspect of this application provides an image processing-based ultrasonic scalpel position recognition device, comprising:

[0013] The module for acquiring intraoperative image information and image acquisition time information to be identified is used to acquire multiple intraoperative image information and multiple image acquisition time information to be identified; the intraoperative image information and the image acquisition time information to be identified correspond one-to-one.

[0014] The modified intraoperative image information generation module is used to perform correction and transformation processing on the multiple intraoperative image information to be identified, and generate multiple modified intraoperative image information.

[0015] The ultrasonic scalpel body pixel information generation module is used to calculate multiple ultrasonic scalpel body pixel information based on a preset ultrasonic scalpel body pixel recognition model and the multiple corrected intraoperative image information.

[0016] The ultrasonic scalpel body position range information generation module is used to generate multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel information, multiple image acquisition time information to be identified, and a preset ultrasonic scalpel body identification pixel position offset correction coefficient.

[0017] The ultrasonic scalpel position recognition information generation module is used to generate multiple ultrasonic scalpel position recognition information based on the multiple intraoperative image information to be recognized, the multiple image acquisition time information to be recognized, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model.

[0018] A third aspect of this application provides a terminal device, the terminal device including a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the steps of the image processing-based ultrasonic scalpel position recognition method described in the first aspect above.

[0019] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, wherein when executed by a processor, the computer program implements the steps of the image processing-based ultrasonic scalpel position recognition method described in the first aspect above.

[0020] Compared with the prior art, the beneficial effects of this application are: this application is adapted to complex clinical scenarios such as tissue occlusion, light and shadow changes, and dynamic extension and retraction of the blade during surgery, thereby improving the accuracy and real-time performance of the ultrasonic scalpel position recognition results during surgery. The recognition results can be used to match various surgical conditions in laparoscopy and open surgery, meeting the high-precision and high-real-time instrument positioning requirements of intraoperative image navigation systems. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 1 of this application;

[0023] Figure 2 This is a schematic diagram of the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 2 of this application;

[0024] Figure 3 This is a schematic diagram of the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 3 of this application;

[0025] Figure 4 This is a schematic diagram illustrating the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 4 of this application;

[0026] Figure 5 This is a schematic diagram illustrating the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 5 of this application;

[0027] Figure 6 This is a schematic diagram illustrating the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment Six of this application;

[0028] Figure 7 This is a schematic diagram of the implementation process of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 7 of this application;

[0029] Figure 8 This is a schematic diagram of the ultrasonic scalpel position recognition device based on image processing provided in an embodiment of this application;

[0030] Figure 9 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0031] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0032] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0033] Figure 1 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 1 of this application is shown below in detail:

[0034] Step S101: Obtain multiple intraoperative image information to be identified and multiple image acquisition time information to be identified; the intraoperative image information to be identified and the image acquisition time information to be identified correspond one-to-one.

[0035] In this embodiment, the multiple intraoperative image information to be identified can be surgical area image data continuously acquired by an intraoperative image acquisition device during the entire ultrasonic scalpel surgical operation, which is used to reflect the placement posture and contour features of the ultrasonic scalpel body at different times. It can be acquired at a frame rate of not less than 30 frames per second, and the image resolution can be not less than 1280×720 pixels. The acquisition time information of the multiple images to be identified can be timestamp data recorded synchronously with each frame of intraoperative image information to be identified, which is used to characterize the order of image acquisition and the time interval between frames, and can be obtained in real time by the surgical supporting industrial control computing unit.

[0036] In this embodiment, specifically, when the laparoscopic surgical instruments are inserted into the cavity and the ultrasonic scalpel is powered on and started, the surgical area image acquisition device is turned on simultaneously. The device continuously captures in-situ images of the surgical area as the ultrasonic scalpel extends, swings, and cuts throughout the entire surgical process, and retains the intraoperative image information to be identified throughout the entire surgical operation cycle without interruption.

[0037] Step S102: Perform correction and transformation processing on the multiple intraoperative image information to be identified to generate multiple corrected intraoperative image information.

[0038] In this embodiment, single-channel grayscale mapping transformation and adaptive noise reduction and smoothing operations are first performed sequentially on multiple intraoperative image information to be identified to eliminate image interference caused by surgical area bleeding reflection, human tissue occlusion and noise from imaging hardware. Then, a contour boundary extraction algorithm is used to strip the edge features of the ultrasonic scalpel instrument. Then, invalid edge removal and instrument target area completion processing are performed on all extracted contours to generate multiple corrected intraoperative image information with clean images, no redundant interference pixels, and complete retention of the ultrasonic scalpel body contour.

[0039] Step S103: Based on the preset ultrasonic scalpel body pixel recognition model, multiple ultrasonic scalpel body pixel information is calculated according to the multiple corrected intraoperative image information.

[0040] In this embodiment, the preset ultrasonic scalpel body pixel recognition model can be preset by humans. This model can adopt the YOLOv8 deep learning object detection model. A special coordinate system for intraoperative imaging is built with the center point of the surgical cannula as the coordinate origin. All instrument feature pixels in each corrected intraoperative image are traversed to locate the extreme pixel coordinates of the distal end of the ultrasonic scalpel head and the proximal end of the handle. Then, numerical conversion is carried out according to the real scene size and pixel conversion standard completed in the preoperative calibration. The coordinate conversion results are then statistically stored frame by frame to generate multiple ultrasonic scalpel body pixel information corresponding to a single frame.

