Ultrasound scan occlusion avoidance control method and related apparatus

Abstract

The application provides an ultrasonic scanning shelter avoidance control method and related device, belonging to the technical field of ultrasonic contrast imaging. The method comprises: acquiring depth image data and first ultrasonic data of a user tissue region specified by a doctor, determining a target scanning tissue of the user according to the depth image data and the first ultrasonic data, determining a shelter type and shelter identification information corresponding to the shelter type when performing ultrasonic scanning according to the target scanning tissue and the first ultrasonic data, wherein the shelter type comprises a bone part shelter and an intestinal gas shelter, determining a pose adjustment parameter of a mechanical arm according to the shelter type and the shelter identification information and the first ultrasonic data, controlling the mechanical arm to perform pose adjustment according to the pose adjustment parameter, so that the mechanical arm moves to a target scanning position, and the target scanning position is scanned by an ultrasonic probe to collect target ultrasonic data of the user tissue region. In this way, the accuracy and reliability of ultrasonic scanning can be improved.

Description

Technical Field

[0001] This application relates to the field of ultrasound contrast imaging technology, and in particular to an ultrasound scanning obstruction avoidance control method and related device. Background Technology

[0002] In current ultrasound scanning procedures, ultrasound scans are performed directly along a pre-set scanning trajectory. However, due to obstruction from the patient's bones or intestinal gas, the accuracy of the scan may not meet the standards for ultrasound diagnosis. To improve the accuracy of ultrasound scans, methods include using image restoration or interpolation algorithms to generate corresponding ultrasound images for obstructed areas; and optimizing the ultrasound scanning area by using the signal-to-noise ratio of the ultrasound image as the optimization objective, thereby driving the ultrasound probe to avoid obstacles during the ultrasound scan.

[0003] However, due to the specificity between different individuals, ultrasound scanning based on a pre-set trajectory may not be fully adapted to the body shape of clinical patients, resulting in inaccurate scanning results; the use of image restoration algorithms to generate ultrasound images may falsify pathological features, thereby affecting the doctor's diagnosis; and using the image signal-to-noise ratio as the optimization target to drive the ultrasound probe for ultrasound scanning, the optimization algorithm cannot effectively distinguish between overall image blurring and structural loss caused by local occlusion, resulting in low scanning efficiency.

[0004] Therefore, improving the efficiency and accuracy of ultrasound scans is an urgent issue that needs to be addressed. Summary of the Invention

[0005] This application provides an ultrasound scanning occlusion avoidance control method and related device, which aims to solve the problems of low scanning efficiency caused by bone / intestinal gas occlusion in real ultrasound scanning tasks, and the insufficient accuracy of ultrasound scanning caused by the use of image restoration algorithms or interpolation algorithms to generate ultrasound images in current ultrasound scanning.

[0006] To achieve the objectives of this application, the following technical solution is provided: In a first aspect, this application provides an ultrasonic scanning obstruction avoidance control method, applied to the host device of an ultrasonic scanning system, the method comprising: Acquire depth image data and first ultrasound data of a user tissue region specified by the doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes changes in the surface curvature of the user tissue region. The user's target tissue to be scanned is determined based on the depth image data and the first ultrasound data; the occlusion type and the corresponding occlusion identification information during ultrasound scanning are determined based on the target tissue to be scanned and the first ultrasound data; the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning; the pose adjustment parameters of the robotic arm are determined based on the occlusion type, the occlusion identification information, and the first ultrasound data; The robotic arm is controlled to adjust its posture according to the posture adjustment parameters so that it moves to the target scanning position; and the target scanning position is scanned by an ultrasonic probe to collect target ultrasound data of the user's tissue area.

[0007] Secondly, this application provides an ultrasonic scanning obstruction avoidance control device, applied to the main equipment of an ultrasonic scanning system, the device comprising: The acquisition unit is used to acquire depth image data and first ultrasound data of a user tissue region specified by a doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes changes in the surface curvature of the user tissue region. The processing unit is configured to determine the user's target tissue for scanning based on the depth image data and the first ultrasound data; and to determine the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data; wherein the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning. The calculation unit is used to determine the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data. The control unit is used to control the robotic arm to adjust its posture according to the posture adjustment parameters so that the robotic arm moves to the target scanning position; and to scan the target scanning position with an ultrasonic probe to collect target ultrasonic data of the user's tissue area.

[0008] Thirdly, this application provides a host device including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing steps in any of the methods of the first aspect of the embodiments of this application.

[0009] Fourthly, this application provides a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in any of the methods of the first aspect of the embodiments of this application.

[0010] Fifthly, this application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, the computer program being operable to cause a computer to perform some or all of the steps described in any method of the first aspect of the embodiments of this application, and the computer program product may be a software installation package.

[0011] By implementing the embodiments of this application, the following beneficial effects are achieved: This application provides an ultrasound scanning occlusion avoidance control method and related device, applied to the host device of an ultrasound scanning system. First, it acquires depth image data and first ultrasound data of a user tissue region specified by a doctor. Second, it determines the user's target scanning tissue based on the depth image data and the first ultrasound data. Then, it determines the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target scanning tissue and the first ultrasound data. The occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning. Next, it determines the pose adjustment parameters of a robotic arm based on the occlusion type, occlusion identification information, and the first ultrasound data. The robotic arm is then controlled to adjust its pose according to the pose adjustment parameters, moving to the target scanning position. Finally, an ultrasound probe scans the target scanning position to collect target ultrasound data of the user tissue region. Thus, by determining the occlusion type and corresponding occlusion marker information during ultrasound scanning based on the target tissue and the first ultrasound data, the robotic arm can adjust its posture according to the occlusion type, thereby actively avoiding occluded objects and optimizing its posture to improve scanning efficiency. Furthermore, by controlling the robotic arm to adjust the posture of the ultrasound probe, the probe actively avoids areas obstructed by bone and intestinal gas, moving to the target scanning position and maintaining stable scanning to obtain target ultrasound data corresponding to the target user tissue area. This directly acquires raw ultrasound data based on real acoustic echo information, avoiding the introduction of false pathological features and thus improving the accuracy of ultrasound scanning. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is an architectural diagram of an ultrasonic scanning system provided in an embodiment of this application; Figure 2This is a schematic diagram of the structure of a host device provided in an embodiment of this application; Figure 3 This is a flowchart illustrating an ultrasonic scanning obstruction avoidance control method provided in an embodiment of this application; Figure 4 This is a schematic diagram of a scenario where ultrasound scanning detects bone occlusion, provided in an embodiment of this application. Figure 5 This is a schematic diagram of an intestinal gas obstruction scenario detected by ultrasound scanning, provided in an embodiment of this application. Figure 6 This is a flowchart illustrating another ultrasonic scanning obstruction avoidance control method provided in an embodiment of this application; Figure 7 This is a block diagram of the functional modules of an ultrasonic scanning obstruction avoidance control device provided in an embodiment of this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0015] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0016] It should be understood that the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document indicates that the preceding and following related objects are in an "or" relationship. In the embodiments of this application, "multiple" refers to two or more.

[0017] In the embodiments of this application, "at least one item" or its similar expression refers to any combination of these items, including any combination of a single item or a plurality of items. "One or more" means one or more, while "multiple" means two or more. For example, "at least one item" of a, b, or c can represent the following seven cases: a, b, c; a and b; a and c; b and c; a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0018] In this application, the term "connection" refers to various connection methods, such as direct connection or indirect connection, to achieve communication between devices. This application does not impose any limitations on this.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] Please see Figure 1 , Figure 1 This is an architecture diagram of an ultrasonic scanning system provided in an embodiment of this application. The ultrasonic scanning system 100 includes an end-effector and sensing module 110, an edge control module 120, and a server cluster 130.

[0021] The end effector and sensing module 110 is used for spatial perception, ultrasonic scanning, and ultrasonic data acquisition of the user's target tissue area. Specifically, the end effector and sensing module 110 includes a depth camera 111, a robotic arm 112, and an ultrasonic probe 113. The depth camera 111 is used to acquire depth image data corresponding to the user's target tissue area to obtain spatial contour information, spatial posture information, and spatial position information of the tissue area on the user's body surface. The depth camera 111 can be a structured light depth camera, a binocular vision depth camera, or a ToF depth camera, used to generate depth image data and three-dimensional point cloud data corresponding to the target tissue area, thereby providing spatial reference for the subsequent pose planning of the robotic arm 112 and the scanning path planning of the ultrasonic probe 113. The robotic arm 112 is used to carry the ultrasonic probe 113 and drive the ultrasonic probe 113 to perform pose adjustment and scanning motion control of the target tissue area. Specifically, the robotic arm 112 can be a six-degree-of-freedom (6DOF) or a seven-degree-of-freedom (7DOF) flexible robotic arm. Its end effector is connected to the ultrasonic probe 113, and it is used to perform spatial position adjustment, attitude adjustment, and trajectory motion control according to the pose adjustment parameters generated by the edge control module 120, so as to realize the automated scanning of the target tissue area by the ultrasonic probe 113. The ultrasonic probe 113 is used to emit ultrasonic signals to the user's tissue area and receive the ultrasonic echo signals corresponding to the target tissue area to generate first ultrasonic data and target ultrasonic data. Specifically, the ultrasonic probe 113 can include a linear array probe, a convex array probe, or a phased array probe, and its output data includes ultrasonic image data and ultrasonic echo data.

[0022] In an optional embodiment, the depth camera 111 in the end effector and sensing module 110 is positioned near the end of the robotic arm 112 and forms a collaborative sensing structure with the ultrasound probe 113, enabling the depth camera 111 to acquire the spatial state of the body surface region corresponding to the ultrasound probe 113 in real time. Furthermore, the end of the robotic arm 112 may also be provided with a flexible buffer execution structure to reduce the rigid impact between the ultrasound probe 113 and the user's body surface during the movement of the robotic arm 112, and to improve the adhesion stability between the ultrasound probe 113 and the target tissue region.

[0023] The edge control module 120 is used for real-time edge-side control, motion control, and ultrasound data analysis and processing during the ultrasound scanning process. The edge control module 120 includes an ultrasound host 121, a robotic arm controller 122, and an edge computing industrial control computer 123. The ultrasound host 121 performs beamforming, image reconstruction, and ultrasound image generation processing on the ultrasound echo data acquired by the ultrasound probe 113 to output corresponding first ultrasound data and target ultrasound data. Furthermore, the ultrasound host 121 also performs ultrasound image enhancement, echo filtering, and tissue structure feature extraction to improve the recognizability of the ultrasound images. The robotic arm controller 122 controls the robotic arm 112 to perform pose and motion control. The robotic arm controller 122 receives pose adjustment parameters and motion trajectory parameters generated by the edge computing industrial control computer 123, and generates joint drive control parameters based on the kinematic model corresponding to the robotic arm 112 to drive the robotic arm 112 to perform spatial displacement motion, posture adjustment motion, and scanning trajectory motion. The edge computing industrial control computer 123 is used to perform fusion analysis on depth image data, first ultrasound data, and target ultrasound data, and generate pose adjustment parameters and scanning control strategies corresponding to ultrasound scanning. Specifically, the edge computing industrial control computer 123 is used to perform processes such as occlusion region identification, occlusion type analysis, scanning quality assessment, and reinforcement learning pose optimization. For example, the edge computing industrial control computer 123 determines the target tissue to be scanned based on the depth image data and the first ultrasound data; further, it determines a first comprehensive quality score based on the occlusion region distribution state, echo continuity state, and tissue structure integrity state corresponding to the target tissue; subsequently, it generates a target pose adjustment action strategy corresponding to the robotic arm 112 through a reinforcement learning algorithm, thereby driving the robotic arm 112 to perform occlusion avoidance scanning.

