Method and system for generating three-dimensional topographic map of tissue region in forward-looking ultrasound scenario in endoscopic surgery

CN122911632APending Publication Date: 2026-10-16SHANGHAI RONGSHENG INNOVATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610976355.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-02
Publication Date
2026-10-16

AI Technical Summary

Technical Problem

[0005]本发明的主要目的在于解决现有技术中在内窥镜手术中前向式超声场景下无法根据超声信号准确获得不同深度范围内的组织分布信息的技术问题

Benefits of technology

[0016]In the technical solution provided by this invention, a robotic arm is controlled to drive a forward-facing ultrasound probe to scan along a preset scanning trajectory, continuously acquiring several frames of initial radio frequency (RF) signals and recording the robotic arm parameters corresponding to each frame of initial RF signals; the initial RF signals of each frame are preprocessed to obtain several frames of standardized RF signals; a tissue recognition model is called to perform tissue recognition on the standardized RF signals to obtain the tissue label distribution along the depth direction; target tissue regions corresponding to different depth ranges are determined based on the tissue label distribution, wherein each target tissue region includes a starting depth and an ending depth; based on the continuity constraints between each sampling frame, the target tissue regions corresponding to different depth ranges are corrected to obtain a corrected target tissue region sequence; based on the calibration relationship between the ultrasound probe and the end effector of the robotic arm and the robotic arm parameters, the probe sampling pose corresponding to each frame of standardized RF signals is determined; based on the corrected target tissue region sequence and the corresponding probe sampling pose, each target tissue region is mapped to a three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region used to characterize the spatial distribution of tissue. This method can accurately obtain tissue distribution information at different depths in a forward-facing ultrasound environment during endoscopic surgery, and then map different tissue regions onto a unified three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region. This provides reliable auxiliary information for intraoperative tissue assessment, instrument path adjustment, and risk assessment. Correspondingly, the system, electronic device, computer-readable storage medium, and computer program product provided by this invention also solve the above-mentioned technical problems and achieve similar technical effects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122911632A_ABST
    Figure CN122911632A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of ultrasonic imaging, and discloses a method and system for generating a three-dimensional topographic map of a tissue region in a forward-mode ultrasonic scene in endoscopic surgery, which are used to solve the technical problem that the prior art cannot accurately obtain tissue distribution information in different depth ranges according to ultrasonic signals in a forward-mode ultrasonic scene in endoscopic surgery. The method comprises the following steps: calling a forward-mode ultrasonic probe to collect initial radio frequency signals and performing preprocessing to obtain standardized radio frequency signals; calling a tissue recognition model to perform tissue recognition on the standardized radio frequency signals to obtain tissue label distribution along a depth direction; determining target tissue regions corresponding to different depth ranges according to the tissue label distribution; determining a probe sampling pose based on a calibration relationship and mechanical arm parameters; and mapping each target tissue region to a three-dimensional space coordinate system according to the sequence of the corrected target tissue regions and the corresponding probe sampling poses to generate a three-dimensional topographic map for representing tissue spatial distribution.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of ultrasound imaging technology, and in particular to a method, system, electronic device, computer storage medium, and computer program product for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery. Background Technology

[0002] In minimally invasive endoscopic surgeries such as spinal surgery, surgeons typically need to manipulate bone or adjacent tissues within narrow passages and limited fields of vision. Due to the complex structure in front of the instruments, limited workspace, and significant obstruction of the field of vision, it is difficult to obtain real-time spatial undulations and layered distribution of the local tissue in front of the instruments by relying solely on preoperative CT or MRI images, the surgeon's experience and intuition, or limited endoscopic visual information. In minimally invasive orthopedic settings, the limited space at the instrument tip necessitates extremely high demands for forward-looking detection and real-time intraoperative perception.

[0003] Existing technologies include several ultrasound 3D imaging or reconstruction methods for scenarios involving in vitro free scanning and recovery of mixed large-field-of-view volume data. However, these methods cannot be applied to minimally invasive orthopedic surgery for several reasons: (1) Existing methods mostly use complete volume data, volume rendering images or conventional three-dimensional volumes as output, focusing on restoring three-dimensional imaging volumes; while in endoscopic surgery, what is needed is to obtain three-dimensional topographic maps, depth changes and cross-sectional information of tissue areas based on RF signals in front of the instrument, to help determine the spatial relationship of tissues in front of the instrument, rather than complete three-dimensional bone tissue reconstruction in a generalized sense. (2) Existing methods usually rely on pose sensors or external positioning information to complete spatial registration. However, in endoscopic surgery, the space at the front end of the instrument is small and the structural integration is limited. It is not only difficult to install an additional complex pose acquisition device, but it also increases the system size, assembly difficulty and the complexity of use during the operation. (3) Existing lateral rotation ultrasound schemes mainly image around the circumference of the instrument. Their detection direction is different from the forward detection scenario of the present invention, and it is difficult to directly use it for continuous three-dimensional terrain perception of the tissue area in front of the instrument axis.

[0004] Therefore, there is an urgent need for a method to generate three-dimensional topographic maps of tissue regions suitable for forward ultrasound scenarios in endoscopic surgery. This method should be able to determine the two layers of tissue regions in front of the instrument based on the sliding window tissue recognition results of one-dimensional RF signals without relying on external pose sensors. It should also combine continuous scanning constraints to correct for incorrect labels, regional breaks, and abrupt depth changes. Furthermore, based on the robotic arm parameters and probe calibration relationship, the sampling pose should be calculated to map different tissue regions to a unified three-dimensional spatial coordinate system, generating three-dimensional topographic maps of each tissue region. This would provide reliable auxiliary information for intraoperative tissue judgment, instrument path adjustment, and risk warning. Summary of the Invention

[0005] The main objective of this invention is to solve the technical problem in the prior art that it is impossible to accurately obtain tissue distribution information at different depths based on ultrasound signals in forward ultrasound scenarios during endoscopic surgery.