[0041] Step S104: Generate multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel information, the multiple image acquisition time information to be identified, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient.

[0042] In this embodiment, the preset ultrasonic scalpel blade identification pixel position offset correction coefficient can be preset manually, and the coefficient is fixed at 0.08. The difference calculation is performed on the ultrasonic scalpel blade pixel information corresponding to adjacent acquisition frames to determine whether the change amplitude of the blade pixel position is greater than or equal to the product of the preset ultrasonic scalpel blade identification pixel position offset correction coefficient and the blade axial span pixel value of the previous frame. Then, the time point corresponding to the critical offset of the blade tip is locked by combining the point change trend of continuous multiple frames, and the relevant data of the blade tip extreme positioning time is obtained. Then, the maximum value of the blade axial span in all frames is extracted, and the acquisition time corresponding to the maximum span and the blade tip extreme positioning time are combined to calculate the complete surgical operation time, thereby generating multiple ultrasonic scalpel blade position range information including the critical point and the entire operation range.

[0043] Step S105: Generate multiple ultrasonic scalpel position recognition information based on the multiple intraoperative image information to be identified, the multiple image acquisition time information to be identified, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model.

[0044] In this embodiment, the preset ultrasonic scalpel position recognition model can be pre-set manually. This model can adopt a ResNet multi-feature fusion deep learning model. First, based on the acquisition time information of multiple images to be recognized and the pixel information of multiple ultrasonic scalpels, a fitting curve information of the temporal change of the scalpel position is constructed. Then, based on the fitting curve, the initial feed rate information of the scalpel head is calculated. The peak position parameters of the scalpel extension and the critical positioning position parameters are extracted from the curve data. Then, the feed rate, position span, operation time and other parameters are normalized and converted according to the pre-reserved benchmark reference data. Nonlinear feature transformation is carried out on each normalized parameter and dynamic interactive feature terms are constructed. All processed feature data are input into the preset ultrasonic scalpel position recognition model for fusion solution, thereby generating multiple ultrasonic scalpel position recognition information that can accurately represent the intraoperative spatial coordinates.

[0045] The ultrasonic scalpel position recognition method based on image processing provided in this application is adapted to complex clinical scenarios such as intraoperative tissue occlusion, sudden changes in light and shadow, and dynamic extension and retraction of the scalpel head, thereby improving the accuracy and real-time performance of the ultrasonic scalpel position recognition results during surgery. The recognition results can be used to match various surgical conditions in laparoscopic and open surgery, meeting the high-precision and high-real-time instrument positioning requirements of intraoperative image navigation systems.

[0046] Figure 2 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 2 of this application is shown. The difference between this method and Embodiment 1 is that step S102 specifically includes:

[0047] Step S201: Perform grayscale mapping processing on the multiple intraoperative image information to be identified to generate multiple intraoperative grayscale image information to be identified.

[0048] In this embodiment, it can be understood that the multiple intraoperative image information to be identified are RGB color images. The multiple intraoperative image information to be identified can be preprocessed to weaken the strong light reflection and image noise interference during the operation. The weighted average method can be used to perform grayscale mapping conversion on the multiple intraoperative image information to be identified, eliminate redundant image information brought by the color channel, and then unify the brightness dimension and pixel dimension of all intraoperative images, thereby generating multiple intraoperative grayscale image information to be identified.

[0049] Step S202: Filter the multiple intraoperative grayscale image information to be identified to generate multiple filtered intraoperative image information.

[0050] In this embodiment, a Gaussian kernel can be used to perform filtering and noise reduction operations on multiple intraoperative grayscale image information to be identified, eliminating dynamic light and shadow interference, tissue speckle interference and inherent noise of the image acquisition equipment in the surgical area, thereby completely preserving the edge contour details of the ultrasonic scalpel body, and generating multiple filtered intraoperative image information.

[0051] In this embodiment, specifically, a 5×5 Gaussian kernel can be used to perform global filtering on the grayscale image information to be identified during the operation, in order to remove scattered noise caused by the complex environment during the operation. The standard deviation of the Gaussian kernel is fixed at 1.0, which effectively balances the image noise reduction effect and the accuracy of contour detail preservation.

[0052] Step S203: Based on the multiple filtered intraoperative image information, perform contour extraction and filling processing to generate multiple corrected intraoperative image information.

[0053] In this embodiment, the Canny edge detection algorithm can be used to calculate the image gradient of multiple filtered intraoperative image information, and then non-maximum suppression and double threshold screening operations can be performed to refine the edge contour of the ultrasonic scalpel to a single pixel width. Then, scattered invalid edges are removed and the complete instrument contour is retained. Finally, the effective contour area of ​​the ultrasonic scalpel is filled with pixels to generate multiple corrected intraoperative image information with complete contours, interference removal, and clean images.