[0024] In an optional embodiment, the edge control module 120 and the end effector and sensing module 110 exchange data via 5.5G Ethernet communication. Depth image data acquired by the depth camera 111 and ultrasonic data acquired by the ultrasonic probe 113 are transmitted in real-time to the edge computing industrial control computer 123 via 5.5G Ethernet communication to achieve low-latency, high-bandwidth data transmission. Simultaneously, the robotic arm pose adjustment parameters and scanning trajectory parameters generated by the edge computing industrial control computer 123 are sent in real-time to the robotic arm controller 122 via 5.5G Ethernet communication, thereby ensuring the real-time performance and stability of the robotic arm 112's pose control.

[0025] The server cluster 130 is used for cloud-based collaborative processing of model training tasks, historical data management tasks, and reinforcement learning optimization tasks in the ultrasound scanning system 100. The server cluster 130 is used to store historical ultrasound scanning data, historical occlusion type data, and historical pose adjustment strategy data, and to train and optimize the occlusion recognition model, scanning quality assessment model, and reinforcement learning strategy model based on a large amount of historical scanning data.

[0026] As can be seen, the end-effector and sensing module 110 is used for spatial perception and ultrasound data acquisition of the target tissue area, the edge control module 120 is responsible for real-time data analysis and robotic arm motion control during the ultrasound scanning process, and the server cluster 130 is responsible for model training and strategy optimization. Furthermore, the depth camera 111 and the ultrasound probe 113 form a multimodal collaborative sensing structure, enabling the ultrasound scanning system 100 to simultaneously acquire spatial structure information and ultrasound tissue information corresponding to the target tissue area; the edge computing industrial control computer 123 achieves real-time identification of bone occlusion and intestinal gas occlusion by fusing depth image data and ultrasound data. The robotic arm controller 122 combines reinforcement learning pose optimization strategies to achieve adaptive pose adjustment and occlusion avoidance control of the robotic arm 112; the server cluster 130 continuously optimizes the scanning strategy model through training with historical data. In this way, it can achieve adaptive optimization of the ultrasound scanning path in scenarios with skeletal occlusion and intestinal gas occlusion, improve the effective imaging coverage of the target tissue area and the quality of ultrasound images, reduce the problem of insufficient adaptability of traditional fixed trajectory scanning methods to complex human body structure scenarios, and at the same time reduce the risk of generating fake pathological features based on image restoration algorithms, thereby improving the authenticity, accuracy and stability of ultrasound diagnostic results.

[0027] The following is combined with Figure 2 The host device in the embodiments of this application will be described. Figure 2 This is a schematic diagram of the structure of a host device provided in an embodiment of this application, such as... Figure 2 As shown, the host device 200 includes a processor 210, a memory 220, a communication interface 230, and one or more programs 221. The processor 210 is communicatively connected to the memory 220 and the communication interface 230 through an internal communication bus.

[0028] The one or more programs 221 are stored in the memory 220 and configured to be executed by the processor 210. The one or more programs 221 include instructions for performing any step in the above method embodiments.

[0029] The processor 210 can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, units, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication unit can be a communication interface, transceiver, transceiver circuit, etc., and the storage unit can be a memory.

[0030] The memory 220 can be volatile memory or non-volatile memory, or it can include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0031] It is understood that the host device 200 may include more or fewer structural elements than those shown in the above block diagram, such as a power module, physical buttons, a Wi-Fi module, a speaker, a Bluetooth module, sensors, a display module, etc., without limitation. It is understood that the host device may be equipped with... Figure 1 The architecture of the ultrasonic scanning system.

[0032] After understanding the software and hardware architecture of this application, the following will be combined with... Figure 3 This application describes an ultrasonic scanning obstruction avoidance control method in its embodiments. Figure 3 This is a flowchart illustrating an ultrasonic scanning obstruction avoidance control method provided in an embodiment of this application. The method is applied to the host device in an ultrasonic scanning system, and specifically includes the following steps: Step S310: Obtain depth image data and first ultrasound data of the user tissue region specified by the doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes the surface curvature changes of the user tissue region.

[0033] In this embodiment, the ultrasound scanning system includes an ultrasound robot, a camera module, and an ultrasound probe mounted at the end of a robotic arm, all connected to a host device. The camera module is used to acquire depth image data of the user's tissue region, and the ultrasound probe is used to acquire first ultrasound data of the corresponding region. The user tissue region specified by the doctor can be the abdomen, thyroid region, breast region, musculoskeletal region, or other tissue regions to be diagnosed, which is mainly used to define the target anatomical region corresponding to the current ultrasound scanning task.

[0034] In a specific embodiment, the doctor inputs information about the user's tissue region to be scanned via a terminal device. The host device controls a robotic arm based on this information to move the ultrasound probe to the corresponding initial scanning position. The ultrasound probe performs an ultrasound scan of the tissue region at the initial scanning position and receives ultrasound echo signals reflected from within the tissue in real time. Subsequently, the host device performs beamforming, envelope detection, and grayscale mapping on the ultrasound echo signals to generate corresponding first ultrasound data. Furthermore, the host device can also mark low-signal-noise regions, strong-reflection regions, and echo-missing regions in the first ultrasound data to provide a data basis for subsequent occlusion type identification. Simultaneously, the camera module performs continuous depth scanning of the user's tissue area at a preset sampling frequency, and performs spatial filtering on the raw depth images acquired by the camera module to eliminate the influence of environmental noise, random speckle noise, and edge burrs on the depth measurement results. Then, outlier removal processing is performed on the filtered depth images to remove outlier data caused by metal reflection, body surface occlusion, or changes in ambient light. Furthermore, multiple frames of depth images acquired at continuous times are fused to generate stable depth image data of the target tissue area.

[0035] It should be noted that depth image data refers to three-dimensional perceptual data used to characterize the spatial geometry of a user's tissue region. It includes not only two-dimensional image texture information of the target region but also spatial distance values ​​corresponding to each pixel, thus forming three-dimensional point cloud data corresponding to the user's tissue region. Depth image data is mainly used to characterize the user's body surface contour, tissue undulations, and the spatial pose relationship of the target region. First ultrasound data refers to the raw ultrasound echo data obtained by the ultrasound probe scanning the user's tissue region in the initial scanning state. This data can include B-mode ultrasound image data, echo intensity data, tissue edge response data, and corresponding ultrasound echo timing information. Furthermore, the end effector of the robotic arm in the ultrasound robot can be equipped with a flexible actuator. This flexible actuator connects to the ultrasound probe and provides flexible buffering and posture compensation capabilities during the contact between the ultrasound probe and the user's body surface. This reduces contact deviations caused by user breathing fluctuations, changes in body surface curvature, or slight movements, ensuring a stable coupling state between the ultrasound probe and the user's tissue region, thereby improving the imaging stability during the first ultrasound data acquisition process. In addition, the depth image data and the first ultrasound data maintain a synchronous correspondence in the time dimension.

[0036] Step S320: Determine the user's target scan tissue based on the depth image data and the first ultrasound data.

[0037] In this embodiment, the target scanned tissue refers to the target tissue structure that the physician needs to focus on observing or analyzing for lesions in the current ultrasound diagnostic task. This can be liver tissue, gallbladder tissue, pancreatic tissue, thyroid tissue, breast tissue, muscle tissue, vascular tissue, or other anatomical tissues to be diagnosed. Because different tissues differ in spatial location, morphology, and ultrasound echo characteristics, the host device needs to identify and locate the target scanned tissue in the current user tissue region based on the fusion analysis results of depth image data and first ultrasound data. The depth image data is mainly used to provide surface spatial structure information corresponding to the user tissue region, including the three-dimensional contour of the target region, changes in surface curvature, spatial boundary positions, and spatial coordinate relationships corresponding to the tissue region. The first ultrasound data is mainly used to characterize the ultrasound echo distribution state within the tissue, including tissue edge echo characteristics, tissue texture characteristics, internal echo intensity distribution, and tissue structural continuity information. Furthermore, the host device can also pre-store a standard human anatomy model database, which includes tissue spatial distribution models, tissue ultrasound texture feature models, and tissue anatomical location correlation models corresponding to different human body parts. By matching the currently acquired depth image data with the standard human anatomy model, the host device can initially determine the range of possible tissue structures in the current user's tissue area. Subsequently, by combining the tissue echo features in the first ultrasound data, the candidate tissue structures are further screened to determine the final target tissue to be scanned.

[0038] In a specific embodiment, the three-dimensional point cloud in the depth image data is first segmented to extract the target spatial region corresponding to the user's tissue region specified by the doctor. Secondly, based on the surface curvature change information and spatial boundary contour information corresponding to the target spatial region, the positional relationship of human structures within the target region is determined. For example, in an abdominal ultrasound scan scenario, the host device can preliminarily determine the approximate spatial range of the liver, gallbladder, and intestinal regions based on the surface contour of the abdominal region, the spatial undulation changes of the rib region, and the positional distribution of the central abdominal region. In a thyroid ultrasound scan scenario, the thyroid region can be preliminarily located based on the spatial contour features of the neck region and the neck curvature distribution. Then, the host device extracts the ultrasound grayscale distribution, tissue boundary continuity, tissue texture features, and echo attenuation features from the first ultrasound data and generates corresponding tissue feature vectors. These tissue feature vectors are then matched with standard tissue ultrasound features in a standard human anatomical model database to determine the corresponding tissue type in the current ultrasound image. After determining the tissue type, the spatial coordinate information in the depth image data is combined to spatially map the tissue region in the ultrasound image, thereby establishing the corresponding positional relationship of the target tissue in the user's body surface space. For example, the host device can map the liver boundary region in the ultrasound image to the corresponding right upper quadrant space region in the depth image, thereby determining the location of the target scanned tissue corresponding to the liver tissue.

[0039] It can be seen that by fusing the spatial structure information in the depth image data and the tissue echo information in the first ultrasound data, accurate identification and spatial positioning of the user's target tissue can be achieved. This enables the acquisition of the internal ultrasound structural features of the tissue and the establishment of a correspondence between the target tissue and the user's body surface space, thereby improving the accuracy of tissue identification and the stability of the ultrasound scan.

[0040] Step S330: Determine the occlusion type and the corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data; the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning.

[0041] In this embodiment, the type of obstruction refers to abnormal tissue structures that interfere with, reflect, attenuate, or block the propagation path of ultrasound waves during ultrasound scanning, thereby reducing the imaging quality of the target tissue. Skeletal obstruction mainly refers to the obstruction phenomenon caused by the ribs, clavicle, spine, or other high acoustic impedance bony structures during ultrasound propagation. Because bone tissue has strong ultrasound reflection capabilities, it typically appears as a bright, strong echo area and a large area of ​​acoustic shadowing behind it in the first ultrasound data, thus preventing effective imaging of some structures of the target tissue. Intestinal gas obstruction mainly refers to the abnormal ultrasound propagation phenomenon caused by intestinal bubbles or gas accumulation areas during ultrasound propagation. Due to the significant difference in acoustic impedance between the gas medium and human soft tissue, ultrasound waves are prone to scattering, refraction, and multipath reflection during propagation, resulting in echo disorder areas, localized high-noise areas, or tissue boundary distortion areas in the first ultrasound data. Occlusion identification information is structured parameter information used to characterize the spatial attributes and occlusion features of the current occluded area. It may include spatial location parameters corresponding to the occluded area, boundary contour parameters of the occluded area, occlusion area ratio parameters, occlusion intensity parameters, occlusion duration parameters, and the corresponding occlusion type number.

