Medical system, collision detection method thereof, computer equipment and storage medium thereof

Abstract

According to the medical system, the collision detection method thereof, the computer equipment and the storage medium thereof, rapid boundary comparison and accurate projection analysis are combined through a dual-algorithm cooperative detection mechanism, and the pose data of the moving part are acquired in real time in the operation of the medical equipment and are converted into the unified reference coordinate system; and millisecond collision risk early warning is realized.

Description

Technical Field

[0001] This application relates to the field of medical device safety control technology, and in particular to a medical system and its collision detection method, and a computer device and its storage medium. Background Technology

[0002] During the operation of medical equipment, the risk of collisions between moving parts such as rotating frames and lifting beds and fixed parts or environmental obstacles urgently needs to be addressed. Current technologies struggle to balance real-time performance and accuracy with single collision detection algorithms: fast collision detection algorithms are prone to false negatives, while high-precision algorithms suffer from excessive computational loads that cannot meet millisecond-level response requirements. Furthermore, the lack of a unified processing framework for various types of detection objects in medical scenarios—including moving parts, fixed bases, and environmental entities—leads to insufficient adaptability to complex environments. In addition, factors such as equipment aging and component wear can cause fixed safety thresholds to fail, and existing systems lack dynamic risk perception mechanisms. These deficiencies create dual risks of equipment damage and even patient safety, necessitating the development of efficient and reliable collision detection solutions. Summary of the Invention

[0003] Therefore, it is necessary to provide a medical system and its collision detection method, as well as a computer device and its storage medium, to address the above problems.

[0004] In a first aspect, the present invention provides a collision detection method for a medical system, comprising: acquiring real-time spatial pose data of a component to be detected; converting the real-time spatial pose data of the component to be detected into a reference coordinate system; performing preliminary collision detection on the component to be detected based on a first collision detection algorithm to determine a preliminary collision detection result; performing secondary collision detection based on the preliminary collision detection result and in combination with a second collision detection algorithm to determine a secondary collision detection result; and outputting a collision detection conclusion based on the preliminary collision detection result and / or the secondary collision detection result.

[0005] In one embodiment, the preliminary collision detection of the component to be detected based on the first collision detection algorithm and the determination of the preliminary collision detection result includes: determining the preliminary collision detection result by fast boundary screening based on the coordinate extreme value comparison method.

[0006] In one embodiment, the step of performing secondary collision detection based on the preliminary collision detection result and in combination with a second collision detection algorithm to determine the secondary collision detection result includes: using the split axis projection method to perform precise geometric analysis on the parts to be detected that have a collision risk in the preliminary collision detection result to determine the secondary collision result.

[0007] In one embodiment, the component to be detected includes one or more of a moving component, a fixed component, and an environmental entity.

[0008] In one embodiment, the first collision detection algorithm and / or the second collision detection algorithm further include: performing full combination collision detection on all components to be detected, wherein the full combination collision detection is a strategy of performing collision detection on all components to be detected by pairing them together.

[0009] In one embodiment, the reference coordinate system is a coordinate system established with the medical device in the medical system as the origin.

[0010] In one embodiment, the collision detection method further includes: determining a preset collision judgment threshold based on a minimum safe space distance threshold between the components to be detected.

[0011] In one embodiment, the preset collision judgment threshold is adaptively adjusted based on at least one parameter among device running time, component wear data, and historical collision records.

[0012] In one embodiment, the collision detection method further includes: determining a safety margin parameter between the components to be detected based on the preliminary collision detection result and / or the secondary collision detection result; comparing the safety margin parameter with the preset collision judgment threshold to determine the collision detection conclusion.

[0013] In one embodiment, the component to be detected includes a convex polygonal component and a non-convex polygonal component.

[0014] For non-convex polygonal components, convexity preprocessing is performed: the components are divided into multiple convex polygonal sub-components by decomposition, or the minimum circumscribed convex hull is constructed by compensation method for component compensation.

[0015] In one embodiment, when the number of decomposed sub-components exceeds a preset number, the method switches to a compensation method.