[0054] In this embodiment, the Sobel operator can be used to calculate the horizontal and vertical gradient information of the image to complete the initial extraction of the image edges. Then, non-maximum suppression is used to refine the edge contours. Next, a fixed high threshold of 80 and a fixed low threshold of 30 are set to complete the dual threshold detection, retaining effective strong edges and associated weak edges, and removing isolated noise edges. Finally, the contour circumscribed rectangle filtering rule is used to filter out non-instrument interference contours, and the effective ultrasonic scalpel contour is filled to obtain the complete instrument grayscale area. The filled image is the corrected intraoperative image information.

[0055] In this embodiment, the preset contour filtering size parameters can be preset manually. The height threshold of the contour circumscribed rectangle is fixed at 40 pixels, the width threshold of the contour circumscribed rectangle is fixed at 15 pixels, and the center offset threshold of the contour is fixed at ±80 pixels. Only the effective contour of the ultrasonic scalpel that meets the size and position conditions is retained, thereby completing the contour filling process.

[0056] The ultrasonic scalpel position recognition method based on image processing provided in this application effectively eliminates image noise and interference contours generated by complex intraoperative environments, significantly improves the integrity and accuracy of ultrasonic scalpel body contour extraction, thereby reducing recognition errors caused by image preprocessing and enhancing adaptability and anti-interference ability to various intraoperative surgical scenarios.

[0057] Figure 3The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 3 of this application is shown. The difference between this method and Embodiment 1 is that step S103 specifically includes:

[0058] Step S301: Based on the preset intraoperative image spatial coordinate system, pixel coordinate extraction processing is performed on the multiple corrected intraoperative image information to generate multiple corrected intraoperative image pixel coordinate information.

[0059] In this embodiment, the preset intraoperative image spatial coordinate system can be preset by humans. A standardized intraoperative imaging coordinate system is established with the center of the surgical cannula as the coordinate origin, the vertical downward direction as the positive direction of the longitudinal axis, and the horizontal rightward direction as the positive direction of the transverse axis. Then, all pixel points in each corrected intraoperative image information are traversed frame by frame. Then, the transverse and longitudinal axis coordinate data corresponding to all pixels are completely collected and classified and recorded, thereby generating multiple corrected intraoperative image pixel coordinate information.

[0060] Step S302: Based on the preset ultrasonic scalpel body pixel recognition model, multiple ultrasonic scalpel body pixel information is calculated according to the multiple corrected intraoperative image information and the multiple corrected intraoperative image pixel coordinate information.

[0061] In this embodiment, the preset ultrasonic scalpel body pixel recognition model can be manually preset, and the model can adopt the YOLOv8 deep learning object detection model. Based on the preset intraoperative image spatial coordinate system and multiple corrected intraoperative image pixel coordinate information, the extreme pixel coordinate points of the farthest end of the ultrasonic scalpel head and the fixed end of the handle are locked. Then, the pixel and actual spatial size conversion coefficients completed in the preoperative calibration are called to complete the conversion operation from pixel coordinates to actual spatial coordinates frame by frame. Then, the key pixel point data of the scalpel body corresponding to each frame are uniformly recorded, thereby generating multiple ultrasonic scalpel body pixel point information.

[0062] In this embodiment, the preset pixel size conversion factor can be preset manually, with a fixed value of 1 pixel corresponding to 0.05 mm, which can adapt to the small imaging scale of the surgical area in minimally invasive surgery and ensure the accuracy of coordinate conversion.

[0063] The ultrasonic scalpel position recognition method based on image processing provided in this application effectively reduces the pixel positioning deviation of the ultrasonic scalpel body, thereby improving the stability and accuracy of the acquisition of scalpel body pixel information, and providing high-precision basic data for subsequent scalpel body position range calculation and instrument position recognition.

[0064] Figure 4 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 4 of this application is shown. The difference between this method and Embodiment 1 is that step S104 specifically includes:

[0065] Step S401: Based on the pixel information of the multiple ultrasonic scalpel bodies and the acquisition time information of the multiple images to be identified, the displacement information of the multiple ultrasonic scalpel bodies is calculated.

[0066] In this embodiment, the acquisition timing of adjacent image frames is determined by the acquisition time information of multiple images to be identified. The pixel information of multiple ultrasonic scalpels corresponding to the previous and next frames is retrieved frame by frame to perform difference calculation. The pixel offset value of the ultrasonic scalpel in a single frame time interval is counted. Then, the inter-frame displacement data is recorded group by group to generate multiple ultrasonic scalpel pixel displacement information.

[0067] Step S402: Based on the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body pixel displacement information, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient, calculate the position information of the multiple ultrasonic scalpel body pixels after displacement.

[0068] In this embodiment, the preset ultrasonic scalpel blade identification pixel position offset correction coefficient can be preset manually, and the coefficient is fixed at 0.08. The displacement information of multiple ultrasonic scalpel blade pixels is weighted and corrected with the preset ultrasonic scalpel blade identification pixel position offset correction coefficient, and then combined with the ultrasonic scalpel blade pixel information of the previous frame to complete the compensation correction of the blade displacement deviation in the current frame. Then, the corrected blade pixel point data is calculated frame by frame to generate multiple ultrasonic scalpel blade pixel displacement position information.