[0042] In a specific embodiment, firstly, tissue boundary extraction processing is performed on the target tissue region in the first ultrasound data to generate a corresponding tissue boundary contour image. Abnormal echo regions in the tissue boundary contour image are then detected to identify candidate occlusion regions with abnormal echo attenuation, missing local structures, or disordered echoes. Next, the ultrasound echo features in the candidate occlusion regions are analyzed. Specifically, echo grayscale distribution features, echo texture continuity features, acoustic shadow extension features, local echo attenuation features, and tissue boundary distortion features are extracted from the candidate regions, and corresponding occlusion feature vectors are generated. When a candidate region contains continuous bright echo areas, and a large area of ​​low-grayscale acoustic shadowing is present behind these bright areas, it is determined that the current candidate region corresponds to a skeletal occlusion. Then, by combining the surface curvature change information in the depth image data and the human bony structure distribution model, the specific bone location corresponding to the occluded area can be further confirmed. When a candidate region exhibits local echo disorder, tissue texture breakage, multiple artifacts, or abnormal enhancement of local speckle patterns, it is determined that the current candidate region corresponds to intestinal gas occlusion. Then, by combining the dynamic echo changes in continuous frame ultrasound data, the dynamic diffusion state and gas accumulation degree corresponding to the gas region can be identified. After identifying the occlusion type, corresponding occlusion identification information is generated based on the spatial location and tissue boundary relationship of the candidate occluded area. Specifically, the two-dimensional coordinates of the occluded area in the ultrasound image are obtained, and combined with the spatial mapping relationship in the depth image data, the two-dimensional occluded area is mapped to the three-dimensional spatial coordinate position corresponding to the user's body surface. Finally, the proportion of the occluded area to the target scanned tissue area is calculated to generate the corresponding occlusion area proportion parameter.

[0043] For easier understanding, please refer to Figure 4 , Figure 4 This is a schematic diagram of an ultrasonic scan detecting a scenario where bone obstruction is observed, provided in an embodiment of this application. As can be seen, the ultrasonic probe is located on one side of the user's body surface and emits ultrasonic signals towards the target tissue area. During the ultrasonic wave propagation path, due to the presence of bone structures in front of the target tissue area, the ultrasonic waves undergo strong reflection when they reach the bone, thus forming a distinct acoustic shadow area behind the bone. Figure 4In the image, the bright areas correspond to areas obscured by bone, while the black areas correspond to acoustic shadowing areas caused by bone reflection. Because bone tissue has high acoustic impedance, ultrasound waves are strongly reflected when they reach the bone interface, preventing them from effectively penetrating the bone structure and resulting in a lack of effective echo signals in the area behind the bone. Specifically, the ultrasound probe emits a fan-shaped beam of ultrasound waves towards the target tissue area. In normal soft tissue areas, this beam can propagate continuously and produce stable echo signals; however, when ultrasound waves propagate to the bone structure, due to the high reflectivity of the corresponding bone region, most of the ultrasound energy is reflected at the bone surface, with only a small amount able to continue propagating into deeper tissues. As a result, the ultrasound echo energy corresponding to the area behind the bone is significantly reduced, thus forming a large area of ​​low-grayscale or echo-free acoustic shadowing in the ultrasound image. Furthermore, Figure 4 The highlighted areas in the ultrasound scan represent the high-echo regions corresponding to bone, with gray values ​​significantly higher than those of the surrounding soft tissue. The acoustic shadowing regions, on the other hand, appear as continuous, large areas of low gray values ​​extending backward along the direction of ultrasound propagation. Because the tissue structure information in the acoustic shadowing regions cannot be effectively imaged, it can lead to missing tissue boundaries, discontinuous tissue texture, and unrecognizable local structures within the target tissue area, thus affecting the accuracy of ultrasound scanning. By identifying the high-echo regions and acoustic shadowing regions of bone and dynamically adjusting the scanning pose of the ultrasound probe, adaptive avoidance of bone-occluded areas is achieved, thereby improving the effective imaging coverage of the target tissue area and the quality of the ultrasound image.

[0044] Similarly, please see Figure 5 , Figure 5 This is a schematic diagram illustrating an ultrasound scan scenario where intestinal gas obstruction is detected, provided in an embodiment of this application. As can be seen, the ultrasound probe is positioned above the user's body surface and emits ultrasound signals towards the target tissue area. During ultrasound propagation, due to the presence of intestinal gas around the target tissue area, the ultrasound waves undergo significant scattering and reflection upon reaching the gas region, resulting in edge distortion, texture disorder, and local structural blurring in the ultrasound image corresponding to the target tissue area. Figure 5In the diagram, the "intestinal gas obstruction" region corresponds to the area where a gaseous medium exists during ultrasound propagation, while the "gas obstruction causing image edge distortion" region corresponds to the area of ​​tissue structure distortion caused by gas scattering interference. Specifically, the ultrasound probe emits a fan-shaped diffused ultrasound beam towards the target tissue area. When ultrasound propagates in normal soft tissue, its propagation path is relatively stable, forming a continuous and regular echo signal. However, when ultrasound propagates to the intestinal gas region, due to the significant acoustic impedance difference between the gas and human soft tissue, strong scattering, multipath reflection, and local random refraction occur at the gas boundary. As a result, some ultrasound waves cannot continue to propagate along their original path, leading to a decrease in the stability of the echo signal in the target tissue area. Furthermore, because intestinal gas has dynamic characteristics, the spatial distribution of the gas region is usually irregularly diffused. Figure 5 In ultrasound scanning, the intestinal gas-occupied area is located in front of the target tissue region. The corresponding ultrasound propagation path is interfered with by the gas medium, resulting in numerous irregular speckle textures and localized aliasing echoes, thus distorting the edge contour of the target tissue region. Specifically, this manifests as blurred boundaries, decreased edge continuity, enhanced random fluctuations in local textures, and loss of tissue structural details. Unlike skeletal occlusion, which forms continuous acoustic shadowing areas, intestinal gas occlusion typically presents as localized texture disorder and boundary distortion, exhibiting strong randomness and dynamic change characteristics. By detecting high-frequency texture disturbances, enhanced localized random speckle patterns, and abrupt boundary gradient changes in the target tissue region, the presence of intestinal gas occlusion during the current ultrasound scan is confirmed. Furthermore, by analyzing the spatial distribution range and echo scattering characteristics of the gas-occupied area, corresponding occlusion identification information is generated. Then, based on the occlusion identification information corresponding to intestinal gas occlusion, the scanning pose and trajectory of the robotic arm are dynamically adjusted. For example, by controlling a robotic arm to perform a localized downward pressure motion on the ultrasound probe, interference from intestinal gas on the ultrasound propagation path can be reduced; or, by adjusting the tilt angle and scanning direction of the ultrasound probe, the incident path of the ultrasound waves can be altered, thereby bypassing localized gas accumulation areas and improving the effective imaging range of the target tissue area. By identifying textured and distorted boundary areas corresponding to intestinal gas and combining this with dynamic adjustment of the robotic arm's pose, adaptive avoidance of gas-obstructed areas is achieved, thereby improving the tissue boundary integrity, echo continuity, and ultrasound image quality of the target tissue area, and enhancing the accuracy and stability of ultrasound scanning in complex tissue scenarios.

[0045] It can be seen that by combining the target tissue and the first ultrasound data, the abnormal echo areas present during the ultrasound scanning process can be structurally analyzed, thereby achieving accurate identification of skeletal obstruction and intestinal gas obstruction, and generating corresponding obstruction identification information. This enables clear differentiation of different obstruction sources and their spatial distribution, thereby improving the integrity of tissue imaging and diagnostic accuracy during the ultrasound scanning process.

[0046] In an optional embodiment, the first ultrasound data includes first ultrasound image data and first ultrasound echo data; the step of determining the occlusion type and the corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data specifically includes the following steps: 331. Extract the ultrasound image data corresponding to the target scanned tissue from the first ultrasound image data to obtain target ultrasound image data; and extract the ultrasound echo data corresponding to the target scanned tissue from the first ultrasound echo data to obtain target ultrasound echo data. 332. Divide the target ultrasound image data into image regions to obtain multiple ultrasound image blocks; 333. Determine the grayscale feature parameters corresponding to each of the plurality of ultrasound image blocks to obtain a plurality of grayscale feature parameters; 334. Based on the multiple grayscale feature parameters, perform occlusion recognition on the multiple ultrasound image blocks to obtain occlusion recognition results; 335. Determine multiple spatial connectivity states corresponding to the multiple ultrasound image blocks based on the occlusion recognition results; and determine the occlusion regions corresponding to the multiple ultrasound image blocks based on the multiple spatial connectivity states to obtain the target occlusion region; 336. Determine the target echo characteristics corresponding to the target obstruction area in the target ultrasonic echo data; 337. Determine the occlusion type of the target occlusion region based on the target echo characteristics; and determine the occlusion identification information of the target occlusion region based on the target echo characteristics and the plurality of spatial connectivity states.

[0047] In this embodiment, the first ultrasound image data refers to a two-dimensional ultrasound image formed by the ultrasound probe based on the ultrasound echo signal through beamforming, envelope detection, and grayscale mapping. It is mainly used to characterize the tissue morphology, tissue boundaries, and local grayscale distribution of the target scanned tissue. The first ultrasound echo data refers to the original ultrasound echo signal data received by the ultrasound probe, which includes acoustic propagation information of different media interfaces within the target tissue, including echo amplitude, echo energy, echo attenuation characteristics, and echo timing characteristics. An ultrasound image block refers to a local image region formed after dividing the target ultrasound image data into regions; each ultrasound image block corresponds to a local tissue region within the target scanned tissue. Grayscale feature parameters are image feature parameters used to characterize the grayscale distribution of ultrasound image blocks, and may include average grayscale value, grayscale variance, grayscale gradient change rate, local texture continuity parameters, and grayscale entropy value. Spatial connectivity state is a structural state parameter used to characterize the spatial continuity relationship between multiple ultrasound image blocks, mainly used to determine whether different abnormal image blocks belong to the same occlusion region. Target echo characteristics refer to the ultrasonic echo propagation characteristics corresponding to the target occlusion area, which may include echo energy distribution characteristics, echo attenuation rate characteristics, echo spectrum distribution characteristics, and echo phase change characteristics. Occlusion identification information is a set of data that provides a structured description of the spatial location, occlusion area, occlusion intensity, and occlusion type of the target occlusion area.

[0048] In a specific embodiment, firstly, tissue region extraction processing is performed on the first ultrasound image data to obtain the target ultrasound image data corresponding to the target scanned tissue. Simultaneously, the tissue echo signal in the first ultrasound echo data is synchronously mapped to extract the target ultrasound echo data corresponding to the target scanned tissue. Next, image region segmentation processing is performed on the target ultrasound image data, and the gray-level distribution state in each ultrasound image block is analyzed to extract the corresponding gray-level feature parameters. Optionally, the target ultrasound image is divided into grids based on a preset window size, so that the target ultrasound image is divided into multiple continuously arranged ultrasound image blocks. For example, by statistically analyzing the mean and variance of pixel gray levels in the image blocks, the echo intensity variation state of the current region is determined; by extracting the local texture gradient change rate, the presence of abrupt changes or breaks in the tissue boundary is analyzed; and by calculating the gray-level entropy value, the complexity of the texture within the current region is determined. Since skeletal occlusion usually manifests as a local strong bright area and a large area of ​​low gray-level shadow behind it, while intestinal gas occlusion usually manifests as texture disorder, enhanced random gray-level fluctuations, and abnormal local speckle, different occlusion regions have significant differences in gray-level feature parameters. Then, occlusion identification is performed on multiple ultrasound image blocks based on multiple grayscale feature parameters. A pre-defined occlusion identification model is used to match and analyze the grayscale feature parameters in each ultrasound image block to identify whether the current image block has occlusion anomalies. When a high grayscale abrupt change is present in an image block accompanied by continuous low grayscale regions, the current image block is determined to have a risk of skeletal occlusion; when a high-frequency texture perturbation and local random speckle enhancement are present in an image block, the current image block is determined to have a risk of intestinal gas occlusion. After completing the image block-level occlusion identification, the spatial adjacency relationship between multiple abnormal image blocks is further analyzed, and corresponding spatial connectivity states are established based on the positional continuity between image blocks. For example, when multiple abnormal image blocks are distributed in a continuous band in the image space, they are determined to belong to the same skeletal occlusion region; when multiple abnormal image blocks are discretely clustered and have a dynamic diffusion trend, they are determined to belong to an intestinal gas occlusion region. Based on multiple spatial connectivity states, region fusion processing is performed on multiple abnormal image blocks to determine the target occlusion region. Finally, the target echo features corresponding to the target occlusion region are extracted from the target ultrasound echo data. Furthermore, by combining target echo characteristics and multiple spatial connectivity states, corresponding occlusion identification information is generated. Specifically, occlusion area coordinate parameters are generated based on the spatial distribution range of the target occlusion area, occlusion intensity parameters are generated based on the degree of change in target echo energy, and corresponding occlusion area boundary parameters and occlusion area proportion parameters are generated by combining the occlusion area area and spatial connectivity state, thereby forming complete occlusion identification information.

