Organ coverage scanning real-time planning method and device for avoiding skeleton shielding

Abstract

The invention discloses an organ coverage scanning real-time planning method and device for avoiding skeleton occlusion, and the method comprises the steps: carrying out the processing based on an obtained ultrasonic image, obtaining the coordinates of an organ region and the coordinates of a skeleton region, and updating an octree map to obtain a perception map; generating a plurality of candidate paths based on the perception map; calculating energy consumption scores of all the ranked candidate paths; performing light projection simulation on each sequenced candidate path to obtain a light projection score; performing weighted calculation on the basis of the energy consumption score and the light projection score of each sequenced candidate path to obtain a pose score; based on all the pose scores, an optimal path is obtained through screening, the ultrasonic robot is controlled to execute path nodes in the optimal path, and a new ultrasonic image is obtained; and circularly executing the process on the basis of the new ultrasonic image until full-coverage scanning of the organ is completed. Skeleton shielding can be dynamically sensed, a scanning path can be autonomously planned, and the organ coverage rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic examination, and in particular to a real-time planning method and device for organ coverage scanning that avoids bone occlusion. Background Art

[0002] In ultrasound imaging-assisted diagnosis, a crucial medical technology, complete organ visualization plays a crucial role in accurately identifying diseases. Ultrasound imaging involves transmitting ultrasound waves into the human body and receiving the reflected signals to form images. Doctors use these images to determine whether an organ is diseased. Therefore, only when the target organ is fully and clearly visualized in ultrasound images can doctors, drawing on their extensive expertise and clinical experience, conduct a detailed analysis of the organ's morphology, structure, and blood flow, ultimately leading to accurate disease identification.

[0003] However, the human body's internal structure is extremely complex, and tissues such as bones pose a significant obstacle to the propagation of ultrasound beams. Bones like the ribs and spine are hard and dense. When the ultrasound beam encounters these bones during propagation, part of the beam is reflected, scattered, or absorbed, thereby blocking the beam's propagation path. This obstruction prevents the ultrasound probe from receiving the complete reflected signal from organs in the obscured area, preventing the full imaging of some critical organs, resulting in missing or blurred images and greatly complicating the accurate diagnosis of diseases.

[0004] Traditional solutions to this problem rely heavily on the ultrasound physician's skill and experience. They constantly adjust the probe's posture and position, changing the angle of the sound beam to avoid obstructions and clearly image the target organ for accurate disease identification. Summary of the Invention

[0005] In order to dynamically sense bone occlusion and autonomously plan the scanning path, improve organ coverage, ensure image quality, and realize an intelligent and efficient ultrasound examination process, the present invention provides a real-time planning method and device for organ coverage scanning that avoids bone occlusion.

[0006] In a first aspect, an embodiment of the present invention provides a real-time planning method for organ coverage scanning that avoids bone occlusion, comprising: Processing is performed based on the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and the pre-built octree map is updated to obtain a perception map; Based on the perception map, generating a plurality of candidate paths, wherein each of the candidate paths connects the same plurality of candidate path nodes; For each candidate path, sort all candidate path nodes in the candidate path to obtain sorted candidate paths, and calculate energy consumption scores of all sorted candidate paths; Performing a ray casting simulation on each of the sorted candidate paths to obtain a ray casting score; Performing weighted calculation based on the energy consumption score and the light projection score of each sorted candidate path to obtain a posture score; Based on all the posture scores, a candidate path with the highest posture score is screened and obtained as an optimal path, and the ultrasonic robot is controlled to execute the path nodes in the optimal path and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; The above process is cyclically executed based on the new ultrasound image until the full coverage scan of the organ is completed.

[0007] Optionally, the processing based on the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and updating the pre-built octree map to obtain the perception map includes: Acquire ultrasound images; Processing the ultrasound image using a pre-trained image segmentation model to obtain a segmentation mask image of the organ region and the coordinates of the organ region; Performing shadow detection based on the segmentation mask of the organ region to obtain a shadow region in the organ region, searching along the normal direction of the shadow region toward the direction of the ultrasonic robot, obtaining a highlight region as a skeleton region, and extracting coordinates of the skeleton region; Based on the coordinates of the organ region and the coordinates of the bone region, a pre-constructed octree map is updated to obtain a perception map, wherein the perception map includes organ voxels, bone voxels, non-organ voxels and unexplored voxels.