[0006] The first aspect of this invention provides a method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, comprising: The robotic arm is controlled to drive the forward-facing ultrasonic probe to scan along a preset scanning trajectory, continuously acquiring several frames of initial radio frequency signals, and recording the robotic arm parameters corresponding to each frame of the initial radio frequency signal. The initial radio frequency signals of each frame are preprocessed to obtain several frames of standardized radio frequency signals; The tissue identification model is invoked to perform tissue identification on the standardized radio frequency signal to obtain the tissue tag distribution along the depth direction; Based on the distribution of the tissue labels, target tissue regions corresponding to different depth ranges are determined, wherein each target tissue region includes a starting depth and an ending depth; Based on the continuity constraint between each sampling frame, the target tissue regions corresponding to different depth ranges are corrected to obtain the corrected target tissue region sequence. Based on the calibration relationship between the ultrasonic probe and the end effector of the robotic arm, and the parameters of the robotic arm, the probe sampling pose corresponding to each frame of standardized radio frequency signal is determined. Based on the corrected target tissue region sequence and the corresponding probe sampling pose, each target tissue region is mapped to a three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region used to characterize the spatial distribution of tissue.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the step of correcting the target tissue region corresponding to different depth ranges based on the continuity constraints between each sampling frame includes: Obtain the target tissue region corresponding to adjacent sampling frames; Based on preset continuity constraints, continuity analysis is performed on the target tissue regions corresponding to adjacent sampling frames; If the target tissue region corresponding to the target sampling frame is detected to not meet the preset continuity condition, the target tissue region corresponding to the target sampling frame is corrected according to the target tissue region corresponding to the adjacent sampling frame to obtain the corrected target tissue region sequence.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the following is characterized in that: The continuity constraints include mechanical scanning sequence constraints, tissue depth sequence constraints, tissue label consistency constraints, and confidence weight constraints.

[0009] Optionally, in a third implementation of the first aspect of the present invention, the step of calling a tissue identification model to perform tissue identification on the standardized radio frequency signal to obtain the tissue tag distribution along the depth direction includes: The standardized radio frequency signal is divided into sliding windows to obtain radio frequency signal segments corresponding to each sliding window; The pre-trained tissue recognition model is invoked to identify the tissue labels of each radio frequency signal segment, resulting in a window-level tissue label sequence along the depth direction, and the recognition confidence of the tissue labels is output. Based on the window-level organization tag sequence, the organization tag distribution corresponding to the standardized radio frequency signal of each frame is mapped.

[0010] Optionally, in a fourth implementation of the first aspect of the present invention, determining the target tissue region corresponding to different depth ranges based on the tissue tag distribution includes: Adjacent depth regions with the same tissue label are merged to obtain several candidate tissue regions; The organization category corresponding to the candidate organization region is determined based on the preset region identification index, and the target organization region is determined; The target organizational region includes at least two different types of organizational regions.

[0011] Optionally, in a fifth implementation of the first aspect of the invention, after generating a three-dimensional topographic map of the tissue region for characterizing the spatial distribution of the tissue, the method further includes: Based on the three-dimensional topographic map, extract profile information along the scanning trajectory or a specified path; Based on the profile information, calculate the distance variation, thickness variation, and spatial undulation variation information between different target tissue regions.

[0012] A second aspect of the present invention provides a system for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery, comprising: The signal acquisition module is used to control the robotic arm to drive the forward ultrasonic probe to scan along a preset scanning trajectory, continuously acquire several frames of initial radio frequency signals, and record the robotic arm parameters corresponding to each frame of the initial radio frequency signal. The preprocessing module is used to preprocess the initial radio frequency signals of each frame to obtain several frames of standardized radio frequency signals; The tissue identification module is used to call the tissue identification model to perform tissue identification on the standardized radio frequency signal and obtain the tissue tag distribution along the depth direction; The region division module is used to determine target tissue regions corresponding to different depth ranges based on the distribution of tissue labels, wherein each target tissue region includes a start depth and an end depth; The correction module is used to correct the target tissue regions corresponding to different depth ranges based on the continuity constraints between each sampling frame, so as to obtain the corrected target tissue region sequence. The pose calculation module is used to determine the probe sampling pose corresponding to each frame of standardized radio frequency signal based on the calibration relationship between the ultrasonic probe and the end effector of the robotic arm and the parameters of the robotic arm. The topographic map generation module is used to map each target tissue region to a three-dimensional spatial coordinate system based on the corrected target tissue region sequence and the corresponding probe sampling pose, and generate a three-dimensional topographic map of the tissue region to characterize the spatial distribution of the tissue.

[0013] A third aspect of the present invention provides a device for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the device for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery to perform the steps of the above-described method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the steps of the method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery.

[0015] A fifth aspect of the present invention provides a computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method for generating a three-dimensional topographic map of a tissue region for a forward ultrasound scene in endoscopic surgery described above.

[0016] In the technical solution provided by this invention, a robotic arm is controlled to drive a forward-facing ultrasound probe to scan along a preset scanning trajectory, continuously acquiring several frames of initial radio frequency (RF) signals and recording the robotic arm parameters corresponding to each frame of initial RF signals; the initial RF signals of each frame are preprocessed to obtain several frames of standardized RF signals; a tissue recognition model is called to perform tissue recognition on the standardized RF signals to obtain the tissue label distribution along the depth direction; target tissue regions corresponding to different depth ranges are determined based on the tissue label distribution, wherein each target tissue region includes a starting depth and an ending depth; based on the continuity constraints between each sampling frame, the target tissue regions corresponding to different depth ranges are corrected to obtain a corrected target tissue region sequence; based on the calibration relationship between the ultrasound probe and the end effector of the robotic arm and the robotic arm parameters, the probe sampling pose corresponding to each frame of standardized RF signals is determined; based on the corrected target tissue region sequence and the corresponding probe sampling pose, each target tissue region is mapped to a three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region used to characterize the spatial distribution of tissue. This method can accurately obtain tissue distribution information at different depths in a forward-facing ultrasound environment during endoscopic surgery, and then map different tissue regions onto a unified three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region. This provides reliable auxiliary information for intraoperative tissue assessment, instrument path adjustment, and risk assessment. Correspondingly, the system, electronic device, computer-readable storage medium, and computer program product provided by this invention also solve the above-mentioned technical problems and achieve similar technical effects. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating a method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, as described in an embodiment of the present invention. Figure 2 This is a schematic diagram of the scanning process for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, as described in an embodiment of the present invention. Figure 3 This is a schematic diagram of the data acquisition process during the scanning process of the method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, as described in an embodiment of the present invention. Figure 4 This is a schematic diagram of the sampling trajectory of the ultrasound probe in the method for generating a three-dimensional topographic map of a tissue region in a forward ultrasound scene during endoscopic surgery, as described in an embodiment of the present invention. Figure 5This is a schematic diagram of an embodiment of a tissue region three-dimensional topographic map generation system for a forward ultrasound scene in endoscopic surgery according to an embodiment of the present invention; Figure 6 This is a schematic diagram of an embodiment of a tissue region three-dimensional topographic map generation device for a forward ultrasound scene in endoscopic surgery according to an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the principle of a computer-readable medium according to an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the invention will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limiting the invention to the embodiments set forth herein. Rather, these exemplary embodiments are provided to make the invention more comprehensive and complete, and to facilitate a full communication of the inventive concept to those skilled in the art. The same reference numerals in the drawings denote the same or similar elements, components, or parts, and therefore repeated descriptions of them will be omitted.