[0049] It can be seen that by combining ultrasound image grayscale feature analysis, spatial connectivity analysis, and ultrasound echo feature analysis, accurate identification of skeletal occlusion and intestinal gas occlusion was achieved. Furthermore, by combining image structural features and acoustic propagation features, the accuracy and robustness of occlusion type determination were improved, thus providing a reliable data foundation for subsequent robotic arm pose adjustment and ultrasound scanning path optimization, and further enhancing the accuracy of ultrasound scanning results.

[0050] In an optional embodiment, determining the occlusion region corresponding to the plurality of ultrasound image blocks based on the plurality of spatial connectivity states to obtain the target occlusion region includes: 3351. Based on the multiple spatial connectivity states, perform neighborhood connectivity region identification on the multiple ultrasound image blocks to obtain multiple connected regions; 3352. Determine the area parameters and spatial distribution parameters of the multiple connected regions corresponding to the multiple regions; 3353. Verify the validity of the multiple connected regions based on the multiple region area parameters and the multiple region spatial distribution parameters to obtain connectivity verification results; and determine the target connected region and misjudged region corresponding to the target ultrasound image data based on the connectivity verification results; 3354. Generate the target occlusion region based on the target connected region and the misjudged region according to the preset occlusion determination rules.

[0051] In this embodiment, neighborhood connectivity region identification refers to performing region clustering analysis on multiple ultrasound image blocks with occlusion features based on the spatial adjacency and grayscale continuity relationships between multiple ultrasound image blocks to determine the spatial connectivity range corresponding to different occlusion regions. A connected region refers to a continuous region composed of multiple spatially adjacent ultrasound image blocks with similar occlusion features, primarily used to characterize candidate occlusion regions in the target ultrasound image. The region area parameter is used to characterize the spatial coverage of the connected region, and may include the pixel area, width, height, and proportion parameters of the connected region. The region spatial distribution parameter characterizes the spatial distribution of the connected region in the target ultrasound image, and may include the region center coordinates, region boundary positions, region shape continuity, and region directional distribution characteristics. The connectivity verification result is the determination of whether a candidate connected region belongs to a true occlusion region, and may include valid occlusion region identifiers and misjudged region identifiers. The target connected region is a true occlusion region after validity verification, while misjudged regions are abnormal regions caused by local noise, abrupt changes in tissue texture, or ultrasound artifacts. The preset occlusion determination rule is a rule model used to generate the final target occlusion region. It is mainly used to filter and merge different connected regions.

[0052] In a specific embodiment, a corresponding image block neighborhood connection graph is first established based on multiple spatial connectivity states and the positional adjacency relationships between each ultrasound image block. Specifically, multiple ultrasound image blocks are treated as nodes in a graph structure, and edge connections are established between spatially adjacent image blocks with similar grayscale features. Based on a breadth-first search algorithm or a region growing algorithm, neighborhood clustering is performed on multiple ultrasound image blocks with occlusion anomalies to identify multiple connected regions. For example, when multiple ultrasound image blocks are spatially distributed in a continuous band and have a consistent grayscale change trend, they are identified as the same connected region; when there are obvious spatial breaks or large grayscale differences between multiple image blocks, they are determined to belong to different connected regions. After obtaining multiple connected regions, regional feature analysis is performed on each connected region, and the pixel coverage area corresponding to each connected region is statistically analyzed to generate the corresponding region area parameters; at the same time, the center coordinates, boundary contours, and regional direction distribution states of each connected region are extracted to generate the corresponding region spatial distribution parameters. Next, based on multiple region area parameters and multiple region spatial distribution parameters, the validity of multiple connected regions is verified. This involves comparing the area parameter of the current connected region with a preset occlusion region area threshold. If the area of ​​the current connected region is lower than the minimum area threshold, the region is determined to be a random noise region or a local ultrasound artifact region. If the area of ​​the current connected region is too large and the grayscale changes within the region are discontinuous, the region is determined to be a false detection region caused by blurred tissue boundaries. Then, structural consistency analysis is performed based on the region spatial distribution parameters. After completing the validity verification, corresponding connectivity verification results are generated, and multiple connected regions are divided into target connected regions and false detection regions based on the verification results. Target connected regions correspond to tissue regions with actual occlusion effects, while false detection regions correspond to abnormal regions caused by tissue texture fluctuations, local artifacts, or random noise. Finally, the final target occlusion region is generated based on preset occlusion determination rules. Specifically, the preset occlusion determination rules may include region area constraint rules, region continuity constraint rules, region direction consistency rules, and region echo consistency rules. For example, when multiple target connected regions are spatially adjacent and have consistent echo characteristics, the host device fuses these regions to form a complete target occlusion region. When some target connected regions overlap with misjudged regions, the corresponding parts of the misjudged regions are removed through region boundary clipping. During continuous ultrasound scanning, the host device can also dynamically update the target occlusion region based on continuous frame target ultrasound image data. When the target occlusion region undergoes spatial changes due to user breathing, tissue movement, or intestinal peristalsis, the host device can correct the boundary range and spatial position parameters of the target occlusion region in real time, thereby improving the stability of occlusion region identification.

[0053] It can be seen that by combining neighborhood connectivity analysis, region area analysis and region spatial distribution verification, multiple abnormal ultrasound image blocks are effectively clustered and misjudged and removed, thereby improving the accuracy and stability of target occlusion region identification, avoiding interference from local noise and ultrasound artifacts on occlusion identification results, improving the robotic arm's avoidance of occlusion regions, and thus improving the efficiency of ultrasound scanning.

[0054] In an optional embodiment, generating the target occlusion region based on the target connected region and the misjudged region specifically includes the following steps: A1. Extract the area parameters and spatial distribution parameters corresponding to the target connected region from the multiple area parameters and the multiple spatial distribution parameters of the region to obtain the target region area parameters and the target region spatial distribution parameters; and extract the area parameters and spatial distribution parameters corresponding to the misjudged region from the multiple area parameters and the multiple spatial distribution parameters of the region to obtain the misjudged region area parameters and the misjudged region spatial distribution parameters. A2. Based on the preset occlusion area determination rules, determine the occlusion area parameters and spatial distribution parameters of the target area to obtain the occlusion area determination result. A3. Based on the occlusion area determination result, determine the candidate occlusion areas in the target connected region to obtain the first occlusion connected region; A4. Based on the misjudged region area parameter and the misjudged region spatial distribution parameter, the first occluded connected region is culled to obtain the target occluded connected region. A5. Merge the region boundaries of the target occlusion connected region to obtain the target occlusion region.

[0055] In this embodiment, the target region area parameter refers to the spatial area description parameter corresponding to the target connected region, which is mainly used to characterize the coverage and scale of the candidate occlusion region in the target ultrasound image. The target region spatial distribution parameter is a parameter used to characterize the spatial structural features of the target connected region in the ultrasound image, which may include the region center position, region boundary contour, region orientation distribution, region aspect ratio, and region continuity parameters. The misjudged region area parameter and the misjudged region spatial distribution parameter are used to describe the spatial size and spatial distribution state of the misjudged region, so as to perform the misjudged region removal process in the future. The occlusion region determination rule is a rule model used to identify the real occlusion region, which is mainly based on the spatial distribution law of bone occlusion and intestinal gas occlusion in ultrasound images. For example, bone occlusion regions are usually distributed in a long strip and have obvious directional consistency; intestinal gas occlusion regions are usually distributed in a clustered discrete distribution and accompanied by irregular changes in local boundaries. The first occlusion connected region refers to the set of candidate occlusion regions formed after the initial occlusion region determination, and the target occlusion connected region is the set of real occlusion regions formed after removing the misjudged regions. Region boundary merging refers to the process of merging the boundaries of multiple spatially adjacent target occlusion connected regions that belong to the same occlusion structure to form a complete target occlusion region.

[0056] In a specific embodiment, firstly, the region area, region boundary contour, region direction distribution, and region continuity parameters are extracted from the target connected region to generate target region area parameters and target region spatial distribution parameters. Simultaneously, corresponding area parameters and spatial distribution parameters are extracted from misjudged regions to generate misjudged region area parameters and misjudged region spatial distribution parameters. Next, based on the region area parameters, an area validity analysis is performed on the current target connected region. If the area of ​​the current target connected region is lower than a preset minimum occlusion area threshold, the region is determined to likely belong to a random noise region; if the region area exceeds a preset maximum occlusion area threshold and the internal structural continuity of the region is poor, the region is determined to likely belong to a region with blurred tissue boundaries. Further, the spatial structural characteristics of the target connected region are determined by combining the target region spatial distribution parameters. For example, when the target connected region is distributed in a long, continuous strip shape and its direction is consistent with the direction of human bony structures, the region is determined to conform to the characteristics of a skeletal occlusion region; when the target connected region is distributed in a discrete cluster shape and its boundary is irregularly diffused, the region is determined to conform to the characteristics of an intestinal gas occlusion region. Then, based on the occlusion region determination results, candidate regions that meet the occlusion structure characteristics are selected from multiple target connected regions and identified as the first occlusion connected region. At this point, the first occlusion connected region may still include some misjudged regions caused by ultrasound artifacts, sudden changes in local tissue texture, or random noise. Therefore, further misjudged region removal processing is required. Further, the spatial overlap relationship between the misjudged regions and the first occlusion connected region is analyzed. When there is a boundary intersection between the misjudged region and the first occlusion connected region, the overlapping region is corrected based on a region boundary clipping algorithm. When the misjudged region is located inside the first occlusion connected region and its area is relatively small, the corresponding region is removed from the first occlusion connected region to avoid local artifact regions affecting the final occlusion region identification result. Finally, to obtain a complete and continuous occlusion region structure, region boundary merging processing is performed on multiple target occlusion connected regions. Specifically, when the spatial distance between multiple target occlusion connected regions is less than a preset boundary fusion threshold and the region orientation distribution is consistent, the host device performs boundary fusion operations on multiple regions to form a continuous and complete target occlusion region. In addition, boundary smoothing can be performed on the target occlusion area to eliminate boundary spikes and local discontinuities generated during the region fusion process, thereby improving the stability and accuracy of the target occlusion area boundary.

[0057] By employing the above methods, including target connectivity region filtering, misjudged region removal, and region boundary fusion, accurate extraction of real occluded regions can be achieved. This effectively reduces the impact of ultrasound artifacts, local tissue texture fluctuations, and random noise on the occluded region recognition results, thereby improving the stability and reliability of target occluded region recognition.