[0008] Optionally, for each candidate path, sorting all candidate path nodes in the candidate path to obtain sorted candidate paths, and calculating energy consumption scores of all sorted candidate paths includes: For each candidate path, sorting all candidate path nodes in the candidate path using a greedy algorithm to obtain sorted path nodes and sorted candidate paths; For each of the sorted candidate paths, calculating the distance between the posture of any sorted path node and an adjacent sorted path node to obtain an initial posture score of the sorted path node; For each of the sorted candidate paths, the initial posture scores of all sorted path nodes are summed to obtain an energy consumption score.

[0009] Optionally, performing a ray casting simulation on each of the sorted candidate paths to obtain a ray casting score includes: For each of the sorted candidate paths, performing a ray projection simulation on each of the sorted path nodes, and determining whether the light propagates to the bone voxel; If yes, stop propagation and the initial score of the ray casting is recorded as 0; If not, continue propagating until it intersects with the boundary of the perception map, and calculate the number of unexplored voxels that the light passes through to obtain the initial score of the light projection; The ray projection initial scores of all sorted path nodes on the sorted candidate path are summed to obtain a ray projection score.

[0010] Optionally, the construction process of the octree map includes: Construct the original octree map; Based on the acquired current posture of the ultrasonic robot, the ultrasonic image is converted into the coordinate system of the original octree map, and the original octree map is updated to obtain an octree map.

[0011] In a second aspect, an embodiment of the present invention provides a real-time planning device for organ coverage scanning that avoids bone occlusion, comprising: An image processing module is used to process the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and to update the pre-built octree map to obtain a perception map; A path generation module, configured to generate a plurality of candidate paths based on the perception map, wherein each candidate path connects the same plurality of candidate path nodes; A first calculation module is configured to sort all candidate path nodes in each candidate path to obtain sorted candidate paths, and calculate energy consumption scores of all sorted candidate paths; a ray projection simulation module, configured to perform ray projection simulation on each of the sorted candidate paths to obtain a ray projection score; A second calculation module is configured to perform weighted calculation based on the energy consumption score and the light projection score of each of the sorted candidate paths to obtain a posture score; a path planning module, configured to screen a candidate path with the highest posture score as an optimal path based on all the posture scores, and control the ultrasonic robot to execute path nodes in the optimal path and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; The loop execution module is used to obtain a new ultrasound image and loop the above process until the full coverage scan of the organ is completed.

[0012] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the real-time planning method for organ coverage scanning that avoids bone occlusion as described in the first aspect.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the real-time planning method for organ coverage scanning that avoids bone occlusion as described in the first aspect is implemented.

[0014] In a fifth aspect, an embodiment of the present invention provides a computer program product comprising instructions, which, when executed on a computer device, enables the computer device to execute the real-time planning method for organ coverage scanning that avoids bone occlusion as described in the first aspect.

[0015] The beneficial effects of the above technical solutions provided in the embodiments of the present invention include at least: An embodiment of the present invention provides a real-time planning method for organ coverage scanning that avoids bone occlusion. A pre-constructed octree map is updated based on the coordinates of the organ area and the coordinates of the bone area to obtain a perception map. Based on the perception map, an optimal path is selected according to the posture score. Only some path nodes in the optimal path are executed, a new ultrasound image is obtained, the perception map is updated, and the path planning is re-performed based on the updated perception map to obtain a new optimal path. The use of dynamic planning and real-time feedback adjustment mechanisms can better adapt to complex in vivo environments and changes in organ positions.

[0016] The method of this embodiment can automatically and quickly locate the position of organs and bones, guide the ultrasonic robot device to scan along the optimal path in real time, improve organ coverage, ensure image quality, reduce the complexity and uncertainty of manual operation, improve the efficiency of ultrasonic scanning, save medical resources and time, enable ultrasonic examinations to be completed more quickly, and help improve the efficiency of diagnosis and treatment. Compared with traditional static path planning or simple experience-based path selection, it has higher flexibility and adaptability.