[0069] Step S403: Based on the position information of the multiple ultrasonic scalpel body pixels after displacement and the acquisition time information of the multiple images to be identified, obtain the time information of the multiple ultrasonic scalpel body pixels after displacement.

[0070] In this embodiment, the position information of multiple ultrasonic scalpel blade pixels after displacement is matched frame by frame with the image acquisition time information of the corresponding frame to be identified. The point position after blade correction is bound with the acquisition time sequence data, and then the precise time node corresponding to each displacement point is recorded, thereby generating multiple ultrasonic scalpel blade pixel displacement time information.

[0071] Step S404: Based on the position information of the multiple ultrasonic scalpel body pixels after displacement and the time information of the multiple ultrasonic scalpel body pixels after displacement, calculate the ultrasonic scalpel body pixel displacement imaging time information.

[0072] In this embodiment, multiple sets of ultrasonic scalpel body pixel displacement position information and corresponding ultrasonic scalpel body pixel displacement time information are integrated. The start time, critical offset time and displacement termination time of the continuous displacement of the scalpel body are statistically analyzed, and then the complete displacement time series of the scalpel body is sorted out, thereby generating ultrasonic scalpel body pixel displacement imaging time information.

[0073] Step S405: Generate multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel displacement information, the multiple ultrasonic scalpel body pixel displacement position information, the multiple ultrasonic scalpel body pixel displacement time information, and the ultrasonic scalpel body pixel displacement imaging time information.

[0074] In this embodiment, the pixel displacement information of multiple ultrasonic scalpel bodies corresponding to all time series is traversed, and the maximum offset and critical offset data of the scalpel body are filtered out. Then, the spatial range of the scalpel body operation is locked by combining the position information after the pixel displacement of multiple ultrasonic scalpel bodies. Then, the effective operating time series range of the scalpel body is determined based on the imaging time information of the ultrasonic scalpel body pixel displacement. The spatial position and time series data are integrated to complete the interval calibration, thereby generating multiple ultrasonic scalpel body position range information that includes the dynamic operating space and time series interval of the ultrasonic scalpel.

[0075] The ultrasonic scalpel position recognition method based on image processing provided in this application accurately locks the dynamic displacement range and operation time interval of the ultrasonic scalpel during surgery, effectively improving the accuracy and timing matching degree of ultrasonic scalpel position range determination, thereby optimizing the overall accuracy and stability of intraoperative instrument dynamic positioning.

[0076] Figure 5 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 5 of this application is shown. The difference between this method and Embodiment 1 above is that step S105 specifically includes:

[0077] Step S501: Based on the acquisition time information of the multiple images to be identified and the pixel information of the multiple ultrasonic scalpel bodies, generate a time-series curve of the position change of the ultrasonic scalpel body pixels.

[0078] In this embodiment, discrete data points can be plotted with the acquisition time information of multiple images to be identified as the horizontal axis and the spatial position corresponding to the pixel information of multiple ultrasonic scalpels as the vertical axis. Then, a smooth interpolation method is used to optimize all the scattered data as a whole, and then a continuous and smooth change curve is obtained by fitting point by point, thereby generating a time series curve of the change in the position of the ultrasonic scalpel pixel.

[0079] In this embodiment, the image acquisition time can be used as the horizontal axis and the axial extension pixel position of the ultrasonic scalpel body can be used as the vertical axis. The time data and scalpel body pixel position data corresponding to each frame can be plotted as a scatter plot. Cubic spline interpolation can be used to smooth the discrete scatter plots, eliminate curve spikes caused by single-frame pixel jitter, and generate a complete time series curve of ultrasonic scalpel body pixel position change that characterizes the dynamic relationship between the ultrasonic scalpel body position and the operation time.

[0080] Step S502: Based on the acquisition time information of the multiple images to be identified and the time-series curve of the pixel position change of the ultrasonic scalpel body, the initial moving speed information of the ultrasonic scalpel body is calculated.

[0081] In this embodiment, the initial operation stage data of the ultrasonic scalpel body pixel position change time series curve can be selected for linear fitting processing. Then, the slope value of the fitted curve is used as the movement rate of the ultrasonic scalpel in the early stage of operation, thereby generating the ultrasonic scalpel body initial movement rate information, which is used to represent the linear fitting rate of the initial segment of the ultrasonic scalpel body pixel position change time series curve within 0.5 seconds before the ultrasonic scalpel starts operation. If a large displacement of the scalpel body position occurs within 0.5 seconds, the linear fitting slope of the curve during the time period from instrument start to the first position displacement is taken out, with the unit being millimeters per second.

[0082] Step S503: Based on the time-series curve of the position change of the ultrasonic scalpel body pixels, extract the spatial extreme values ​​of the multiple ultrasonic scalpel body pixel information to obtain the pixel information of the multiple ultrasonic scalpel body extension positions.

[0083] In this embodiment, all ultrasonic scalpel pixel information corresponding to the time-series curve of the change in the position of ultrasonic scalpel pixel can be traversed and numerical comparison and screening can be carried out to screen out the maximum extension space pixel position during the entire ultrasonic scalpel operation, thereby generating multiple ultrasonic scalpel extension position pixel information to characterize the limit space pixel position of the blade extension during the entire ultrasonic scalpel surgical operation.