[0058] Step S340: Determine the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data.

[0059] In this embodiment, the pose adjustment parameters refer to a set of control parameters used to control the movement state of the robotic arm and the spatial posture changes of the ultrasound probe. These parameters guide the robotic arm to adjust the spatial position and orientation of the ultrasound probe during ultrasound scanning to avoid obstructed areas in the current scanning path and improve the effective imaging quality of the target tissue. The pose adjustment parameters may include displacement adjustment parameters, rotation angle adjustment parameters, posture tilt parameters, probe normal correction parameters, scanning trajectory offset parameters, and probe contact pressure adjustment parameters at the robotic arm end effector. The displacement adjustment parameters characterize the translational change of the robotic arm end effector in three-dimensional space; the rotation angle adjustment parameters characterize the rotational change of the ultrasound probe around a preset coordinate axis; the posture tilt parameters characterize the change in the incident angle of the ultrasound probe relative to the target tissue region; the probe normal correction parameters adjust the normal angle relationship between the ultrasound probe and the target tissue surface; the scanning trajectory offset parameters correct the deviation between the current scanning path of the robotic arm and the target scanning path; and the probe contact pressure adjustment parameters adjust the contact pressure between the ultrasound probe and the user's body surface to improve ultrasound coupling stability.

[0060] Specifically, firstly, a spatial occlusion model for the current ultrasound scanning scenario is constructed based on the spatial location parameters corresponding to the occlusion area and the spatial coordinate relationship of the target tissue. Secondly, parameters such as tissue boundary integrity, tissue texture continuity, local signal-to-noise ratio distribution, and the proportion of the visible area of ​​the target tissue are extracted from the target tissue to generate a tissue quality assessment result corresponding to the current scanning state. Further, based on the tissue quality assessment result and the influence range of the occlusion area, the occlusion influence direction corresponding to the current ultrasound probe is determined. For example, when the skeletal occlusion area is located above the target tissue, it is determined that there is longitudinal acoustic shadowing occlusion in the current ultrasound propagation path; when the intestinal gas area is located on the side of the target tissue, it is determined that there is lateral scattering interference in the current ultrasound propagation path. After determining the occlusion influence direction, spatial constraint analysis is performed on the current pose of the robotic arm by combining the spatial structure information of the body surface in the depth image data. Specifically, the curvature of the user's body surface, the spatial boundary of the tissue area, and the reachable range of the robotic arm's movement are analyzed to determine the safe movement area that the robotic arm is allowed to adjust. Then, a corresponding pose adjustment strategy is generated based on the occlusion type. When the obstruction type is skeletal obstruction, the tilt angle between the ultrasound probe and the body surface is increased, and the ultrasound probe is laterally offset along the bone edge to allow the ultrasound waves to bypass the high-reflectivity areas of the bone and re-enter the target tissue area. Simultaneously, the normal direction of the ultrasound probe is corrected to ensure effective coverage of the target tissue. When the obstruction type is intestinal gas obstruction, the host device first controls the robotic arm to appropriately increase the probe contact pressure to reduce the local gas accumulation area; simultaneously, the scanning direction and spatial path of the ultrasound probe are adjusted so that the ultrasound waves enter the target tissue area from the direction of weaker gas interference; furthermore, the robotic arm scanning trajectory is locally replanned to bypass areas of dense gas. Finally, based on the robotic arm kinematic model, the generated pose adjustment strategy is solved using inverse kinematics to generate the corresponding robotic arm joint adjustment parameters and end effector motion parameters. The generated pose adjustment parameters are then subjected to motion smoothing to avoid abrupt movements of the robotic arm during pose adjustment.

[0061] It can be seen that by combining the occlusion type, occlusion identification information, and first ultrasound data, the occlusion influence status during the current ultrasound scanning process is analyzed, and corresponding robotic arm pose adjustment parameters are generated. This enables the robotic arm to adopt different spatial avoidance strategies for bone occlusion and intestinal gas occlusion, thereby achieving adaptive optimization of the ultrasound probe scanning path and probe posture, further improving the effective imaging coverage of the target tissue area, the accuracy of ultrasound scanning, and the efficiency of ultrasound scanning.

[0062] In an optional embodiment, determining the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data specifically includes the following steps: 341. Determine the first comprehensive quality score corresponding to the first ultrasound data based on the occlusion identification information; and determine the pose and motion constraint strategy corresponding to the occlusion type. 342. Determine the pose motion range parameters of the robotic arm according to the pose motion constraint strategy; 343. Determine the first state parameters of reinforcement learning corresponding to the first scanning state based on the first comprehensive quality score; the first scanning state is the state during the current ultrasound scan. 344. Input the first state parameter and the pose motion range parameter into the reinforcement learning algorithm to obtain multiple pose adjustment action strategies; 345. Determine multiple candidate scanning poses corresponding to the multiple pose adjustment action strategies; and control the robotic arm to move to the multiple candidate scanning poses so as to collect multiple scanning results through the ultrasonic probe; 346. Determine the second scanning state parameters corresponding to the multiple candidate scanning poses based on the multiple scanning results, and determine the second comprehensive quality score based on the second scanning state parameters; 347. Determine the comprehensive quality score change parameter between the second comprehensive quality score and the first comprehensive quality score; and determine the target pose adjustment action strategy based on the comprehensive quality score change parameter; 348. Determine the pose adjustment parameters of the robotic arm according to the target pose adjustment action strategy.

[0063] In this embodiment, the first comprehensive quality score is a comprehensive evaluation parameter used to characterize the imaging quality of the target tissue under the current ultrasound scanning state. It is mainly used to quantify the degree of influence of the occluded area on the ultrasound imaging effect during the current scanning process. The first comprehensive quality score can be calculated jointly based on parameters such as the integrity of the target tissue boundary, the continuity of tissue texture, the local signal-to-noise ratio, the proportion of the occluded area, the echo stability, and the effective visible area of ​​the tissue. The pose and motion constraint strategy refers to the pre-established pose adjustment constraint rules for the robotic arm for different occlusion types. It is mainly used to limit the direction of movement, the amplitude of movement, and the range of probe posture changes of the robotic arm during the pose optimization process. For example, in the scenario of skeletal occlusion, the pose and motion constraint strategy prioritizes constraining the robotic arm to perform lateral displacement and probe tilt adjustment along the bone edge direction; in the scenario of intestinal gas occlusion, it prioritizes constraining the robotic arm to perform local compression actions and multi-angle around-scanning actions. The pose motion range parameter is the movable space parameter of the robotic arm generated based on the pose and motion constraint strategy, which can include the spatial range of movement allowed at the end of the robotic arm, the angle range of rotation allowed by the probe, and the range of contact pressure allowed to be applied by the probe. The first state parameter is a description parameter of the current scan state used as input to the reinforcement learning algorithm. It characterizes the imaging state, occlusion state, and robotic arm motion state of the target tissue during the current ultrasound scan. The second comprehensive quality score is a quality assessment result obtained by re-scanning after the robotic arm performs candidate actions. It is mainly used to evaluate the optimization effect of the current pose adjustment action on the ultrasound imaging quality.

[0064] In a specific embodiment, firstly, the tissue boundary integrity parameter, tissue texture continuity parameter, effective visible area parameter of the target tissue, and local echo stability parameter are extracted from the target tissue region. Combined with the occlusion area ratio and occlusion intensity parameters, a first comprehensive quality score is calculated. The calculation formula can be: ;in, The score represents the overall quality at the current moment, with a value range of [0,1]. A higher score indicates that the cut surface is unobstructed and the anatomical structure is more complete. These are the weighting coefficients for the target anatomical structure matching score, the occlusion avoidance score based on the fused occlusion type, and the effective signal-to-noise ratio score, respectively, satisfying... Fixed values ​​are preset based on clinical diagnostic needs; The score is given for the matching degree of the target anatomical structure, with a value range of [0,1]. The occlusion avoidance score is a fusion of occlusion types, with a value range of [0,1]. The effective signal-to-noise ratio score has a value range of [0,1].

[0065] Furthermore, by combining the kinematic model of the robotic arm, the spatial structure model of the user's body surface, and the current spatial position of the target tissue, the spatial boundary range of the allowable movement of the robotic arm's end effector is determined, and the corresponding probe rotation angle constraint range and contact pressure constraint range are generated. Then, the first comprehensive quality score, the spatial distribution parameters of the occlusion area, the current pose parameters of the robotic arm, the integrity parameters of the target tissue boundary, and the probe contact state parameters are fused and encoded to form the state vector corresponding to the reinforcement learning algorithm. The first state parameters and the pose movement range parameters are then input into the reinforcement learning algorithm to generate multiple pose adjustment action strategies. Specifically, based on the current state vector, the reinforcement learning algorithm predicts the possible changes in imaging quality caused by different pose adjustment actions and outputs multiple candidate action strategies. For example, these include probe lateral offset actions, probe tilt angle adjustment actions, local compression actions, and arc scanning actions. Then, based on the multiple pose adjustment action strategies, multiple candidate scanning poses are generated, and the robotic arm is controlled to move sequentially to the multiple candidate scanning poses to acquire multiple scanning results through the ultrasound probe. During the pose adjustment process of the robotic arm, the host device monitors the probe contact pressure and the spatial position of the target tissue in real time to avoid excessive movement or tissue displacement during the robotic arm's motion. Simultaneously, the host device performs quality analysis on multiple scan results, extracting parameters such as tissue boundary integrity, tissue texture continuity, echo stability, and effective visible tissue area, and generating multiple second scan state parameters. Then, a second comprehensive quality score is calculated based on these second scan state parameters. Next, the host device calculates the comprehensive quality score change parameter based on the difference between the second and first comprehensive quality scores. Specifically, when the comprehensive quality score change parameter is greater than a preset quality improvement threshold, it indicates that the current pose adjustment action effectively improves the ultrasound imaging quality; conversely, it indicates that the current action strategy optimization effect is poor. Finally, based on the comprehensive quality score change parameter, the target pose adjustment action strategy with the best quality improvement effect is selected from multiple pose adjustment action strategies, and the final pose adjustment parameters are generated based on this target pose adjustment action strategy, including robotic arm end-effector displacement parameters, probe tilt angle parameters, trajectory offset parameters, and probe contact pressure parameters.

[0066] It can be seen that by combining comprehensive quality scoring analysis, pose and motion constraint control, and reinforcement learning pose optimization, adaptive optimization of the robotic arm scanning pose is achieved. This enables the system to dynamically learn the optimal pose adjustment strategy according to different occlusion types, thereby improving the effective imaging coverage of the target tissue area and the quality of ultrasonic scanning, and improving the efficiency of ultrasonic scanning.

[0067] In an optional embodiment, determining the first comprehensive quality score corresponding to the first ultrasound data based on the occlusion identification information specifically includes the following steps: 3411. Determine the distribution parameters of the occlusion area corresponding to the target scanning organization based on the occlusion identification information; 3412. Determine the proportion parameter of the obstructed area during ultrasonic scanning based on the obstructed area distribution parameter; and determine the obstruction impact parameter corresponding to the obstruction type based on the obstruction identification information; 3413. Determine the echo continuity parameter corresponding to the target scanned tissue based on the target ultrasound echo data; and determine the tissue boundary integrity parameter corresponding to the target scanned tissue based on the target ultrasound image data; 3414. Determine the scanning quality status parameters corresponding to the target scanned tissue based on the occlusion area ratio parameter, the occlusion influence parameter, the echo continuity parameter, and the tissue boundary integrity parameter; 3415. Determine the first comprehensive quality score corresponding to the first ultrasound data based on the scan quality status parameters.