[0019] Subject to the technical concept of this invention, the features, structures, characteristics or other details described in a particular embodiment may be combined in one or more other embodiments in a suitable manner.

[0020] In the description of specific embodiments, the features, structures, characteristics, or other details described in this invention are intended to enable those skilled in the art to fully understand the embodiments. However, it is not excluded that those skilled in the art can practice the technical solutions of this invention without one or more of the specific features, structures, characteristics, or other details.

[0021] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0022] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0023] The terms “and / or” or “and / or” include all combinations of any one or more of the listed items.

[0024] See Figures 1-4The first embodiment of the method for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery, as described in this invention, includes: It is understood that the executing entity of this invention can be a three-dimensional topographic map generation system for tissue regions in forward ultrasound scenes during endoscopic surgery, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0025] S101. Control the robotic arm to drive the forward ultrasonic probe to scan along the preset scanning trajectory, continuously acquire several frames of initial radio frequency signals, and record the robotic arm parameters corresponding to each frame of initial radio frequency signal. The scanning process in this step is as follows: Figure 2 As shown, a forward-facing ultrasound probe is first integrated into the instrument's front end. This probe is driven by a robotic arm and scans the area in front. During scanning, it continuously scans along a preset scanning trajectory within the surgical channel to probe deep tissues at the bottom of the channel. The direct result of the scan is the acquisition of continuous one-dimensional radio frequency (RF) signals, resulting in an initial RF signal sequence. This initial RF sequence can be determined by setting sampling frames at fixed acquisition time intervals to define the corresponding time points. (Continue reading...) Figure 3 When performing a specific scan, it is necessary to first set the scanning angle range and preset the scanning trajectory. Based on actual usage requirements, a single arc or multi-angle trajectory can be used for scanning. While moving along the scanning trajectory, the probe is also controlled to swing along the trajectory through a mechanical transmission device. During this process, multiple frames of initial radio frequency signals are continuously acquired, and the correspondence between radio frequency signals and trajectory parameters is established.

[0026] In one specific implementation, in a minimally invasive spinal endoscopic surgery scenario, a miniature forward-facing ultrasound probe is mounted on the front end of an endoscopic instrument, with the probe's sound beam pointing in the direction of the instrument's movement. A robotic arm or mechanical control mechanism drives the probe to scan along a single arc trajectory. A sequence of one-dimensional radio frequency signal frames is continuously acquired at equal arc length intervals along the preset arc trajectory, and the arc length parameters, robotic arm joint angle parameters, and scanning direction angle corresponding to each frame are recorded simultaneously, thereby establishing a correspondence between radio frequency frames and trajectory parameters. (Continue reading...) Figure 4 In another embodiment, a miniature forward-facing ultrasonic probe can be driven by a robotic arm or mechanical control mechanism to perform arc scanning at multiple rotation angles, and the arc trajectories in multiple directions can be combined to form a multi-directional sampling trajectory similar to a "rice" shape.

[0027] In another specific implementation, a preset trajectory with equal arc length sampling is used to determine multiple sampling positions of the initial radio frequency signal. Specifically, the scanning trajectory, scanning start point, scanning end point, scanning direction angle, and sampling interval are preset in the robotic arm control system. Let the arc length sampling interval of the preset trajectory be... Then the first Trajectory parameters corresponding to each sampling position The expression is: ; In this case, the robotic arm moves to the... When the sampling point reaches the first arc length, the control module triggers the forward ultrasonic probe to acquire the first... The initial radio frequency signal of the frame, denoted as ,in, Indicates the echo time. Also records the first... Robotic arm parameters corresponding to the initial radio frequency signal of the frame The expression is: ; in, This indicates the joint angle parameters read by the robotic arm control system. This represents the arc length parameter on the preset trajectory. This indicates the current scanning direction angle. Therefore, the sampling position in this invention is not a spatial coordinate measured by an external pose sensor, but rather a positional information determined by the robotic arm control system based on a preset trajectory and joint angle parameters. These robotic arm parameters are recorded synchronously with the initial radio frequency signal acquisition process and are used to subsequently calculate the spatial sampling pose of each frame of the initial radio frequency signal.

[0028] After continuous acquisition, an initial radio frequency signal sequence arranged in scanning order is obtained. The expression is: ; in, Indicates the number of frames captured.

[0029] This acquisition method is applicable to minimally invasive endoscopic surgeries where the instrument tip space is limited, probe size is limited, and the surgical field of view is limited. It can acquire one-dimensional radio frequency signals through a miniature forward-facing ultrasound probe, so that even when high-quality two-dimensional ultrasound images cannot be obtained, the tissue distribution at different depths in front of the instrument can still be identified. This can reduce the impact of the limited number of miniature probe array elements, imaging field of view, and data bandwidth on the tissue perception results. Furthermore, this acquisition method does not rely on optical positioning devices, electromagnetic positioning devices, inertial measurement units, or additional pose sensors.

[0030] S102. Preprocess the initial radio frequency signals of each frame to obtain several frames of standardized radio frequency signals; After obtaining the initial radio frequency signals for each frame, preprocessing is performed first. Preprocessing operations in this step may include depth axis transformation, depth range clipping, and robust normalization.