[0084] Step S504: Generate multiple ultrasonic scalpel position recognition information based on the initial moving speed information of the ultrasonic scalpel body, multiple ultrasonic scalpel body extension position pixel information, multiple ultrasonic scalpel body pixel information, multiple ultrasonic scalpel body position range information, and a preset ultrasonic scalpel position recognition model.

[0085] In this embodiment, the preset ultrasonic scalpel position recognition model can be pre-set manually. This model can adopt a ResNet multi-feature fusion deep learning model. First, dimensionless and nonlinear transformation processing is performed on the initial movement rate information of the ultrasonic scalpel body, the pixel information of multiple ultrasonic scalpel body extension positions, the pixel information of multiple ultrasonic scalpel body, and the position range information of multiple ultrasonic scalpel body. Then, multi-dimensional interactive feature terms are constructed to characterize the dynamic displacement characteristics and spatial deformation characteristics of the ultrasonic scalpel. Then, the processed feature information is input into the preset ultrasonic scalpel position recognition model to complete the multi-parameter fusion solution calculation, thereby generating multiple ultrasonic scalpel position recognition information that accurately characterize the real-time spatial coordinates during ultrasonic scalpel surgery.

[0086] The ultrasonic scalpel position recognition method based on image processing provided in this application improves the extraction accuracy of the temporal position features and spatial extreme value features of the ultrasonic scalpel, and enhances the ability to represent the dynamic movement and deformation behavior of the ultrasonic scalpel during surgery. As a result, the ultrasonic scalpel position recognition results are more consistent with the actual working state during surgery, and effectively improve the accuracy and timeliness of instrument positioning.

[0087] Figure 6 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment Six of this application is shown. The difference between this method and Embodiment Five is that step S504 specifically includes:

[0088] Step S601: Based on the initial moving rate information of the ultrasonic scalpel body and the preset reference moving rate information of the ultrasonic scalpel body, the normalized information of the initial moving rate of the ultrasonic scalpel body is calculated.

[0089] In this embodiment, the preset ultrasonic scalpel body reference movement rate information can be preset by humans. The reference movement rate is fixed at 1 mm per second. The initial movement rate information of the ultrasonic scalpel body is compared with the preset ultrasonic scalpel body reference movement rate information to complete the dimensionless correction processing of the initial movement rate data, thereby generating the normalized information of the initial movement rate of the ultrasonic scalpel body.

[0090] Step S602: Based on the pixel information of the multiple ultrasonic scalpel body extension positions, the pixel information of the multiple ultrasonic scalpel bodies, the position range information of the multiple ultrasonic scalpel bodies, and the preset intraoperative wound calibration aperture parameter information, calculate the normalized information of the positions of the multiple ultrasonic scalpel body extension pixels, the normalized information of the positions of the multiple ultrasonic scalpel body pixels, and the normalized information of the position boundaries of the multiple ultrasonic scalpel bodies.

[0091] In this embodiment, the preset intraoperative wound calibration aperture parameter information can be preset manually. The intraoperative wound calibration aperture is fixed at 10 mm. The pixel information of multiple ultrasonic scalpel body extension positions, multiple ultrasonic scalpel body pixel information, and multiple ultrasonic scalpel body position range information are compared with the preset intraoperative wound calibration aperture parameter information to complete the dimensionless unified processing of various spatial position parameters, thereby generating multiple ultrasonic scalpel body extension pixel position normalization information, multiple ultrasonic scalpel body pixel position normalization information, and multiple ultrasonic scalpel body position boundary normalization information.

[0092] Step S603: Based on the normalized information of the initial moving speed of the ultrasonic scalpel body, the normalized information of the positions of multiple ultrasonic scalpel body extended pixels, the normalized information of the positions of multiple ultrasonic scalpel body pixels, the normalized information of the positions of multiple ultrasonic scalpel body boundaries, and the preset ultrasonic scalpel position recognition model, generate multiple ultrasonic scalpel position recognition information.

[0093] In this embodiment, the preset ultrasonic scalpel position recognition model can be preset by humans. The model can adopt the ResNet multi-feature fusion deep learning model, which inputs all normalized rate features and spatial position features into the preset ultrasonic scalpel position recognition model, thereby completing multi-dimensional parameter fusion and regression calculation, and generating multiple ultrasonic scalpel position recognition information that accurately represents the real-time spatial positioning status during ultrasonic scalpel surgery.

[0094] The ultrasonic scalpel position recognition method based on image processing provided in this application eliminates the recognition calculation deviation caused by the difference in the dimensions of parameters of different dimensions, unifies the standard of model input data, thereby improving the stability of ultrasonic scalpel position recognition model operation, and thus improving the accuracy and reliability of ultrasonic scalpel position recognition results during surgery.