[0017] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0018] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is an example diagram of an ultrasonic robot device provided in an embodiment of the present invention; Figure 2 A flowchart of a real-time planning method for organ coverage scanning to avoid bone occlusion provided in an embodiment of the present invention; Figure 3 This is an example diagram of an ultrasound image provided in an embodiment of the present invention; Figure 4 Schematic diagram of the ultrasound probe provided in an embodiment of the present invention in different positions; Figure 5 This is an example diagram of a perception map provided in an embodiment of the present invention; Figure 6 This is an example diagram of a light projection simulation provided in an embodiment of the present invention; Figure 7 A schematic diagram of screening the optimal path provided in an embodiment of the present invention; Figure 8 Schematic diagram of a real-time planning device for organ coverage scanning that avoids bone occlusion provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0021] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," "far," "near," "front," and "back" and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0022] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0023] The inventors discovered that ultrasound imaging-assisted diagnosis often relies on the physician's skill and experience to address the problem of bones (such as ribs and spine) blocking the sound beam propagation path, preventing the full imaging of some key organs. This requires the physician to constantly adjust the probe posture and position to avoid obstructed areas. This not only increases the physician's operational burden but can also result in missed scans, affecting diagnostic accuracy. This is especially true when intercostal spaces are narrow or organ positions vary, making it difficult for manual experience to optimize the path in real time.

[0024] To solve the above problems, the inventors have conducted research and development and proposed a real-time planning method and device for organ coverage scanning that avoids bone occlusion. The method and device can dynamically sense bone occlusion and autonomously plan the scanning path, thereby improving organ coverage, ensuring image quality, and realizing an intelligent and efficient ultrasound examination process.

[0025] First, some nouns or terms that appear in the description of the embodiments of this application are subject to the following interpretations: Octree map: It is a hierarchical data structure used to efficiently represent 3D space occupancy information. Octree map recursively divides the 3D space into eight equal-sized sub-regions, each of which is called a voxel, and stores these voxels in a tree structure. The basic structure of the octree map includes: Root node: represents the entire three-dimensional space, usually a cube, which defines the initial space boundary.

[0026] ‌Child nodes‌: The root node recursively divides the space into eight equal-sized sub-cubes, each corresponding to a child node. Each child node can also continue to be divided until a certain condition (such as maximum depth or spatial resolution) is met.

[0027] Leaf nodes: When the data (such as a point cloud) in a cube is dense, or the partitioning depth reaches a preset threshold, the partitioning stops and a leaf node is generated. Leaf nodes typically store data related to that spatial region, such as point cloud density, color, or other attributes.

[0028] Example 1 This embodiment proposes a real-time planning method for organ coverage scanning to avoid bone occlusion. This method can be applied to Figure 1 The ultrasonic robot device shown in the figure may include a robotic arm and a computer. The computer deploys a real-time planning algorithm for organ coverage scanning that avoids bone occlusion, and controls the robot's motion trajectory to autonomously scan the target organ (such as the human kidney). Figure 2 The method of this embodiment may specifically include the following steps: Step S101: Processing the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and updating the pre-built octree map to obtain a perception map; Step S102: generating multiple candidate paths based on the perception map, wherein each candidate path connects the same multiple candidate path nodes; Step S103: for each candidate path, sort all candidate path nodes in the candidate path to obtain sorted candidate paths, and calculate the energy consumption scores of all sorted candidate paths; Step S104: Perform a ray projection simulation on each sorted candidate path to obtain a ray projection score; Step S105: Perform weighted calculation based on the energy consumption score and the light projection score of each sorted candidate path to obtain a posture score; Step S106: Based on all posture scores, a candidate path with the highest posture score is screened and obtained as the optimal path, and the ultrasonic robot is controlled to execute the path nodes in the optimal path and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; Step S107 : cyclically executing the above steps S101 to S106 based on the new ultrasound image until the full coverage scan of the organ is completed.