[0031] First, the echo time of the initial radio frequency signal is converted into a forward depth z along the probe beam direction: ; in, Represents the equivalent speed of sound within an organization. This indicates the forward depth along the direction of the probe's acoustic beam. After conversion, the... The frame radio frequency signal is represented as .

[0032] Subsequently, based on the effective detection distance of the forward-facing ultrasound probe and the depth range of the target tissue, a fixed effective depth range was set. And perform depth range clipping on the radio frequency signal: ; By clipping the depth range in this step, potential near-field ringing of the probe, invalid shallow interference, and low signal-to-noise ratio signals in the far field can be eliminated from the radio frequency signal, allowing subsequent tissue identification to focus on the effective depth range that is meaningful for tissue identification.

[0033] After completing the depth range clipping, the signal within the effective depth range is robustly normalized: ; in, This represents the median of the RF signal after clipping. and These represent the 95th and 5th percentiles of the clipped signal amplitude, respectively. To prevent constants with a denominator of zero.

[0034] After processing, a standardized radio frequency signal is obtained, which is then used... It is indicated that the standardized radio frequency signal is a one-dimensional radio frequency signal.

[0035] This normalization method is performed only within the effective depth range and does not use the maximum and minimum values ​​of the entire original radio frequency signal for simple scaling. It can reduce the impact of individual abnormal peaks, acoustic coupling changes, probe contact state changes, and incident angle changes on the amplitude scale of the entire frame signal.

[0036] S103. Call the tissue identification model to perform tissue identification on the standardized radio frequency signal to obtain the tissue tag distribution along the depth direction; After obtaining the standardized radio frequency (RF) signal, the continuous waveform of the standardized RF signal is converted into a sequence of tissue tags arranged along the depth direction. Specifically, this includes: dividing the standardized RF signal into sliding windows to obtain RF signal segments corresponding to each sliding window; calling a pre-trained tissue recognition model to identify the tissue tags of each RF signal segment, obtaining a window-level tissue tag sequence along the depth direction, and outputting the recognition confidence of the tissue tags; and mapping the tissue tag distribution corresponding to the standardized RF signal of each frame according to the window-level tissue tag sequence.

[0037] In one specific implementation, the standardized radio frequency signal obtained in S102 is first processed... Divide the window along the forward depth direction. Indicates the length of the sliding window. Let the sliding step size be denoted as , then the ... The first frame in the normalized radio frequency signal A sliding window The expression is: ; in, This indicates the starting depth position of the sliding window. Each sliding window corresponds to a radio frequency signal segment within a continuous depth range, used to represent the tissue echo characteristics within that depth range. Sliding window length. and step length Maintaining consistency during training and inference ensures that tissue identification results across different sampled frames have a uniform depth scale.

[0038] Subsequently, a pre-trained tissue recognition model is used to classify tissues within the sliding window. This tissue recognition model can be a one-dimensional Vision Transformer tissue recognition model (referred to as 1D ViT). This model takes a one-dimensional radio frequency (RF) signal segment corresponding to each sliding window as input, without converting the RF signal into a two-dimensional ultrasound image. Specifically, the tissue recognition model first divides each RF signal window into several one-dimensional RF signal segments (patch) and maps each segment to a feature vector; then, a one-dimensional depth position encoding is added to preserve the depth order of different signal segments within the window; next, the encoder in the Transformer structure extracts the correlation features between segments of different depths within the window; finally, the classification head of the Transformer structure outputs the tissue label and recognition confidence score corresponding to the window.

[0039] The above one-dimensional radio frequency signal segment embedding and position encoding process is expressed as follows: ; in, This represents the window feature sequence of the encoder of the input tissue recognition model. This indicates a one-dimensional radio frequency signal segment embedding operation. This represents a one-dimensional depth location encoding.

[0040] After inputting the one-dimensional radio frequency signal segment corresponding to the sliding window into the trained tissue recognition model, the tissue label and recognition confidence score are obtained: ; in, This represents the trained tissue identification model. Indicates the first Frame number Predicted organization labels for one-dimensional radio frequency signal segments corresponding to each sliding window This indicates the confidence level of the prediction result.

[0041] The organization labels include a first target organization, a second target organization, and a non-target area. The first and second target organizations are used to generate a two-layer three-dimensional terrain map of the organization regions; the non-target areas are used to represent background, invalid signals, and organization signals that do not participate in the generation of the three-dimensional terrain map.

[0042] Based on the above steps, complete the process for the first... After identifying all one-dimensional radio frequency signal segments corresponding to the sliding windows in the frame radio frequency signal, a window-level organization label sequence arranged along the depth direction is obtained: ; in, Indicates the first The number of sliding windows in the frame of radio frequency signal. This tag sequence reflects the distribution of tissue categories at different depth locations in the frame of radio frequency signal.

[0043] The purpose of this step is to convert the continuous waveform of the one-dimensional radio frequency signal into a sequence of tissue tags arranged along the depth direction. Based on the specific steps of this embodiment, it is clear that this invention does not search for individual echo peaks in the radio frequency signal, nor does it extract tissue interfaces. Instead, it utilizes a tissue recognition model to identify tissue echo features within a continuous depth window, providing a foundation for subsequent determination of two-layer tissue regions, multi-frame continuity correction, and the generation of three-dimensional topographic maps.

[0044] S104. Determine the target tissue regions corresponding to different depth ranges based on the distribution of tissue labels; Obtain the first from step S103 After organizing the frame into a window-level label sequence, this step maps the recognition result corresponding to each sliding window back to the depth axis, merges adjacent depth intervals with the same organization label, and obtains several candidate organization regions. The organization category corresponding to the candidate organization region is determined according to the preset region recognition index, and the target organization region is determined. The target organization region includes at least two different categories of organization regions, and each target organization region is divided by a start depth and an end depth.

[0045] In one specific implementation, after obtaining the first After the window-level organization label sequence of the frame, the sliding window recognition results are mapped back to the depth axis to form a one-dimensional organization label distribution for that frame. Adjacent sliding windows with the same organization labels are merged to obtain candidate organization regions. Among them, the first... Each candidate organization region is denoted as: ; in, Indicates the starting depth of the candidate tissue region. Indicates the termination depth of the candidate organization region.