[0068] In this embodiment, the occlusion region distribution parameters are a set of parameters used to characterize the spatial distribution and structural characteristics of occlusion regions in the target scanned tissue, reflecting the coverage and spatial concentration of skeletal occlusion or intestinal gas occlusion in the target tissue region. The occlusion region distribution parameters include the spatial coordinate distribution information of the occlusion region, the continuity distribution characteristics of the occlusion region, the directional characteristics of the occlusion region, and the aggregation degree parameters of the occlusion region. The occlusion region proportion parameter is used to quantify the proportion of the occlusion region in the overall area of ​​the target scanned tissue, directly reflecting the degree of influence of occlusion on the effective imaging area. The occlusion influence parameter is a weighted parameter used to characterize the intensity of the influence of different occlusion types on ultrasound propagation and imaging quality. For example, skeletal occlusion usually has high reflection and acoustic shadowing effects, and its occlusion influence parameter is relatively high, while intestinal gas occlusion manifests as scattering and noise interference, and its influence parameter reflects the degree of local structural disorder. The echo continuity parameter is a parameter used to characterize the spatial continuity and temporal stability of the target scanned tissue in ultrasound echo data, reflecting the integrity of the acoustic propagation path within the tissue. The tissue boundary integrity parameter is used to characterize the edge clarity and structural recognizability of the target tissue in ultrasound images. It can be quantified by features such as edge gradient change, boundary closure degree, and contour continuity.

[0069] In a specific embodiment, firstly, by mapping the spatial coordinate information of the occluded area, a distribution model of the occluded area in a three-dimensional body surface coordinate system is constructed, thereby obtaining the occluded area distribution parameters. Then, based on the occluded area distribution parameters, the occluded area proportion parameter is calculated. Specifically, by statistically analyzing the pixel coverage ratio of the occluded area in the corresponding image space of the target scanned tissue, and combining the spatial scale information in the depth image data, the two-dimensional pixel proportion is mapped to the actual spatial proportion, thus obtaining a more physically meaningful occluded area proportion parameter. Simultaneously, based on the classification results of different occlusion types in the occlusion identification information, corresponding occlusion influence weight parameters are introduced to reflect the differentiated impact of different occlusion types on ultrasound imaging quality. Next, by analyzing the amplitude changes and phase consistency of the echo signal during spatial propagation, the continuity of the echo path within the target tissue is quantified. When the echo signal exhibits a continuous and stable attenuation trend, it indicates good tissue structure continuity; when the echo signal shows obvious breaks or random fluctuations, it indicates the presence of occlusion or tissue structure abnormalities. Simultaneously, tissue boundary integrity parameters are extracted from the target ultrasound image data. The tissue contour is extracted using an edge detection operator, and the degree of boundary closure and contour continuity are calculated to assess the identifiability of the tissue structure in the image. Then, the occlusion area proportion parameter, occlusion influence parameter, echo continuity parameter, and tissue boundary integrity parameter are fused in multiple dimensions to construct the scanning quality status parameters corresponding to the target tissue. Specifically, these status parameters represent the comprehensive state of tissue imaging quality during the current ultrasound scan, reflecting not only the degree of occlusion but also the visibility of the tissue structure and the stability of echo propagation, thus achieving a multi-dimensional unified representation of ultrasound imaging quality. Finally, the scanning quality status parameters are normalized and mapped to a preset quality scoring function model to obtain the first comprehensive quality score corresponding to the first ultrasound data.

[0070] It can be seen that by fusing and modeling multi-dimensional parameters such as the distribution characteristics of the occlusion area, the occlusion ratio characteristics, the weight of the occlusion influence, and the continuity of ultrasound echoes and the integrity of tissue boundaries, a multi-scale quantitative evaluation of ultrasound scan quality is achieved. This enables the first comprehensive quality score to accurately reflect the imaging quality level under the current scan state and provides a reliable state evaluation basis for subsequent pose optimization based on reinforcement learning, thereby improving the adaptive capability of the ultrasound scan control strategy.

[0071] In an optional embodiment, determining the second scanning state parameters corresponding to the plurality of candidate scanning poses based on the plurality of scanning results specifically includes the following steps: 3461. Perform tissue identification on each of the multiple scanning results to obtain multiple tissue regions; 3462. Based on the multiple scanning results, determine the occlusion distribution state corresponding to each scanned tissue region in the multiple scanned tissue regions, thereby obtaining multiple occlusion distribution states; and based on the multiple scanning results, determine multiple echo continuity states corresponding to each scanned tissue region in the multiple scanned tissue regions; and based on the multiple scanning results, determine multiple tissue structure integrity states corresponding to each scanned tissue region in the multiple scanned tissue regions. 3463. Determine the scanning quality change state corresponding to the multiple scanning results based on the multiple occlusion distribution states, the multiple echo continuity states, and the multiple tissue structure integrity states; 3464. Generate the second scanning state parameters based on the scanning quality change status.

[0072] In this embodiment, tissue identification involves matching and analyzing the prior anatomical information of the target tissue with ultrasound echo characteristics to ensure that each scan result corresponds to a specific tissue structural unit, such as liver tissue, kidney tissue, or vascular tissue. The scanned tissue region refers to the local representation of the target tissue in the image space or echo space obtained by scanning with an ultrasound probe under a candidate scan pose, reflecting the actual visible range of tissue imaging under that pose. Occlusion distribution state is a state parameter characterizing the spatial distribution characteristics of occlusion structures within or around the tissue region. It describes the degree and distribution of the impact of bone or gas occlusion on image quality under that scan pose. Echo continuity state characterizes the continuity of the scanned tissue region in the ultrasound echo signal, mainly reflecting whether there are abnormal phenomena such as breaks, attenuation, or enhanced scattering during the propagation of sound waves within the tissue. Tissue structure integrity state characterizes the clarity of the target tissue's boundaries and the degree of structural recognizability in the scan result, mainly reflecting whether the tissue outline is complete, whether the boundaries are continuous, and whether the internal structure is stably identifiable. The scan quality change state characterizes the differences in imaging quality between different candidate scan poses. It is quantified by comprehensively comparing the changing trends of occlusion distribution, echo continuity, and tissue structure integrity under different pose conditions, thus reflecting the improvement or degradation effect of pose adjustment on ultrasound imaging quality. The second scan state parameter is a reinforcement learning feedback state input constructed based on the above scan quality change state, used to characterize the comprehensive imaging effect after the current candidate pose is executed.

[0073] In a specific embodiment, firstly, based on a preset target tissue model, feature matching analysis is performed on the ultrasound images in each scan result to extract the corresponding scanned tissue region. For example, when the scan result corresponding to the candidate pose includes liver structural features, it is identified as the liver scanned tissue region. Next, the occlusion structures within the tissue region are detected. By analyzing local bright reflection areas, acoustic shadow areas, and echo missing areas, the occlusion distribution state corresponding to the tissue region is determined. Further, the continuity of the echo signal in the scan result is analyzed. By calculating the temporal stability and spatial consistency of the echo amplitude, the echo continuity state is determined. When the echo signal shows stable attenuation and uniform spatial distribution, it indicates that the tissue imaging continuity under the scan pose is good; when the echo signal shows breaks or strong fluctuations, it indicates that there is obvious occlusion or inhomogeneity of the medium. At the same time, the integrity of the tissue structure is assessed. The degree of closure and continuity of the tissue boundary are analyzed by edge detection and structural contour extraction methods to determine the integrity state of the tissue structure. After completing the state analysis, the occlusion distribution state, echo continuity state, and tissue structure integrity state are fused in multiple dimensions to generate a scan quality change state. This state is used to uniformly characterize the trend of image quality changes among different candidate scan poses. Finally, a second scan state parameter is generated based on the scan quality change state. Specifically, the multidimensional state features are vectorized and encoded, and mapped to the reinforcement learning state space, so that they can serve as input parameters for the subsequent policy optimization model, thereby providing a data foundation for the optimal selection of pose adjustment strategies.

[0074] It can be seen that by performing multi-dimensional fusion modeling of tissue recognition, occlusion distribution analysis, echo continuity analysis, and structural integrity analysis corresponding to candidate scanning poses, a refined quantitative expression of the imaging quality changes of different scanning poses is achieved. This constructs more accurate second scanning state parameters, providing high-quality feedback for subsequent reinforcement learning strategy optimization, improving the accuracy and adaptability of the robot arm's pose optimization during ultrasound scanning, and thus enhancing the efficiency of ultrasound scanning.

[0075] Step S350: Control the robotic arm to adjust its posture according to the posture adjustment parameters so that the robotic arm moves to the target scanning position; and scan the target scanning position with an ultrasound probe to collect target ultrasound data of the user tissue area.

[0076] In this embodiment, the target scanning location refers to the target spatial location suitable for ultrasound scanning, determined based on the occlusion avoidance analysis results and the spatial distribution of the target tissue. The ultrasound propagation path corresponding to this location can effectively avoid occlusion from bone sites or intestinal gas, thereby improving the effective imaging quality of the target tissue. Target ultrasound data refers to the ultrasound data acquired by scanning the user tissue area corresponding to the target scanning location after the ultrasound probe has completed pose adjustment. This data may include B-mode ultrasound image data corresponding to the target tissue, tissue echo intensity data, tissue boundary continuity data, tissue texture distribution data, and corresponding dynamic echo time-series data. The target ultrasound data is mainly used for subsequent physician diagnosis, lesion analysis, and tissue structure assessment.

[0077] In a specific embodiment, firstly, the pose adjustment parameters are decomposed into joint trajectories based on the robotic arm's kinematic model to generate corresponding robotic arm joint motion sequences. These sequences are then sent to the robotic arm controller to control each joint of the robotic arm to perform pose adjustments according to preset motion trajectories. During the robotic arm's movement, the host device acquires real-time angle feedback data from each joint of the robotic arm and spatial pose data from the end effector. Based on the feedback results, it performs closed-loop correction on the current motion state of the robotic arm to ensure that the ultrasonic probe at the end of the robotic arm can accurately move to the target scanning position. Further, as the ultrasonic probe approaches the user's body surface, the flexible end effector gradually reduces the movement speed of the robotic arm's end effector and adjusts the contact pressure between the ultrasonic probe and the user's body surface, ensuring that the ultrasonic probe contacts the target tissue area surface in a flexible, conforming manner. Then, based on real-time contact pressure data acquired by a force feedback sensor, the contact state of the ultrasonic probe is dynamically adjusted to avoid user discomfort due to excessive contact pressure or unstable ultrasonic coupling due to insufficient contact pressure. After completing the pose adjustment, the ultrasonic probe performs an ultrasonic scan of the target scanning position. Specifically, the ultrasound probe emits ultrasound signals towards the target tissue region and receives ultrasound echo signals reflected from within the tissue. Then, the ultrasound echo signals undergo beamforming, envelope detection, dynamic gain compensation, and grayscale mapping to generate corresponding target ultrasound data. Furthermore, the host device can perform real-time analysis of tissue boundary integrity, tissue texture continuity, and local signal-to-noise ratio in the target ultrasound data to assess the imaging quality corresponding to the current target scanning position. When the tissue imaging quality in the target ultrasound data falls below a preset threshold, the host device can regenerate pose adjustment parameters and control the robotic arm to perform secondary pose adjustments to further optimize the scanning position and posture of the ultrasound probe. The host device can also dynamically track the user's tissue region using depth image data. When the target tissue position changes due to the user's breathing, slight movement, or tissue deformation, the host device can update the target scanning position in real time and synchronously adjust the robotic arm's motion to maintain the spatial alignment between the ultrasound probe and the target tissue region.

[0078] As can be seen, by controlling the pose adjustment parameters to optimize the pose of the ultrasound probe and performing ultrasound scanning on the user's tissue area at the target scanning position, the ultrasound probe can actively avoid areas obstructed by bone and intestinal gas, thereby obtaining real, complete and high-quality target ultrasound data. At the same time, through closed-loop pose control and dynamic contact adjustment mechanism, the spatial positioning stability and tissue imaging accuracy during ultrasound scanning are improved, further reducing the risk of artifacts and false pathological features introduced by traditional image restoration methods.