[0095] Figure 7 The flowchart illustrating the implementation of the ultrasonic scalpel position recognition method based on image processing provided in Embodiment 7 of this application is shown. The difference between this method and Embodiment 6 above is that step S603 specifically includes:

[0096] Step S701: Based on the normalized information of the initial moving rate of the ultrasonic scalpel body, the normalized information of the positions of multiple extended pixels of the ultrasonic scalpel body, the normalized information of the positions of multiple pixels of the ultrasonic scalpel body, and the normalized information of the position boundaries of the ultrasonic scalpel body, nonlinear feature mapping processing is performed to generate the initial moving rate feature information of the ultrasonic scalpel body, the position feature information of multiple extended pixels of the ultrasonic scalpel body, the position feature information of multiple pixels of the ultrasonic scalpel body, and the position boundary feature information of multiple ultrasonic scalpel bodies.

[0097] In this embodiment, the initial movement rate of the ultrasonic scalpel can be normalized by performing a nonlinear exponential transformation on the rate dimension to compensate for the impact of changes in the initial movement rate of the ultrasonic scalpel on positioning accuracy, thereby generating initial movement rate feature information of the ultrasonic scalpel. Logarithmic transformation can be performed on the normalized position information of multiple extended pixel points of the ultrasonic scalpel to compress the dynamic range of the scalpel's spatial dimension and extract the structured features of the instrument's extension, thereby generating position feature information of multiple extended pixel points of the ultrasonic scalpel. Low-order nonlinear transformation can be performed on the normalized position information of multiple ultrasonic scalpel pixel points to refine the minute offset features of single-frame pixel positions, thereby generating position feature information of multiple ultrasonic scalpel pixel points. High-order nonlinear transformation can be performed on the normalized position boundary information of multiple ultrasonic scalpels to enhance the feature discrimination of the scalpel's working boundary, thereby generating position boundary feature information of multiple ultrasonic scalpels.

[0098] Step S702: Generate multiple ultrasonic scalpel position recognition information based on the initial moving rate feature information of the ultrasonic scalpel body, the position feature information of multiple ultrasonic scalpel body extended pixels, the position feature information of multiple ultrasonic scalpel body pixels, the position boundary feature information of multiple ultrasonic scalpel body, and the preset ultrasonic scalpel position recognition model.

[0099] In this embodiment, the preset ultrasonic scalpel position recognition model can be preset by humans. This model can adopt the ResNet multi-feature fusion deep learning model to integrate and summarize all the rate features, spatial position features and boundary features that have been processed by nonlinear mapping, and then complete the multi-dimensional deep feature fusion and weighted fitting calculation. It fully explores the complex nonlinear correlation features of dynamic displacement, spatial extension and boundary offset during ultrasonic scalpel surgery, and then outputs accurate positioning data through model regression calculation, thereby generating multiple ultrasonic scalpel position recognition information.

[0100] The ultrasonic scalpel position recognition method based on image processing provided in this application fully explores the nonlinear dynamic characteristics of ultrasonic scalpel operations, effectively compensates for the shortcomings of traditional linear feature fitting in adapting to the dynamic deformation and variable speed movement of the machine, greatly improves the model's feature analysis capability for complex intraoperative conditions, and significantly enhances the accuracy, robustness and scene adaptability of ultrasonic scalpel position recognition during surgery.

[0101] Corresponding to the method in the above embodiments, Figure 8 The diagram shows a structural block diagram of an ultrasonic scalpel position recognition device based on image processing provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiments of this application are shown. Figure 8 The example image processing-based ultrasonic scalpel position recognition device can be the execution subject of the image processing-based ultrasonic scalpel position recognition method provided in the aforementioned embodiment 1.

[0102] Reference Figure 8 The image processing-based ultrasonic scalpel position recognition device includes:

[0103] The module 810 for acquiring intraoperative image information and image acquisition time information to be identified is used to acquire multiple intraoperative image information and multiple image acquisition time information to be identified; the intraoperative image information and the image acquisition time information to be identified correspond one-to-one.

[0104] The modified intraoperative image information generation module 820 is used to perform correction and transformation processing on the multiple intraoperative image information to be identified, and generate multiple modified intraoperative image information.

[0105] The ultrasonic scalpel body pixel information generation module 830 is used to calculate multiple ultrasonic scalpel body pixel information based on a preset ultrasonic scalpel body pixel recognition model and the multiple corrected intraoperative image information.

[0106] The ultrasonic scalpel body position range information generation module 840 is used to generate multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel information, multiple image acquisition time information to be identified, and a preset ultrasonic scalpel body identification pixel position offset correction coefficient.

[0107] The ultrasonic scalpel position recognition information generation module 850 is used to generate multiple ultrasonic scalpel position recognition information based on the multiple intraoperative image information to be recognized, the multiple image acquisition time information to be recognized, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model.

[0108] The process by which each module in the image processing-based ultrasonic scalpel position recognition device provided in this application implements its respective function can be specifically referred to the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0109] It should be understood that the sequence number of each step in the above embodiments does not imply 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 this application.

[0110] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0111] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0112] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0113] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0114] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0115] The ultrasonic scalpel position recognition method based on image processing provided in this application can be applied to terminal devices such as mobile phones, tablets, wearable devices, in-vehicle devices, augmented reality / virtual reality devices, laptops, super mobile personal computers, netbooks, and personal digital assistants. This application does not impose any restrictions on the specific type of terminal device.