[0046] The average confidence level of candidate tissue regions is defined as: ; in, Indicates belonging to the candidate organization region A set of sliding window numbers.

[0047] The main objective of this embodiment is to determine two layers of tissue regions based on tissue labels, depth order, region length, and average confidence level. These two layers include a first target tissue region and a second target tissue region, while other regions are classified as non-target regions. The first target tissue region and the second target tissue region are denoted as follows: ; ; in, Indicates the first target organization area. This represents the second target organizational region; the two organizational regions satisfy the following depth order constraint: ; in, Indicates the initial depth of the first target tissue region. Indicates the termination depth of the first target organization region; Indicates the starting depth of the second target tissue region. This indicates the termination depth of the second target organization region.

[0048] Meanwhile, candidate tissue regions need to meet minimum region length and minimum confidence requirements. When multiple candidate regions exist under the same tissue label, the candidate region with a larger region length, higher average confidence, and better internal label continuity is selected as the target tissue region.

[0049] In a preferred embodiment, the purpose of distinguishing between the first target tissue region and the second target tissue region is to differentiate between bone tissue and adjacent tissues, preventing problems such as excessive grinding during surgery. The first and second target tissue regions are respectively a bone tissue region (e.g., lamina, facet joints, or cortical bone) and a soft tissue region adjacent to the bone tissue (e.g., ligamentum flavum, intervertebral disc, muscle tissue). The specific tissue types of the first and second target tissue regions depend on the actual surgical scenario and the spatial position of the probe relative to the tissue.

[0050] After the above processing, each frame of one-dimensional normalized radio frequency signal is represented as two tissue regions, each described by a start depth and an end depth. It is important to note that this step yields the identification results of two tissue regions, not the tissue interface extraction results.

[0051] In this step, by using sliding window tissue identification and region merging, continuous radio frequency signals can be converted into two-layer tissue region results, avoiding insufficient information caused by relying solely on a single echo peak or traditional interface extraction, and further improving identification accuracy.

[0052] S105. Based on the continuity constraints between each sampling frame, the target tissue regions corresponding to different depth ranges are corrected to obtain the corrected target tissue region sequence. Because forward ultrasound signals during endoscopic surgery are susceptible to changes in acoustic coupling, probe contact status, noise, and artifacts, single-frame tissue identification results may exhibit transient mislabeling, regional breaks, and abrupt depth changes. Therefore, this step utilizes the continuity of multiple frames during the scanning process to perform continuity correction and erroneous information filtering on the two tissue layers.

[0053] During the correction process, the target tissue regions corresponding to adjacent sampling frames are first acquired. Based on preset continuity constraints, a continuity analysis is performed on the target tissue regions corresponding to adjacent sampling frames. If the target tissue region corresponding to the target sampling frame is found to not meet the preset continuity conditions, the target tissue region corresponding to the target sampling frame is corrected according to the target tissue regions corresponding to adjacent sampling frames, resulting in the corrected target tissue region sequence. The continuity constraints include mechanical scanning order constraints, tissue depth order constraints, tissue label consistency constraints, and confidence weight constraints.

[0054] In one specific implementation, for the first Frame number Layered organizational regions, whose representative depth is defined as: ; Based on the representative depth, region length, tissue label, and average confidence level, information from adjacent frames can be compared. When the tissue region of a certain frame undergoes a significant abrupt change relative to the preceding and following frames, and this abrupt change does not conform to the spatial continuous change pattern under the mechanical scanning trajectory, the corresponding region of that frame is determined to be erroneous information.

[0055] This embodiment of continuity correction introduces the following four types of constraints: mechanical scanning order constraint, two-layer tissue depth order constraint, tissue label consistency constraint, and confidence weight constraint. Specifically, since the radio frequency signal is continuously acquired along a preset mechanical trajectory and the spatial positions of adjacent frames are continuous, the depth change of the same tissue layer should remain smooth; the first target tissue region is always located before the second target tissue region, and the order of the two tissue layers is not allowed to be reversed during the correction process; the tissue labels of the same tissue layer in consecutive frames should remain stable, and labels that appear briefly and cannot be supported by the preceding and following frames are judged as erroneous labels; frames with higher average confidence have higher weight in continuity correction, while frames with lower average confidence have lower weight in correction.

[0056] In one specific implementation, the start and end depths of the two tissue layers are optimized for continuity as follows: ; in, and This indicates the start and end depths of the tissue region identified in a single frame. and This indicates the starting and ending depths after continuity correction. This represents the weight determined by the average confidence level of the region. This represents the weights of the multi-frame smoothing constraint.

[0057] The optimization result simultaneously satisfies the depth order constraint of two tissue layers: ; After this step, a continuous and stable two-layer tissue region sequence is obtained. This step is not a simple smoothing of tissue labels, but rather a constraint correction of multi-frame tissue region results by combining the mechanical scanning order, the spatial order of the two tissue layers, and the recognition confidence level. In this embodiment, correction is performed by constraining the mechanical scanning order, the depth order of the two tissue layers, the consistency of tissue labels, and the confidence level weight. This can reduce erroneous labels, region breaks, and abrupt depth changes caused by noise, artifacts, changes in acoustic coupling, and changes in probe contact state, thereby improving the continuity and stability of the results.

[0058] S106. Based on the calibration relationship between the ultrasonic probe and the end effector of the robotic arm and the parameters of the robotic arm, determine the probe sampling pose corresponding to each frame of standardized radio frequency signal. After completing the identification and continuity correction of the two-layer tissue regions, the spatial sampling pose of the frame is calculated based on the robotic arm parameters corresponding to each frame of RF signal.

[0059] The robotic arm control system is collecting the first... The robot arm's joint angle parameters, trajectory arc length parameters, and scanning direction angle are recorded synchronously during each frame of radio frequency signal. Based on the robot arm's forward kinematics model, the position and orientation of the robot arm's end effector in the base coordinate system are calculated; then, combined with the probe's installation and calibration relationship relative to the robot arm's end effector, the position and orientation of the ultrasonic probe in the [missing information] frame are obtained. Spatial pose during frame sampling.