[0079] For easier understanding, please refer to Figure 6 , Figure 6This is a flowchart illustrating another ultrasonic scanning obstruction avoidance control method provided in this application embodiment. As can be seen, Figure 6 Following the logic of "initial positioning - distortion-free acquisition - classification recognition - quantitative evaluation - intelligent decision-making - closed-loop execution," a complete control system was constructed, from active obstacle avoidance of the physical acoustic path to the output of raw images. This system achieves accurate identification and active avoidance of two types of obstructions: skeletal and intestinal gas. Specifically, step S601, "scanning initialization and initial target area positioning," is executed first. Based on the patient's three-dimensional point cloud and target anatomical structure information, combined with a general anatomical model, the initial pose of the six-degree-of-freedom robotic arm of the ultrasound probe is determined, establishing the mapping relationship between the scanning coordinate system and the target area. Next, step S602, "raw ultrasound signal acquisition and distortion-free preprocessing," is executed. Raw radiofrequency echo data and raw B-mode ultrasound images are acquired simultaneously through the ultrasound probe. No pixel modification or image restoration is performed throughout the process; only data synchronization and format standardization are completed, preserving all original features of the signal and image. Then, step S603, "Merging occlusion regions classified by occlusion type and performing distortion-free identification," is executed. Combining the amplitude stability characteristics of radiofrequency echoes with the grayscale and boundary gradient characteristics of B-mode images, occlusion regions are located and skeletal occlusion and intestinal gas occlusion are distinguished, generating distortion-free occlusion identification information. Next, step S604, "Merging Target Anatomical Structure Integrity Quantitative Assessment of Occlusion Type," is executed. A comprehensive quality scoring system integrating anatomical structure matching degree, occlusion avoidance degree, and signal-to-noise ratio is constructed. Differentiated penalty weights are set for the two types of occlusion to quantitatively assess the effectiveness of the current scanning section. Finally, step S605, "Occlusion Avoidance Reinforcement Learning Decision Reasoning of Occlusion Type," is executed. Based on the scoring results, the motion space for probe pose adjustment is dynamically constrained for different occlusion types, outputting the optimal 6DoF ultrasound probe pose adjustment amount, achieving intelligent obstacle avoidance decision-making adapted to individual anatomical differences. Finally, step S606, "Robotic arm inverse kinematics solution and pose update, synchronous acquisition of ultrasound images and output of scan results," is executed. A smooth motion trajectory is generated through inverse kinematics solution, driving the robotic arm to update the probe pose. Data is acquired again and the process iterates until the score meets the standard, outputting a distortion-free, original, standardized ultrasound image, thus completing the scan. This approach ensures the originality of ultrasound data and diagnostic compliance while improving the accuracy of occlusion avoidance and scanning efficiency, effectively solving the problems of image distortion and high diagnostic risk in existing technologies.

[0080] As can be seen, by implementing the above-described embodiment of the ultrasonic scanning occlusion avoidance control method, the host device acquires depth image data of the user's tissue region through the camera module and first ultrasonic data through the ultrasonic probe. Based on the depth image data and the first ultrasonic data, the target tissue structure is identified to determine the target tissue to be scanned. Next, based on the target tissue and the first ultrasonic data, the local abnormal echo regions generated during the current scanning process are analyzed to identify the occlusion type and generate corresponding occlusion identification information. The occlusion types include skeletal occlusion and intestinal gas occlusion. Then, based on the occlusion type, occlusion identification information, and the first ultrasonic data, robotic arm pose adjustment parameters are constructed and generated. The robotic arm is then controlled to drive the ultrasonic probe to perform pose adjustment, enabling the ultrasonic probe to actively avoid skeletal and intestinal gas occlusion regions in space, thereby moving to the target scanning position and maintaining stable scanning to obtain the target ultrasonic data corresponding to the target user tissue region. This allows the robotic arm's pose to be adjusted according to the occlusion type during scanning, thereby actively avoiding occluded objects and optimizing the pose. Without relying on image restoration or interpolation reconstruction, it directly acquires raw ultrasound data based on real acoustic echo information, avoiding the introduction of false pathological features, and significantly improving the imaging accuracy and diagnostic reliability of ultrasound scans.

[0081] The above primarily describes the solutions of the embodiments of this application from the perspective of the method execution process. It is understood that, in order to achieve the above functions, the host device includes the corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments provided herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware 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.

[0082] This application embodiment can divide the host device into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0083] When dividing each function into modules according to its corresponding function. Figure 7This is a functional module block diagram of an ultrasonic scanning obstruction avoidance control device provided in an embodiment of this application. The ultrasonic scanning obstruction avoidance control device 700 is applied to the host device in an ultrasonic scanning system. The ultrasonic scanning obstruction avoidance control device 700 includes: The acquisition unit 701 is used to acquire depth image data and first ultrasound data of a user tissue region specified by a doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes changes in the surface curvature of the user tissue region. Processing unit 702 is configured to determine the user's target tissue for scanning based on the depth image data and the first ultrasound data; and to determine the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data; the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning. The calculation unit 703 is used to determine the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data. The control unit 704 is used to control the robotic arm to perform posture adjustment according to the posture adjustment parameters so that the robotic arm moves to the target scanning position; and to scan the target scanning position with an ultrasonic probe to collect target ultrasonic data of the user tissue area.

[0084] In an optional embodiment, the first ultrasound data includes first ultrasound image data and first ultrasound echo data. Specifically, the processing unit 702, in determining the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data, is used for: The ultrasound image data corresponding to the target scanned tissue is extracted from the first ultrasound image data to obtain target ultrasound image data; and the ultrasound echo data corresponding to the target scanned tissue is extracted from the first ultrasound echo data to obtain target ultrasound echo data. The target ultrasound image data is divided into image regions to obtain multiple ultrasound image blocks; Determine the grayscale feature parameters corresponding to each of the plurality of ultrasound image blocks to obtain a plurality of grayscale feature parameters; Based on the multiple grayscale feature parameters, the multiple ultrasound image blocks are occlusion identified to obtain the occlusion identification result; Based on the occlusion recognition results, multiple spatial connectivity states corresponding to the multiple ultrasound image blocks are determined; and based on the multiple spatial connectivity states, occlusion regions corresponding to the multiple ultrasound image blocks are determined to obtain the target occlusion region. Determine the target echo characteristics corresponding to the target obstruction area in the target ultrasonic echo data; The occlusion type of the target occlusion region is determined based on the target echo characteristics; and the occlusion identification information of the target occlusion region is determined based on the target echo characteristics and the plurality of spatial connectivity states.

[0085] In an optional embodiment, the processing unit 702, in determining the occlusion regions corresponding to the plurality of ultrasound image blocks based on the plurality of spatial connectivity states to obtain the target occlusion region, is specifically used for: Based on the multiple spatial connectivity states, the multiple ultrasound image blocks are used to identify neighborhood connectivity regions, resulting in multiple connected regions; Determine multiple area parameters and multiple spatial distribution parameters of the multiple connected regions; The validity of the multiple connected regions is verified based on the multiple region area parameters and the multiple region spatial distribution parameters to obtain connectivity verification results; and the target connected regions and misjudged regions corresponding to the target ultrasound image data are determined based on the connectivity verification results. The target occlusion region is generated based on the target connected region and the misjudged region using a preset occlusion determination rule.

[0086] In an optional embodiment, the processing unit 702, in generating the target occlusion region based on the target connected region and the misjudged region, is specifically configured to: Extract the area parameters and spatial distribution parameters corresponding to the target connected region from the plurality of area parameters and spatial distribution parameters to obtain the target region area parameters and spatial distribution parameters; and extract the area parameters and spatial distribution parameters corresponding to the misjudged region from the plurality of area parameters and spatial distribution parameters to obtain the misjudged region area parameters and spatial distribution parameters. The occlusion area parameters and spatial distribution parameters of the target area are determined according to the preset occlusion area determination rules to obtain the occlusion area determination result. Based on the occlusion region determination result, candidate occlusion regions in the target connected region are determined to obtain the first occlusion connected region; Based on the misjudged region area parameter and the misjudged region spatial distribution parameter, the first occluded connected region is culled to obtain the target occluded connected region; The target occlusion region is obtained by merging the region boundaries of the target occlusion region.

[0087] In an optional embodiment, the calculation unit 703 is specifically used for determining the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data, in the following ways: The first comprehensive quality score corresponding to the first ultrasound data is determined based on the occlusion identification information; and the pose and motion constraint strategy corresponding to the occlusion type is determined. The pose motion range parameters of the robotic arm are determined according to the pose motion constraint strategy. The first state parameters of reinforcement learning corresponding to the first scanning state are determined based on the first comprehensive quality score; the first scanning state is the state during the current ultrasound scan. The first state parameter and the pose motion range parameter are input into the reinforcement learning algorithm to obtain multiple pose adjustment action strategies; Determine multiple candidate scanning poses corresponding to the multiple pose adjustment action strategies; and control the robotic arm to move to the multiple candidate scanning poses so as to collect multiple scanning results through the ultrasonic probe; Based on the multiple scanning results, a second scanning state parameter corresponding to the multiple candidate scanning poses is determined, and a second comprehensive quality score is determined based on the second scanning state parameter; Determine the change parameters of the comprehensive quality score between the second comprehensive quality score and the first comprehensive quality score; and determine the target pose adjustment action strategy based on the change parameters of the comprehensive quality score. The pose adjustment parameters of the robotic arm are determined based on the target pose adjustment action strategy.

[0088] In an optional embodiment, the calculation unit 703, in determining the first comprehensive quality score corresponding to the first ultrasound data based on the occlusion identification information, is specifically used for: The distribution parameters of the occlusion area corresponding to the target scanned tissue are determined based on the occlusion identification information. The occlusion area ratio parameter during ultrasonic scanning is determined based on the occlusion area distribution parameter; and the occlusion impact parameter corresponding to the occlusion type is determined based on the occlusion identification information. The echo continuity parameter corresponding to the target scanned tissue is determined based on the target ultrasound echo data; and the tissue boundary integrity parameter corresponding to the target scanned tissue is determined based on the target ultrasound image data. The scanning quality status parameters corresponding to the target scanned tissue are determined based on the occlusion area ratio parameter, the occlusion effect parameter, the echo continuity parameter, and the tissue boundary integrity parameter. The first comprehensive quality score corresponding to the first ultrasound data is determined based on the scan quality status parameters.

[0089] In an optional embodiment, the calculation unit 703, in determining the second scanning state parameters corresponding to the plurality of candidate scanning poses based on the plurality of scanning results, is specifically used for: Each of the multiple scanning results is subjected to tissue identification to obtain multiple tissue regions. Based on the multiple scanning results, determine the occlusion distribution state corresponding to each of the multiple scanned tissue regions, thereby obtaining multiple occlusion distribution states; and based on the multiple scanning results, determine multiple echo continuity states corresponding to each of the multiple scanned tissue regions; and based on the multiple scanning results, determine multiple tissue structure integrity states corresponding to each of the multiple scanned tissue regions. The scanning quality change states corresponding to the multiple scanning results are determined based on the multiple occlusion distribution states, the multiple echo continuity states, and the multiple tissue structure integrity states. The second scanning status parameter is generated based on the scanning quality change status.