[0029] In order to more clearly explain the above-mentioned real-time planning method for organ coverage scanning to avoid bone occlusion, each step is described in detail below.

[0030] In step S101 above, the planning and decision-making of the ultrasonic robot during the scanning process need to rely on perception data. If the perception map construction time is too long, it will lead to insufficient real-time performance and inability to achieve real-time planning, causing lag and affecting scanning efficiency. Therefore, an efficient perception map construction method is needed. The specific process of constructing the perception map can include the following steps: Step S1011: Acquire an ultrasound image; In the above step S1011, refer to Figure 4 , by installing the ultrasonic probe on the ultrasonic robot to collect ultrasonic images in the current posture, we can get the following Figure 3Ultrasound image shown.

[0031] Step S1012: Process the ultrasound image using a pre-trained image segmentation model to obtain a segmentation mask of the organ region and the coordinates of the organ region; In the above step S1012, the image segmentation model can adopt the deep learning image segmentation model in the existing technology, such as U-Net, DeepLabV3+, etc. The image segmentation model is pre-trained and can segment the target organs (such as liver, kidneys, muscles, etc.) on the ultrasound image, obtain the segmentation mask map of the organ area, and extract the coordinates of the organ area from the segmentation mask map to achieve fast and accurate positioning of the organ area and improve the efficiency of perceptual map construction.

[0032] Step S1013: Perform shadow detection based on the segmentation mask of the organ region to obtain the shadow region in the organ region, search along the normal direction of the shadow region toward the direction of the ultrasonic robot, and use the obtained highlight region as the skeleton region, and extract the coordinates of the skeleton region; In step S1013, bones in the ultrasound image are highly reflective of the acoustic beam, forming a prominent highlight area and a typical acoustic shadow effect behind them. This acoustic shadow appears as a distinct dark or missing area in the organ region recognition results, becoming a key clue for identifying bone occlusion. Therefore, after successfully identifying the location of the organ region, the shadow area within the organ region is found. The normal direction of the shadow area is searched toward the direction of the ultrasound robot, and the resulting highlight area is regarded as the bone region. This highlight area corresponds to the topmost highlight area in the ultrasound image.

[0033] First, shadow detection is performed based on the segmentation mask of the organ region to identify shadow areas with significant missing or sudden grayscale drops in the organ region boundary. Specifically, the grayscale value of each area on the organ region boundary is calculated. If the grayscale value of the area is lower than the preset grayscale threshold, the area is considered to be a shadow area. The expression of the shadow area can be: In the above formula (1), represents the region of the segmentation mask image, represents the grayscale value, Indicates the preset grayscale threshold, usually set to 40.

[0034] Then, search along the normal direction of the shadow area toward the direction of the ultrasound robot. Immediately in front of the shadow area (close to the probe side), there is usually a bright area. This is because the bone surface reflects most of the sound waves back to the probe, forming a strong echo. The bright area corresponding to the ultrasound image is the bone area. The expression of the bone area can be: In the above formula (2), Indicates the search height; Indicates the height threshold of the shadow area; Indicates the search width; Indicates the height threshold of ultrasound image; Indicates the bone judgment threshold, usually set to 150.

[0035] In this embodiment, the coordinates of the ultrasound image are based on the upper left corner as the origin, and the x and y values ​​increase from the lower right corner. Only the topmost part of the ultrasound image may contain bones, so the search height is Should be greater than the height threshold of the ultrasound image , which means searching at the top of the ultrasound image and searching for the height Should be greater than the height threshold of the shadow area , indicating searching from the shaded area toward the direction of the ultrasonic robot.

[0036] Finally, the coordinate information of the identified bone region is obtained to obtain the coordinates of the bone region.

[0037] In this embodiment, a highlighted region in the ultrasound image is identified. This highlighted region and the shadowed region immediately below it form a typical "highlight-shadow" pair, a hallmark of bone in ultrasound images. Leveraging this characteristic to identify bone regions significantly improves the interference resistance and accuracy of bone detection. Bone localization can be achieved using only a single ultrasound image frame, without relying on external optical or electromagnetic tracking equipment. This reduces system complexity and cost, and avoids multi-sensor calibration errors.