[0060] Let the fixed calibration matrix between the probe coordinate system and the robotic arm end effector coordinate system be: Based on the parameters of the robotic arm The calculated end-effector pose is as follows Then collect the first The probe pose corresponding to the frame RF signal is: ; For the In-frame depth is The organization identification result, its position in the unified three-dimensional spatial coordinate system is: ; Therefore, the tissue region in each frame of radio frequency signal can be mapped to a unified three-dimensional space according to its depth range. It can be seen that the pose estimation in this embodiment comes from the robot arm parameters, the preset scanning trajectory, and the probe installation and calibration relationship. It does not rely on optical positioning devices, electromagnetic positioning devices, inertial measurement units, or other external pose sensors. The radio frequency sampling pose can be estimated only based on the robot arm parameters, the preset trajectory parameters, and the probe installation and calibration relationship, which helps to reduce the complexity of the instrument front-end structure and the difficulty of system integration.

[0061] S107. Based on the corrected target tissue region sequence and the corresponding probe sampling pose, map each target tissue region to a three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region used to characterize the spatial distribution of tissue.

[0062] After obtaining the corrected target tissue region sequence output in S105 and the probe sampling pose output in S106, this step can combine the two continuously corrected tissue regions with the spatial sampling pose corresponding to each frame to generate a three-dimensional topographic map of the two tissue regions.

[0063] In one specific implementation, for the first Frame number The corrected depth range for the layered organizational region is: ; According to the The probe's spatial pose in each frame is used to map the initial depth, final depth, and center depth of the tissue region into three-dimensional spatial points. After spatial mapping of all sampled frames, the point sets of the first and second target tissue regions are obtained.

[0064] The two organizational region point sets are denoted as follows: ; ; in, , and They represent the first Frame number The positions of the starting depth point, ending depth point, and center depth point of the layered organization region in three-dimensional space.

[0065] By meshing and surface fitting the point sets of the two organizational regions, a three-dimensional terrain representation is obtained: ; in, This represents a three-dimensional topographic map of the first target area. This represents a three-dimensional topographic map of the second target area. This represents the point set meshing and surface fitting functions.

[0066] The three-dimensional topographic map is used to represent the spatial distribution, depth variation, thickness range, and relative positional relationship of the two layers of tissue in front of the endoscopic instrument. This invention does not perform complete three-dimensional volumetric data reconstruction; instead, it generates a three-dimensional topographic map for intraoperative spatial perception based on the sampling pose calculated from one-dimensional radiofrequency tissue identification results and robotic arm parameters.

[0067] After generating a 3D topographic map of the two tissue regions, profile information can be extracted along the scanning trajectory and the path specified by the doctor. Using this profile information, the depth, thickness, and spatial undulation trends of the first and second target tissue regions along the specified path can be observed.

[0068] Furthermore, it can also calculate the distance changes between two layers of tissue and the distribution of abnormal areas, providing auxiliary information for intraoperative tissue assessment, instrument path adjustment, and risk warning.

[0069] This invention is applicable to scenarios in minimally invasive endoscopic surgery where the instrument tip space is limited, probe size is restricted, and the surgical field of view is limited. It can identify tissue distribution at different depths in front of the instrument using one-dimensional radio frequency signals acquired by a miniature forward-facing ultrasound probe, without relying on high-quality two-dimensional ultrasound images. Therefore, it can reduce the impact of limited miniature probe element count, imaging field of view, and data bandwidth on tissue perception results. Furthermore, this invention can calculate the sampling pose of the radio frequency signal based on robotic arm parameters, preset trajectory parameters, and probe installation calibration relationships, without relying on optical positioning devices, electromagnetic positioning devices, inertial measurement units, or other external pose sensors, which helps reduce the complexity of the instrument tip structure and the difficulty of system integration.

[0070] Furthermore, in this embodiment, when performing tissue region identification, it can convert continuous radio frequency signals into two-layer tissue region results through sliding window tissue identification and region merging, avoiding insufficient information caused by relying solely on a single echo peak or traditional interface extraction; through mechanical scanning sequence, two-layer tissue depth sequence, tissue label consistency and confidence weight constraints, it can reduce erroneous labels, region breaks and abrupt depth changes caused by noise, artifacts, changes in acoustic coupling and changes in probe contact state, thereby improving the continuity and stability of the results.

[0071] Furthermore, the two-layer tissue region three-dimensional topographic map generated in this embodiment can express the spatial distribution, depth variation, thickness range and relative positional relationship of the tissue region in front of the instrument, providing basic data for profile analysis, interlayer distance assessment and abnormal area indication; compared with general three-dimensional ultrasound reconstruction methods, this invention is more suitable for the anterior detection needs of miniature forward ultrasound probes in endoscopic surgery, and the output results are more convenient for intraoperative tissue judgment, instrument path adjustment and risk indication.

[0072] The above describes the method for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery according to embodiments of the present invention. The following describes the system for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery according to embodiments of the present invention. (See reference...) Figure 5 One embodiment of the tissue region three-dimensional topographic map generation system for forward ultrasound scenes in endoscopic surgery according to the present invention includes: The signal acquisition module 501 is used to control the robotic arm to drive the forward ultrasonic probe to scan along a preset scanning trajectory, continuously acquire several frames of initial radio frequency signals, and record the robotic arm parameters corresponding to each frame of the initial radio frequency signal. Preprocessing module 502 is used to preprocess the initial radio frequency signals of each frame to obtain several frames of standardized radio frequency signals; The tissue identification module 503 is used to call the tissue identification model to perform tissue identification on the standardized radio frequency signal and obtain the tissue tag distribution along the depth direction; The region segmentation module 504 is used to determine target tissue regions corresponding to different depth ranges based on the tissue label distribution, wherein each target tissue region includes a start depth and an end depth; The correction module 505 is used to correct the target tissue regions corresponding to different depth ranges based on the continuity constraints between each sampling frame, so as to obtain the corrected target tissue region sequence. The pose calculation module 506 is used to determine the probe sampling pose corresponding to each frame of standardized radio frequency signal based on the calibration relationship between the ultrasonic probe and the end of the robotic arm and the parameters of the robotic arm. The topographic map generation module 507 is used to map each target tissue region to a three-dimensional spatial coordinate system based on the corrected target tissue region sequence and the corresponding probe sampling pose, and generate a three-dimensional topographic map of the tissue region to characterize the spatial distribution of the tissue.