[0116] For example, the terminal device may be a station in a WLAN, a cellular phone, a cordless phone, a session initiation protocol phone, a wireless local loop station, a personal digital processing device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a set-top box, a user premises equipment, and / or other devices for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved public terrestrial mobile networks, etc.

[0117] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 9 As shown, the terminal device 9 of this embodiment includes: at least one processor 90 ( Figure 9Only one is shown in the image (i.e., a memory 91 is provided), and the memory 91 stores a computer program 92 that can run on the processor 90. When the processor 90 executes the computer program 92, it implements the steps in the above-described embodiments of the image processing-based ultrasonic scalpel position recognition method, for example... Figure 1 Steps S101 to S105 are shown. Alternatively, when the processor 90 executes the computer program 92, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 8 The functions of modules 810 to 850 are shown.

[0118] The terminal device 9 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 90 and a memory 91. Those skilled in the art will understand that... Figure 9 This is merely an example of terminal device 9 and does not constitute a limitation on terminal device 9. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmission devices, network access devices, buses, etc.

[0119] The processor 90 may be a central processing unit, or it may be other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0120] In some embodiments, the memory 91 may be an internal storage unit of the terminal device 9, such as a hard disk or memory of the terminal device 9. The memory 91 may also be an external storage device of the terminal device 9, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc., equipped on the terminal device 9. Furthermore, the memory 91 may include both internal and external storage units of the terminal device 9. The memory 91 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of computer programs. The memory 91 can also be used to temporarily store data that has been sent or will be sent.

[0121] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0122] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0123] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0124] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0125] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0126] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0127] Those skilled in the art will recognize that the units and algorithm steps of the various examples 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0128] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0129] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for ultrasonic scalpel position recognition based on image processing, characterized in that, include: Acquire information on multiple intraoperative images to be identified, as well as the acquisition time information of multiple images to be identified; The intraoperative image information to be identified corresponds one-to-one with the image acquisition time information to be identified; Based on the multiple intraoperative image information to be identified, a correction and transformation process is performed to generate multiple corrected intraoperative image information; Based on the preset ultrasonic scalpel blade pixel recognition model, multiple ultrasonic scalpel blade pixel information are calculated according to the multiple corrected intraoperative image information. Based on the pixel information of the multiple ultrasonic scalpel bodies, the acquisition time information of multiple images to be identified, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient, multiple ultrasonic scalpel body position range information are generated. Based on the multiple intraoperative image information to be identified, the multiple image acquisition time information to be identified, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model, multiple ultrasonic scalpel position recognition information are generated.

2. The ultrasonic scalpel position recognition method based on image processing as described in claim 1, characterized in that, The step of performing correction and transformation processing on the multiple intraoperative image information to be identified, and generating multiple corrected intraoperative image information, specifically includes: Grayscale mapping processing is performed on the multiple intraoperative image information to be identified to generate multiple intraoperative grayscale image information to be identified; The multiple intraoperative grayscale image information to be identified is filtered to generate multiple filtered intraoperative image information. Based on the multiple filtered intraoperative image information, contour extraction and filling processes are performed to generate multiple corrected intraoperative image information.

3. The ultrasonic scalpel position recognition method based on image processing as described in claim 1, characterized in that, The step of calculating multiple ultrasonic scalpel pixel information based on the preset ultrasonic scalpel pixel recognition model and the multiple corrected intraoperative image information specifically includes: Based on a preset intraoperative image spatial coordinate system, pixel coordinate extraction is performed on the multiple corrected intraoperative image information to generate multiple corrected intraoperative image pixel coordinate information. Based on the preset ultrasonic scalpel blade pixel recognition model, multiple ultrasonic scalpel blade pixel information is calculated according to the multiple corrected intraoperative image information and the multiple corrected intraoperative image pixel coordinate information.

4. The ultrasonic scalpel position recognition method based on image processing as described in claim 1, characterized in that, The step of generating multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel information, the multiple image acquisition time information to be identified, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient specifically includes: Based on the pixel information of the multiple ultrasonic scalpel bodies and the acquisition time information of the multiple images to be identified, the displacement information of the multiple ultrasonic scalpel bodies pixels is calculated. Based on the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body pixel displacement information, and the preset ultrasonic scalpel body identification pixel position offset correction coefficient, the position information of multiple ultrasonic scalpel body pixels after displacement is calculated. Based on the position information of the multiple ultrasonic scalpel blade pixels after displacement and the acquisition time information of the multiple images to be identified, the time information of the multiple ultrasonic scalpel blade pixels after displacement is obtained; Based on the position information and time information of the multiple ultrasonic scalpel body pixels after displacement, the ultrasonic scalpel body pixel displacement imaging time information is calculated. Based on the multiple ultrasonic scalpel body pixel displacement information, the multiple ultrasonic scalpel body pixel displacement position information, the multiple ultrasonic scalpel body pixel displacement time information, and the ultrasonic scalpel body pixel displacement imaging time information, multiple ultrasonic scalpel body position range information are generated.