[0038] Step S1014: Based on the coordinates of the organ region and the coordinates of the bone region, the pre-built octree map is updated to obtain a perception map.

[0039] In the above step S1014, the perception map includes organ voxels, bone voxels, non-organ voxels and unexplored voxels. Figure 5 In the process of updating the octree map with the coordinates of the organ region and the coordinates of the bone region, the voxels falling in the projected organ region are marked as organ voxels; the voxels falling in the projected bone region are marked as bone voxels; the voxels within the field of view of the ultrasound image (i.e., the area covered by the sound beam) but not marked as organs or bones are marked as non-organ voxels (indicating that the space is penetrable but there is no target organ); the area that has not yet been observed is marked as unexplored voxels.

[0040] The construction process of the octree map can specifically include the following steps: Step S10141: construct an original octree map; Step S10142: Based on the acquired current posture of the ultrasonic robot, the ultrasonic image is converted into the coordinate system of the original octree map, and the original octree map is updated to obtain the octree map.

[0041] In this embodiment, 2D image information (organ region coordinates and bone region coordinates) is fused into the octree map to construct a 3D semantic perception map containing organ voxels, bone voxels, non-organ voxels, and unexplored voxels, which facilitates the subsequent automatic path planning of the ultrasound robot, achieves complete coverage of the organ region, and reduces the risk of missed diagnosis.

[0042] In step S103, for each candidate path, all candidate path nodes in the candidate path are sorted to obtain sorted candidate paths, and the specific process of calculating the energy consumption scores of all sorted candidate paths may include the following steps: Step S1031: For each candidate path, use a greedy algorithm to sort all candidate path nodes in the candidate path to obtain sorted path nodes and sorted candidate paths; In step S1031, for each candidate path, all candidate path nodes are sorted using a greedy algorithm in ascending order of initial energy consumption scores to obtain sorted path nodes. All sorted path nodes constitute the sorted candidate path. The initial energy consumption score is consistent with the calculation method and principle of the energy consumption score described below (see steps S1032 and S1033), and will not be further described here.

[0043] Step S1032: For each sorted candidate path, calculate the distance between the posture of any sorted path node and the adjacent sorted path node to obtain the initial posture score of the sorted path node; Step S1033: For each sorted candidate path, the initial posture scores of all sorted path nodes are summed according to the following formula (3) to obtain the energy consumption score: In the above formula (3), It represents the initial score of the posture of the i-th sorted path node, that is, the L2 norm between the robot posture at the i-th sorted path node and the robot posture at the i-1-th sorted path node. The 0th sorted path node is the current robot posture.

[0044] In the above step S104, a specific process of performing a ray casting simulation on each sorted candidate path and obtaining a ray casting score may include the following steps: Step S1041: for each sorted candidate path, perform ray projection simulation on each sorted path node and determine whether the light propagates to the bone voxel; if so, execute step S1042; if not, execute step S1043; In the above step S1041, the propagation of sound waves is simulated using a raycast simulation method. Figure 6 , the light may pass through the bone voxels, non-organ voxels and unexplored voxels. When the light propagates to the bone voxels, it does not continue to propagate, that is, step S1042 is executed; when the light does not propagate to the bone voxels, it continues to propagate to the boundary of the perception map (i.e. Figure 6 ), and count the number of unexplored voxels passed through, i.e., execute step S1043.

[0045] Here, the spatial coordinate equation p(s) for light propagation is as follows: In the above formula (4), The coordinate vector representing the point at which the light is emitted; Indicates the direction of propagation.

[0046] Step S1042: stop propagation, and the initial score of the ray projection is recorded as 0; Step S1043: Continue propagating until it intersects the boundary of the perception map, and calculate the number of unexplored voxels that the ray passes through to obtain the initial ray projection score; Step S1044: According to the following formula (5), the initial ray projection scores of all sorted path nodes on the sorted candidate path are summed to obtain the ray projection score: In the above formula (5), Indicates the number of path nodes after sorting; is the initial score of the ray casting of the i-th sorted path node.