[0073] The system provided in this invention can accurately obtain tissue distribution information at different depths in a forward ultrasound scenario during endoscopic surgery, and then map different tissue regions to a unified three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region, thereby providing reliable auxiliary information for intraoperative tissue judgment, instrument path adjustment and risk warning.

[0074] In another embodiment of this application, the correction module 502 is specifically used for: Obtain the target tissue region corresponding to adjacent sampling frames; Based on preset continuity constraints, continuity analysis is performed on the target tissue regions corresponding to adjacent sampling frames; If the target tissue region corresponding to the target sampling frame is detected to not meet the preset continuity condition, the target tissue region corresponding to the target sampling frame is corrected according to the target tissue region corresponding to the adjacent sampling frame to obtain the corrected target tissue region sequence.

[0075] In another embodiment of this application, the continuity constraint includes mechanical scanning sequence constraint, tissue depth sequence constraint, tissue label consistency constraint, and confidence weight constraint.

[0076] In another embodiment of this application, the tissue identification module 503 is specifically used for: The standardized radio frequency signal is divided into sliding windows to obtain radio frequency signal segments corresponding to each sliding window; The pre-trained tissue recognition model is invoked to identify the tissue labels of each radio frequency signal segment, resulting in a window-level tissue label sequence along the depth direction, and the recognition confidence of the tissue labels is output. Based on the window-level organization tag sequence, the organization tag distribution corresponding to the standardized radio frequency signal of each frame is mapped.

[0077] In another embodiment of this application, the region division module 504 is specifically used for: Adjacent depth regions with the same tissue label are merged to obtain several candidate tissue regions; The organization category corresponding to the candidate organization region is determined based on the preset region identification index, and the target organization region is determined; The target organizational region includes at least two different types of organizational regions.

[0078] In another embodiment of this application, the topographic map generation module 507 is further configured to: Based on the three-dimensional topographic map, extract profile information along the scanning trajectory or a specified path; Based on the profile information, calculate the distance variation, thickness variation, and spatial undulation variation information between different target tissue regions.

[0079] The system provided in this invention is suitable for scenarios in minimally invasive endoscopic surgery where the instrument tip space is limited, the probe size is limited, and the surgical field of view is limited. It can identify the tissue distribution at different depths in front of the instrument using one-dimensional radio frequency signals acquired by a miniature forward-facing ultrasound probe, without relying on high-quality two-dimensional ultrasound images. Therefore, it can reduce the impact of limited miniature probe element count, imaging field of view, and data bandwidth on tissue perception results. Furthermore, this invention can calculate the sampling pose of the radio frequency signal based on robotic arm parameters, preset trajectory parameters, and probe installation calibration relationships, without relying on optical positioning devices, electromagnetic positioning devices, inertial measurement units, or other external pose sensors, which helps reduce the complexity of the instrument tip structure and the difficulty of system integration.

[0080] Furthermore, when performing tissue region identification, the system provided in this embodiment can convert continuous radio frequency signals into two-layer tissue region results through sliding window tissue identification and region merging, avoiding insufficient information caused by relying solely on a single echo peak or traditional interface extraction. By constraining mechanical scanning sequence, two-layer tissue depth sequence, tissue label consistency, and confidence weight, it can reduce erroneous labels, region breaks, and abrupt depth changes caused by noise, artifacts, changes in acoustic coupling, and changes in probe contact state, thereby improving the continuity and stability of the results.

[0081] Furthermore, the two-layer tissue region three-dimensional topographic map generated by the system provided in this embodiment can express the spatial distribution, depth variation, thickness range and relative positional relationship of the tissue region in front of the instrument, providing basic data for profile analysis, interlayer distance assessment and abnormal area indication; compared with general three-dimensional ultrasound reconstruction methods, this invention is more suitable for the anterior detection needs of miniature forward ultrasound probes in endoscopic surgery, and the output results are more convenient for intraoperative tissue judgment, instrument path adjustment and risk indication.

[0082] Based on the same inventive concept, this specification also provides an electronic device for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery. The electronic device for generating three-dimensional topographic maps of tissue regions in a forward ultrasound scene during endoscopic surgery is described in detail below from the perspective of hardware processing.

[0083] Figure 6 This is a schematic diagram of an electronic device provided as an embodiment of this specification. Refer to the following... Figure 6 The electronic device 600 according to this embodiment of the present invention will be described. Figure 6 The electronic device 600 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0084] like Figure 6 As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0085] The storage unit stores program code that can be executed by the processing unit 610, causing the processing unit 610 to perform the steps described in the processing method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 1 The steps of the method shown.

[0086] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only memory unit (ROM) 6203.

[0087] The storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0088] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0089] Electronic device 600 can also communicate with one or more external devices 100 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although... Figure 6 As not shown in the diagram, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0090] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described in this invention can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this invention can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the method described above according to this invention. When the computer program is executed by a data processing device, it enables the computer-readable medium to implement the method described above, i.e.: as... Figure 1 The steps of the method shown.

[0091] Figure 7 This is a schematic diagram of a computer-readable medium provided for embodiments of this specification.

[0092] accomplish Figure 1 The computer program of the method shown can be stored on one or more computer-readable media. A computer-readable medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0093] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0094] In addition, the present invention also provides a computer program product, including a computer program / instruction that, when executed by a processor, implements a method for generating a three-dimensional topographic map of a tissue region for a forward ultrasound scene in endoscopic surgery as described in any of the above embodiments.