5. The ultrasonic scalpel position recognition method based on image processing as described in claim 1, characterized in that, The step of generating multiple ultrasonic scalpel position recognition information based on the multiple intraoperative image information to be recognized, the multiple image acquisition time information to be recognized, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model specifically includes: Based on the acquisition time information of the multiple images to be identified and the pixel information of the multiple ultrasonic scalpel bodies, a time-series curve of the position change of the ultrasonic scalpel body pixels is generated. Based on the acquisition time information of the multiple images to be identified and the time-series curve of the pixel position change of the ultrasonic scalpel body, the initial moving speed information of the ultrasonic scalpel body is calculated. Based on the time-series curve of the position change of the ultrasonic scalpel body pixel points, the spatial extreme values ​​of the multiple ultrasonic scalpel body pixel point information are extracted to obtain the pixel point information of the multiple ultrasonic scalpel body extension positions. Based on the initial moving speed information of the ultrasonic scalpel body, the pixel information of multiple ultrasonic scalpel body extension positions, the pixel information of multiple ultrasonic scalpel body, the position range information of multiple ultrasonic scalpel body, and the preset ultrasonic scalpel position recognition model, multiple ultrasonic scalpel position recognition information are generated.

6. The ultrasonic scalpel position recognition method based on image processing as described in claim 5, characterized in that, The step of generating multiple ultrasonic scalpel position recognition information based on the initial moving speed information of the ultrasonic scalpel body, multiple ultrasonic scalpel body extension position pixel information, multiple ultrasonic scalpel body pixel information, multiple ultrasonic scalpel body position range information, and a preset ultrasonic scalpel position recognition model specifically includes: Based on the initial moving rate information of the ultrasonic scalpel body and the preset reference moving rate information of the ultrasonic scalpel body, the normalized information of the initial moving rate of the ultrasonic scalpel body is calculated. Based on the pixel information of the multiple ultrasonic scalpel extension positions, the pixel information of the multiple ultrasonic scalpels, the position range information of the multiple ultrasonic scalpels, and the preset intraoperative wound calibration aperture parameter information, the normalized information of the positions of the multiple ultrasonic scalpel extension pixels, the normalized information of the positions of the multiple ultrasonic scalpel pixels, and the normalized information of the positions of the multiple ultrasonic scalpels are calculated. Based on the normalized information of the initial movement rate of the ultrasonic scalpel body, the normalized information of the positions of multiple extended pixels of the ultrasonic scalpel body, the normalized information of the positions of multiple pixels of the ultrasonic scalpel body, the normalized information of the position boundaries of the ultrasonic scalpel body, and the preset ultrasonic scalpel position recognition model, multiple ultrasonic scalpel position recognition information are generated.

7. The ultrasonic scalpel position recognition method based on image processing as described in claim 6, characterized in that, The step of generating multiple ultrasonic scalpel position recognition information based on the normalized information of the initial moving speed of the ultrasonic scalpel body, the normalized information of the positions of multiple extended pixels of the ultrasonic scalpel body, the normalized information of the positions of multiple pixels of the ultrasonic scalpel body, the normalized information of the position boundaries of the ultrasonic scalpel body, and the preset ultrasonic scalpel position recognition model specifically includes: Based on the normalized information of the initial moving rate of the ultrasonic scalpel body, the normalized information of the positions of multiple extended pixels of the ultrasonic scalpel body, the normalized information of the positions of multiple pixels of the ultrasonic scalpel body, and the normalized information of the position boundaries of the ultrasonic scalpel body, nonlinear feature mapping processing is performed to generate the initial moving rate feature information of the ultrasonic scalpel body, the position feature information of multiple extended pixels of the ultrasonic scalpel body, the position feature information of multiple pixels of the ultrasonic scalpel body, and the position boundary feature information of multiple ultrasonic scalpel bodies. Based on the initial movement rate feature information of the ultrasonic scalpel body, the position feature information of multiple ultrasonic scalpel body extension pixels, the position feature information of multiple ultrasonic scalpel body pixels, the position boundary feature information of multiple ultrasonic scalpel body, and the preset ultrasonic scalpel position recognition model, multiple ultrasonic scalpel position recognition information are generated.

8. An ultrasonic scalpel position recognition device based on image processing, characterized in that, include: The module for acquiring intraoperative image information and acquisition time information of images to be identified is used to acquire multiple intraoperative image information and acquisition time information of multiple images to be identified. The intraoperative image information to be identified corresponds one-to-one with the image acquisition time information to be identified; The modified intraoperative image information generation module is used to perform correction and transformation processing on the multiple intraoperative image information to be identified, and generate multiple modified intraoperative image information. The ultrasonic scalpel body pixel information generation module is used to calculate multiple ultrasonic scalpel body pixel information based on a preset ultrasonic scalpel body pixel recognition model and the multiple corrected intraoperative image information. The ultrasonic scalpel body position range information generation module is used to generate multiple ultrasonic scalpel body position range information based on the multiple ultrasonic scalpel body pixel information, multiple image acquisition time information to be identified, and a preset ultrasonic scalpel body identification pixel position offset correction coefficient. The ultrasonic scalpel position recognition information generation module is used to generate multiple ultrasonic scalpel position recognition information based on the multiple intraoperative image information to be recognized, the multiple image acquisition time information to be recognized, the multiple ultrasonic scalpel body pixel information, the multiple ultrasonic scalpel body position range information, and the preset ultrasonic scalpel position recognition model.

9. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images