[0047] In this embodiment, the ray casting simulation may be performed using the algorithm shown in the following table: Table 1 Example of ray casting simulation algorithm S_total=0; Initialize node queue Q←{root node} While Q is not empty Take a node from Q Skip this node End If node∈VisitedSet Skip this node End Add node to VisitedSet If node is a leaf node If node is a bone voxel Stop the spread Break End If node memory exists in the point set Calculate the angle between the light direction and the line connecting the node center If angle < threshold S_total += node.Score End End End Else Add all child nodes of node to Q End End In the above algorithm, the core logic is to traverse the nodes in the octree that intersect with the light, and immediately stop propagation when encountering a skeleton voxel (the skeleton blocks the ultrasound); for non-skeleton leaf nodes, calculate the angle between the light direction and the node center; only when the angle is less than the threshold, accumulate the energy score of the node to calculate the energy accumulation (S_total) on the light path.

[0048] In the above step S105, the energy consumption score and the light projection score of each sorted candidate path are weighted according to the following formula (6) to obtain the posture score: In the above formula (6), represents the pose score; Indicates the number of path nodes after sorting; Indicates the preset attenuation coefficient; is the initial score of the ray projection of the i-th sorted path node; is the initial pose score of the i-th sorted path node.

[0049] Since the longer the execution distance is, the more uncertainty there is in the movement process of the sorted path node, and the shorter the execution distance is, the smaller the change is in the middle of the sorted path node, so the credibility is higher, so when calculating the posture score, the attenuation coefficient is multiplied to make the candidate node with a closer execution distance have a higher score weight, and the candidate node with a longer execution distance has a lower score weight, so that the obtained posture score is more reasonable and accurate.

[0050] In the above step S106, based on all the posture scores, the particle swarm algorithm can be used to screen the candidate path with the highest posture score as the optimal path, that is, Figure 7 The execution path shown in is the optimal path, and real-time rolling optimization is performed. Path nodes are used to search for the global path, and execution is stopped after each execution of a partial path node, that is, the number of executed path nodes is less than the number of candidate path nodes, and a new ultrasound image is obtained. The optimal search pose corresponding to the optimal path is Figure 4 In the search posture shown in FIG, the ultrasound probe is in the search posture to obtain a new ultrasound image.

[0051] In the above step S107, the process of the above steps S101 to S106 is cyclically executed based on the new ultrasound image until the full coverage scan of the organ is completed, that is, the perception map is timely updated based on the new ultrasound image and the current optimal path is planned in real time based on the latest perception map, so as to guide the ultrasound robot device in real time to avoid bone occlusion and scan the target organ.

[0052] In this embodiment, an ultrasound image is processed based on an image segmentation model to obtain a segmentation mask of the organ region and its coordinates. Shadow detection is then performed based on the segmentation mask to obtain shadow regions within the organ region. A search is then performed along the normal direction of the shadow region toward the ultrasound robot. The resulting highlighted regions corresponding to the ultrasound image are identified as skeletal regions, and the coordinates of the skeletal regions are extracted to effectively avoid skeletal occlusion. This reduces attenuation and interference during ultrasound propagation, resulting in clearer and more accurate organ images and improved diagnostic accuracy and reliability. A pre-built octree map is updated based on the coordinates of the organ and skeletal regions to obtain a perception map. Based on the perception map, an optimal path is selected based on pose scores. The optimal path is dynamically adjusted using a rolling optimization approach, where only a portion of the optimal path nodes are executed. New ultrasound images are acquired, the perception map is updated, and path planning is re-performed based on the updated perception map to obtain a new optimal path. This dynamic planning and real-time feedback adjustment mechanism can better adapt to complex in vivo environments and changes in organ position.