[0095] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0096] In summary, the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that in practice, general-purpose data processing devices such as microprocessors or digital signal processors (DSPs) can be used to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0097] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the present invention is not inherently related to any specific computer, virtual device, or electronic device, and various general-purpose devices can also implement the present invention. The above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0098] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0099] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for generating three-dimensional topographic maps of tissue regions in a forward-facing ultrasound scene during endoscopic surgery, characterized in that, include: The robotic arm is controlled to drive the forward-facing ultrasonic probe to scan along a preset scanning trajectory, continuously acquiring several frames of initial radio frequency signals, and recording the robotic arm parameters corresponding to each frame of the initial radio frequency signal. The initial radio frequency signals of each frame are preprocessed to obtain several frames of standardized radio frequency signals; The tissue identification model is invoked to perform tissue identification on the standardized radio frequency signal to obtain the tissue tag distribution along the depth direction; Based on the distribution of the tissue labels, target tissue regions corresponding to different depth ranges are determined, wherein each target tissue region includes a starting depth and an ending depth; Based on the continuity constraint between each sampling frame, the target tissue regions corresponding to different depth ranges are corrected to obtain the corrected target tissue region sequence. Based on the calibration relationship between the ultrasonic probe and the end effector of the robotic arm, and the parameters of the robotic arm, the probe sampling pose corresponding to each frame of standardized radio frequency signal is determined. Based on the corrected target tissue region sequence and the corresponding probe sampling pose, each target tissue region is mapped to a three-dimensional spatial coordinate system to generate a three-dimensional topographic map of the tissue region used to characterize the spatial distribution of tissue.

2. The method for generating three-dimensional topographic maps of tissue regions for forward ultrasound scenes in endoscopic surgery according to claim 1, characterized in that, The step of correcting the target tissue region corresponding to different depth ranges based on the continuity constraint between each sampling frame includes: Obtain the target tissue region corresponding to adjacent sampling frames; Based on preset continuity constraints, continuity analysis is performed on the target tissue regions corresponding to adjacent sampling frames; If the target tissue region corresponding to the target sampling frame is detected to not meet the preset continuity condition, the target tissue region corresponding to the target sampling frame is corrected according to the target tissue region corresponding to the adjacent sampling frame to obtain the corrected target tissue region sequence.

3. The method for generating three-dimensional topographic maps of tissue regions for forward ultrasound scenes in endoscopic surgery according to claim 2, characterized in that: The continuity constraints include mechanical scanning sequence constraints, tissue depth sequence constraints, tissue label consistency constraints, and confidence weight constraints.

4. The method for generating three-dimensional topographic maps of tissue regions for forward ultrasound scenes in endoscopic surgery according to claim 1, characterized in that, The step of calling the tissue identification model to perform tissue identification on the standardized radio frequency signal to obtain the tissue tag distribution along the depth direction includes: The standardized radio frequency signal is divided into sliding windows to obtain radio frequency signal segments corresponding to each sliding window; The pre-trained tissue recognition model is invoked to identify the tissue labels of each radio frequency signal segment, resulting in a window-level tissue label sequence along the depth direction, and the recognition confidence of the tissue labels is output. Based on the window-level organization tag sequence, the organization tag distribution corresponding to the standardized radio frequency signal of each frame is mapped.

5. The method for generating three-dimensional topographic maps of tissue regions for forward ultrasound scenes in endoscopic surgery according to claim 4, characterized in that, The step of determining the target tissue regions corresponding to different depth ranges based on the tissue tag distribution includes: Adjacent depth regions with the same tissue label are merged to obtain several candidate tissue regions; The organization category corresponding to the candidate organization region is determined based on the preset region identification index, and the target organization region is determined; The target organizational region includes at least two different types of organizational regions.

6. The method for generating a three-dimensional topographic map of a tissue region for a forward ultrasound scene in endoscopic surgery according to any one of claims 1-5, characterized in that, After generating a three-dimensional topographic map of the tissue region to characterize the spatial distribution of the tissue, the following steps are also included: Based on the three-dimensional topographic map, extract profile information along the scanning trajectory or a specified path; Based on the profile information, calculate the distance variation, thickness variation, and spatial undulation variation information between different target tissue regions.

7. A system for generating three-dimensional topographic maps of tissue regions in a forward-facing ultrasound scene during endoscopic surgery, characterized in that, The tissue region three-dimensional topographic map generation system for forward ultrasound scenes in endoscopic surgery includes: The signal acquisition module is used to control the robotic arm to drive the forward ultrasonic probe to scan along a preset scanning trajectory, continuously acquire several frames of initial radio frequency signals, and record the robotic arm parameters corresponding to each frame of the initial radio frequency signal. The preprocessing module is used to preprocess the initial radio frequency signals of each frame to obtain several frames of standardized radio frequency signals; The tissue identification module is used to call the tissue identification model to perform tissue identification on the standardized radio frequency signal and obtain the tissue tag distribution along the depth direction; The region division module is used to determine target tissue regions corresponding to different depth ranges based on the distribution of tissue labels, wherein each target tissue region includes a start depth and an end depth; The correction module is used to correct the target tissue regions corresponding to different depth ranges based on the continuity constraints between each sampling frame, so as to obtain the corrected target tissue region sequence. The pose calculation module is used to determine the probe sampling pose corresponding to each frame of standardized radio frequency signal based on the calibration relationship between the ultrasonic probe and the end effector of the robotic arm and the parameters of the robotic arm. The topographic map generation module is used to map each target tissue region to a three-dimensional spatial coordinate system based on the corrected target tissue region sequence and the corresponding probe sampling pose, and generate a three-dimensional topographic map of the tissue region to characterize the spatial distribution of the tissue.

8. A device for generating three-dimensional topographic maps of tissue regions in a forward-facing ultrasound scene during endoscopic surgery, characterized in that, The device for generating three-dimensional topographic maps of tissue regions for forward ultrasound scenes in endoscopic surgery includes: a memory and at least one processor, wherein the memory stores instructions. The at least one processor invokes the instructions in the memory to cause the tissue region three-dimensional topographic map generation device for a forward ultrasound scene in endoscopic surgery to perform the steps of the tissue region three-dimensional topographic map generation method for a forward ultrasound scene in endoscopic surgery as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program / instructions thereon, characterized in that, When the program / instruction is executed by the processor, it implements the steps of the method for generating a three-dimensional topographic map of a tissue region for a forward ultrasound scene in endoscopic surgery as described in any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method for generating a three-dimensional topographic map of a tissue region for a forward ultrasound scene in endoscopic surgery as described in any one of claims 1-6 are implemented.