[0090] As can be seen, this application provides an ultrasound scanning occlusion avoidance control device 700, applied to the host device of an ultrasound scanning system. This device 700 acquires depth image data and first ultrasound data of a user's tissue region specified by a doctor. Based on the depth image data and the first ultrasound data, it determines the user's target tissue to be scanned. Based on the target tissue and the first ultrasound data, it determines the occlusion type and corresponding occlusion identification information during ultrasound scanning. The occlusion types include bone occlusion and intestinal gas occlusion. Based on the occlusion type, occlusion identification information, and the first ultrasound data, it determines the posture adjustment parameters of a robotic arm. Based on the posture adjustment parameters, it controls the robotic arm to adjust its posture so that it moves to the target scanning position and scans the target scanning position using an ultrasound probe to collect target ultrasound data of the user's tissue region. This improves the accuracy and reliability of ultrasound scanning.

[0091] This application also provides a computer-readable storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes a host device.

[0092] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer includes a host device.

[0093] It should be noted that, for the sake of simplicity, the above embodiments are all described as a series of actions. Those skilled in the art should understand that this application is not limited to the described order of actions, as some steps in the embodiments of this application can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of this application.

[0094] In the above embodiments, the descriptions of each embodiment in this application have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0095] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0096] The steps of the methods or algorithms described in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, portable hard disk, read-only optical disk (CD-ROM), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor.

[0097] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated.

[0098] The modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both. For example, for devices and products applied to or integrated into a chip, all modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For devices and products applied to or integrated into a terminal device, all modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components within the terminal device, or at least some modules / units can be implemented using software programs that run on a processor integrated within the terminal device, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0099] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above descriptions are merely specific embodiments of the embodiments of this application and are not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solutions of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A method for controlling obstruction during ultrasonic scanning, characterized in that, A host device used in an ultrasonic scanning system, the method comprising: Acquire depth image data and first ultrasound data of a user tissue region specified by the doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes changes in the surface curvature of the user tissue region. The user's target tissue to be scanned is determined based on the depth image data and the first ultrasound data; the occlusion type and corresponding occlusion identification information during ultrasound scanning are determined based on the target tissue to be scanned and the first ultrasound data; the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning; the pose adjustment parameters of the robotic arm are determined based on the occlusion type, the occlusion identification information and the first ultrasound data; The robotic arm is controlled to adjust its posture according to the posture adjustment parameters so that it moves to the target scanning position; and the target scanning position is scanned by an ultrasonic probe to collect target ultrasonic data of the user's tissue area. The first ultrasound data includes first ultrasound image data and first ultrasound echo data; the step of determining the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data includes: The ultrasound image data corresponding to the target scanned tissue is extracted from the first ultrasound image data to obtain target ultrasound image data; and the ultrasound echo data corresponding to the target scanned tissue is extracted from the first ultrasound echo data to obtain target ultrasound echo data. The target ultrasound image data is divided into image regions to obtain multiple ultrasound image blocks; Determine the grayscale feature parameters corresponding to each of the plurality of ultrasound image blocks to obtain a plurality of grayscale feature parameters; Based on the multiple grayscale feature parameters, the multiple ultrasound image blocks are occlusion identified to obtain the occlusion identification result; Based on the occlusion recognition results, multiple spatial connectivity states corresponding to the multiple ultrasound image blocks are determined; and based on the multiple spatial connectivity states, occlusion regions corresponding to the multiple ultrasound image blocks are determined to obtain the target occlusion region. Determine the target echo characteristics corresponding to the target obstruction area in the target ultrasonic echo data; The occlusion type of the target occlusion region is determined based on the target echo characteristics; and the occlusion identification information of the target occlusion region is determined based on the target echo characteristics and the plurality of spatial connectivity states. The step of determining the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data includes: The first comprehensive quality score corresponding to the first ultrasound data is determined based on the occlusion identification information; and the pose and motion constraint strategy corresponding to the occlusion type is determined. The pose motion range parameters of the robotic arm are determined according to the pose motion constraint strategy. The first state parameters of reinforcement learning corresponding to the first scanning state are determined based on the first comprehensive quality score; the first scanning state is the state during the current ultrasound scan. The first state parameter and the pose motion range parameter are input into the reinforcement learning algorithm to obtain multiple pose adjustment action strategies; Determine multiple candidate scanning poses corresponding to the multiple pose adjustment action strategies; and control the robotic arm to move to the multiple candidate scanning poses so as to collect multiple scanning results through the ultrasonic probe; Based on the multiple scanning results, a second scanning state parameter corresponding to the multiple candidate scanning poses is determined, and a second comprehensive quality score is determined based on the second scanning state parameter; Determine the change parameters of the comprehensive quality score between the second comprehensive quality score and the first comprehensive quality score; and determine the target pose adjustment action strategy based on the change parameters of the comprehensive quality score. The pose adjustment parameters of the robotic arm are determined based on the target pose adjustment action strategy.

2. The method as described in claim 1, characterized in that, The step of determining the occlusion region corresponding to the multiple ultrasound image blocks based on the multiple spatial connectivity states to obtain the target occlusion region includes: Based on the multiple spatial connectivity states, the multiple ultrasound image blocks are used to identify neighborhood connectivity regions, resulting in multiple connected regions; Determine multiple area parameters and multiple spatial distribution parameters of the multiple connected regions; The validity of the multiple connected regions is verified based on the multiple region area parameters and the multiple region spatial distribution parameters to obtain connectivity verification results; and the target connected regions and misjudged regions corresponding to the target ultrasound image data are determined based on the connectivity verification results. The target occlusion region is generated based on the target connected region and the misjudged region using a preset occlusion determination rule.

3. The method as described in claim 2, characterized in that, The step of generating the target occlusion region based on the target connected region and the misjudged region includes: Extract the area parameters and spatial distribution parameters corresponding to the target connected region from the plurality of area parameters and spatial distribution parameters to obtain the target region area parameters and spatial distribution parameters; and extract the area parameters and spatial distribution parameters corresponding to the misjudged region from the plurality of area parameters and spatial distribution parameters to obtain the misjudged region area parameters and spatial distribution parameters. The occlusion area parameters and spatial distribution parameters of the target area are determined according to the preset occlusion area determination rules to obtain the occlusion area determination result. Based on the occlusion region determination result, candidate occlusion regions in the target connected region are determined to obtain the first occlusion connected region; Based on the misjudged region area parameter and the misjudged region spatial distribution parameter, the first occluded connected region is culled to obtain the target occluded connected region; The target occlusion region is obtained by merging the region boundaries of the target occlusion region.

4. The method as described in claim 1, characterized in that, The step of determining the first comprehensive quality score corresponding to the first ultrasound data based on the occlusion identification information includes: The distribution parameters of the occlusion area corresponding to the target scanned tissue are determined based on the occlusion identification information. The occlusion area ratio parameter during ultrasonic scanning is determined based on the occlusion area distribution parameter; and the occlusion impact parameter corresponding to the occlusion type is determined based on the occlusion identification information. The echo continuity parameter corresponding to the target scanned tissue is determined based on the target ultrasound echo data; and the tissue boundary integrity parameter corresponding to the target scanned tissue is determined based on the target ultrasound image data. The scanning quality status parameters corresponding to the target scanned tissue are determined based on the occlusion area ratio parameter, the occlusion effect parameter, the echo continuity parameter, and the tissue boundary integrity parameter. The first comprehensive quality score corresponding to the first ultrasound data is determined based on the scan quality status parameters.

5. The method as described in claim 1, characterized in that, The step of determining the second scanning state parameters corresponding to the multiple candidate scanning poses based on the multiple scanning results includes: Each of the multiple scanning results is subjected to tissue identification to obtain multiple tissue regions. Based on the multiple scanning results, determine the occlusion distribution state corresponding to each of the multiple scanned tissue regions, thereby obtaining multiple occlusion distribution states; and based on the multiple scanning results, determine multiple echo continuity states corresponding to each of the multiple scanned tissue regions; and based on the multiple scanning results, determine multiple tissue structure integrity states corresponding to each of the multiple scanned tissue regions. The scanning quality change states corresponding to the multiple scanning results are determined based on the multiple occlusion distribution states, the multiple echo continuity states, and the multiple tissue structure integrity states. The second scanning status parameter is generated based on the scanning quality change status.

6. An ultrasonic scanning obstruction avoidance control device, characterized in that, A main unit used in an ultrasonic scanning system, the device comprising: The acquisition unit is used to acquire depth image data and first ultrasound data of a user tissue region specified by a doctor; the first ultrasound data includes ultrasound grayscale distribution, tissue edge echo characteristics, internal tissue echo intensity distribution, and tissue structure continuity information; the depth image data includes changes in the surface curvature of the user tissue region. The processing unit is configured to determine the user's target tissue for scanning based on the depth image data and the first ultrasound data; and to determine the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data; wherein the occlusion type includes skeletal occlusion and intestinal gas occlusion during ultrasound scanning. The calculation unit is used to determine the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data. The control unit is used to control the robotic arm to adjust its posture according to the posture adjustment parameters so that the robotic arm moves to the target scanning position; and to scan the target scanning position with an ultrasonic probe to collect target ultrasonic data of the user tissue area. The first ultrasound data includes first ultrasound image data and first ultrasound echo data; the step of determining the occlusion type and corresponding occlusion identification information during ultrasound scanning based on the target tissue and the first ultrasound data includes: The ultrasound image data corresponding to the target scanned tissue is extracted from the first ultrasound image data to obtain target ultrasound image data; and the ultrasound echo data corresponding to the target scanned tissue is extracted from the first ultrasound echo data to obtain target ultrasound echo data. The target ultrasound image data is divided into image regions to obtain multiple ultrasound image blocks; Determine the grayscale feature parameters corresponding to each of the plurality of ultrasound image blocks to obtain a plurality of grayscale feature parameters; Based on the multiple grayscale feature parameters, the multiple ultrasound image blocks are occlusion identified to obtain the occlusion identification result; Based on the occlusion recognition results, multiple spatial connectivity states corresponding to the multiple ultrasound image blocks are determined; and based on the multiple spatial connectivity states, occlusion regions corresponding to the multiple ultrasound image blocks are determined to obtain the target occlusion region. Determine the target echo characteristics corresponding to the target obstruction area in the target ultrasonic echo data; The occlusion type of the target occlusion region is determined based on the target echo characteristics; and the occlusion identification information of the target occlusion region is determined based on the target echo characteristics and the plurality of spatial connectivity states. The step of determining the pose adjustment parameters of the robotic arm based on the occlusion type, the occlusion identification information, and the first ultrasonic data includes: The first comprehensive quality score corresponding to the first ultrasound data is determined based on the occlusion identification information; and the pose and motion constraint strategy corresponding to the occlusion type is determined. The pose motion range parameters of the robotic arm are determined according to the pose motion constraint strategy. The first state parameters of reinforcement learning corresponding to the first scanning state are determined based on the first comprehensive quality score; the first scanning state is the state during the current ultrasound scan. The first state parameter and the pose motion range parameter are input into the reinforcement learning algorithm to obtain multiple pose adjustment action strategies; Determine multiple candidate scanning poses corresponding to the multiple pose adjustment action strategies; and control the robotic arm to move to the multiple candidate scanning poses so as to collect multiple scanning results through the ultrasonic probe; Based on the multiple scanning results, a second scanning state parameter corresponding to the multiple candidate scanning poses is determined, and a second comprehensive quality score is determined based on the second scanning state parameter; Determine the change parameters of the comprehensive quality score between the second comprehensive quality score and the first comprehensive quality score; and determine the target pose adjustment action strategy based on the change parameters of the comprehensive quality score. The pose adjustment parameters of the robotic arm are determined based on the target pose adjustment action strategy.

7. A host device, characterized in that, include: Processor, memory, communication interface, and one or more programs; The one or more programs are stored in the memory and configured to be executed by the processor, the programs including instructions for performing the steps of the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1-5.

Images