[0053] Traditional ultrasound scanning methods often require doctors to rely on experience and repeatedly adjust the probe position to find the appropriate scanning angle when facing bone obstruction. This method is not only time-consuming and labor-intensive, but also difficult to guarantee the optimal scanning path every time, and can easily miss important lesion information. The method of this embodiment can automatically and rapidly locate the position of organs and bones, guiding the ultrasonic robotic device to scan along the optimal path in real time. This reduces the complexity and uncertainty of manual operation, improves the efficiency of ultrasound scanning, saves medical resources and time, enables ultrasound examinations to be completed more quickly, and is conducive to improving diagnosis and treatment efficiency. Compared with traditional static path planning or simple experience-based path selection, it has greater flexibility and adaptability.

[0054] Example 2 Based on the same inventive concept, see Figure 8 The present application also proposes a real-time planning device for organ coverage scanning that avoids bone occlusion, including: An image processing module 101 is configured to process the acquired ultrasound image to obtain coordinates of the organ region and the bone region, and to update a pre-built octree map to obtain a perception map; A path generation module 102 is configured to generate a plurality of candidate paths based on the perception map, wherein each candidate path connects the same plurality of candidate path nodes; The first calculation module 103 is configured to sort all candidate path nodes in each candidate path to obtain sorted candidate paths, and calculate energy consumption scores of all sorted candidate paths; A ray casting simulation module 104 is configured to perform ray casting simulation on each sorted candidate path to obtain a ray casting score; The second calculation module 105 is configured to perform weighted calculation based on the energy consumption score and the light projection score of each sorted candidate path to obtain a posture score; A path planning module 106 is configured to screen a candidate path with the highest posture score based on all posture scores as the optimal path, control the ultrasound robot to execute path nodes in the optimal path, and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; The loop execution module 107 is used to obtain a new ultrasound image and loop the above process until the full coverage scan of the organ is completed.

[0055] The implementation principle and technical effects of the real-time planning device for organ coverage scanning that avoids bone occlusion provided in the embodiment of the present invention are similar to those of the first embodiment and will not be repeated here.

[0056] Example 3 Based on the same inventive concept, an embodiment of the present application also proposes a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the real-time planning method for organ coverage scanning that avoids bone occlusion as in Example 1 is implemented.

[0057] The computer-readable storage medium may be included in the device / apparatus described in the above embodiments, or may exist independently and not be incorporated into the device / apparatus. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the first embodiment of the present invention.

[0058] According to an embodiment of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, such as, but not limited to, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0059] Example 4 Based on the same inventive concept, an embodiment of the present application also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, a real-time planning method for organ coverage scanning that avoids bone occlusion is implemented as in Example 1.

[0060] Example 5 Based on the same inventive concept, an embodiment of the present application proposes a computer program product comprising instructions. When the computer program product is run on a computer device, the computer device executes the real-time planning method for organ coverage scanning that avoids bone occlusion in embodiment one.

[0061] The principles of solving the problems described above by the apparatus, client, medium, and related equipment in the embodiments of the present invention are similar to those of the aforementioned methods, so their implementation can refer to the implementation of the aforementioned methods, and repeated details will not be repeated.

[0062] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage) containing computer-usable program code.

[0063] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0064] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0066] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. The present disclosure is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and variations may be made without departing from the scope of the present disclosure. The scope of the present disclosure is limited solely by the appended claims. Thus, to the extent such modifications and variations fall within the scope of the claims and their equivalents, the present disclosure is intended to include such modifications and variations.

Claims

1. A real-time planning method for organ coverage scanning to avoid bone occlusion, characterized in that: include: Processing is performed based on the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and the pre-built octree map is updated to obtain a perception map; Based on the perception map, generating a plurality of candidate paths, wherein each of the candidate paths connects the same plurality of candidate path nodes; For each candidate path, sort all candidate path nodes in the candidate path to obtain sorted candidate paths, and calculate energy consumption scores of all sorted candidate paths; Performing a ray casting simulation on each of the sorted candidate paths to obtain a ray casting score; Performing weighted calculation based on the energy consumption score and the light projection score of each sorted candidate path to obtain a posture score; Based on all the posture scores, a candidate path with the highest posture score is screened and obtained as an optimal path, and the ultrasonic robot is controlled to execute the path nodes in the optimal path and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; The above process is cyclically executed based on the new ultrasound image until the full coverage scan of the organ is completed.

2. The real-time planning method for organ coverage scanning to avoid bone occlusion according to claim 1 is characterized in that: The processing is performed based on the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and the pre-built octree map is updated to obtain the perception map, including: Acquire ultrasound images; Processing the ultrasound image using a pre-trained image segmentation model to obtain a segmentation mask image of the organ region and the coordinates of the organ region; Performing shadow detection based on the segmentation mask of the organ region to obtain a shadow region in the organ region, searching along the normal direction of the shadow region toward the direction of the ultrasonic robot, obtaining a highlight region as a skeleton region, and extracting coordinates of the skeleton region; Based on the coordinates of the organ region and the coordinates of the bone region, a pre-constructed octree map is updated to obtain a perception map, wherein the perception map includes organ voxels, bone voxels, non-organ voxels and unexplored voxels.

3. The real-time planning method for organ coverage scanning to avoid bone occlusion according to claim 2, characterized in that: For each candidate path, sorting all candidate path nodes in the candidate path to obtain sorted candidate paths, and calculating energy consumption scores of all sorted candidate paths, including: For each candidate path, sorting all candidate path nodes in the candidate path using a greedy algorithm to obtain sorted path nodes and sorted candidate paths; For each of the sorted candidate paths, calculating the distance between the posture of any sorted path node and an adjacent sorted path node to obtain an initial posture score of the sorted path node; For each of the sorted candidate paths, the initial posture scores of all sorted path nodes are summed to obtain an energy consumption score.

4. The real-time planning method for organ coverage scanning to avoid bone occlusion according to claim 3 is characterized in that: The performing of a ray casting simulation on each of the sorted candidate paths to obtain a ray casting score includes: For each of the sorted candidate paths, performing a ray projection simulation on each of the sorted path nodes, and determining whether the light propagates to the bone voxel; If yes, stop propagation and the initial score of the ray casting is recorded as 0; If not, continue propagating until it intersects with the boundary of the perception map, and calculate the number of unexplored voxels that the light passes through to obtain the initial score of the light projection; The ray projection initial scores of all sorted path nodes on the sorted candidate path are summed to obtain a ray projection score.

5. The real-time planning method for organ coverage scanning to avoid skeleton occlusion according to claim 2, characterized in that: The construction process of the octree map includes: Construct the original octree map; Based on the acquired current posture of the ultrasonic robot, the ultrasonic image is converted into the coordinate system of the original octree map, and the original octree map is updated to obtain an octree map.

6. A real-time planning device for organ coverage scanning that avoids bone occlusion, characterized in that: include: An image processing module is used to process the acquired ultrasound image to obtain the coordinates of the organ region and the coordinates of the bone region, and to update the pre-built octree map to obtain a perception map; A path generation module, configured to generate a plurality of candidate paths based on the perception map, wherein each candidate path connects the same plurality of candidate path nodes; A first calculation module is configured to sort all candidate path nodes in each candidate path to obtain sorted candidate paths, and calculate energy consumption scores of all sorted candidate paths; a ray projection simulation module, configured to perform ray projection simulation on each of the sorted candidate paths to obtain a ray projection score; A second calculation module is configured to perform weighted calculation based on the energy consumption score and the light projection score of each of the sorted candidate paths to obtain a posture score; a path planning module, configured to screen a candidate path with the highest posture score as an optimal path based on all the posture scores, and control the ultrasonic robot to execute path nodes in the optimal path and acquire a new ultrasound image, wherein the number of executed path nodes is less than the number of candidate path nodes; The loop execution module is used to obtain a new ultrasound image and loop the above process until the full coverage scan of the organ is completed.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the real-time planning method for organ coverage scanning avoiding bone occlusion as described in any one of claims 1 to 5 is implemented.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the real-time planning method for organ coverage scanning that avoids bone occlusion is implemented as described in any one of claims 1-5.

9. A computer program product comprising instructions, which, when executed on a computer device, causes the computer device to execute the real-time planning method for organ coverage scanning avoiding bone occlusion according to any one of claims 1 to 5.

Images