[0016] In a second aspect, the present invention provides a collision detection system for a medical system, comprising: an acquisition module for acquiring real-time spatial pose data of a component to be detected; a coordinate system transformation module for transforming the real-time spatial pose data to a unified reference coordinate system; a collision detection module for executing a first collision detection algorithm and / or a second collision detection algorithm to determine preliminary collision detection results and / or secondary collision detection results; and a decision output module for generating and outputting collision detection conclusions.

[0017] In one embodiment, the collision detection module includes: a fast detection unit for executing a first collision detection algorithm to determine a preliminary collision detection result; and a precise verification unit for executing a second collision detection algorithm to determine a secondary collision detection result.

[0018] In one embodiment, the collision detection system of the medical system further includes a threshold adjustment module for dynamically adjusting a preset collision judgment threshold based on at least one parameter among equipment operating time, component wear data, and historical collision records.

[0019] In one embodiment, the coordinate system transformation module includes a dynamic calibration unit for periodically verifying the position of the origin of the reference coordinate system using a laser rangefinder.

[0020] In a third aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the collision detection algorithm of the medical system described above.

[0021] In a fourth aspect, the present invention provides a medical system comprising a patient support device for supporting a patient; a medical device for treating / diagnosing the patient; and the aforementioned computer-readable storage medium and / or collision detection system.

[0022] This invention employs a dual-algorithm collaborative detection mechanism, combining rapid boundary comparison with precise projection analysis to acquire real-time pose data of moving parts during medical device operation and convert it to a unified reference coordinate system, achieving millisecond-level collision risk warning. For non-convex geometric components in medical devices, an adaptive convexity processing strategy is adopted, intelligently switching between decomposition and compensation modes based on complexity. Simultaneously, a dynamic threshold adjustment system is established, integrating multi-dimensional parameters such as device operating time, component wear data, and historical collision records to optimize safety margin judgment criteria in real time. A comprehensive collision detection framework covers the collision probability of all moving parts, fixed bases, and environmental entities, forming a closed-loop protection system from data acquisition and algorithm judgment to risk decision-making. This technical solution significantly improves the operational safety and environmental adaptability of medical devices, maintaining high-efficiency response capabilities while ensuring detection accuracy, effectively avoiding the risks of false positives and false negatives, extending equipment lifespan, and ensuring patient safety, providing reliable all-round collision protection for medical devices. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of a medical system in some embodiments of the present invention.

[0024] Figure 2 This is a flowchart illustrating a collision detection method for a medical system in some embodiments of the present invention.

[0025] Figure 3 This is a schematic diagram illustrating the process of converting the real-time spatial pose data of the component to be detected to a reference coordinate system in some embodiments of the present invention.

[0026] Figure 4a This is a schematic diagram of the motion state of the probe in the SPECT device in some embodiments of the present invention.

[0027] Figure 4b This is a schematic diagram of the motion state of the probe in the SPECT device in some embodiments of the present invention.

[0028] Figure 5 This is a schematic diagram of the process for preliminary collision detection in some embodiments of the present invention.

[0029] Figure 6 This is a schematic diagram of the process for secondary collision detection in some embodiments of the present invention.

[0030] Figure 7 This is a schematic diagram of the split-axis projection method in some embodiments of the present invention.

[0031] Figure 8 This is a schematic diagram of the architecture of a medical device collision detection system in some embodiments of the present invention.

[0032] Figure 9 This is a schematic diagram of the architecture of a computer device in some embodiments of the present invention. Detailed Implementation

[0033] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. However, those skilled in the art should understand that this application can be implemented without these details. In other instances, to avoid unnecessarily obscuring various aspects of this application, well-known methods, processes, systems, components, and / or circuits have been described at a higher level. It will be apparent to those skilled in the art that various changes can be made to the disclosed embodiments, and the general principles defined in this application can be applied to other embodiments and application scenarios without departing from the principles and scope of this application. Therefore, this application is not limited to the embodiments shown, but conforms to the broadest scope consistent with the scope of the claims.

[0034] The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising” and “including” as used herein only indicate the presence of the stated feature, integral, step, operation, component, and / or part, but do not exclude the presence or addition of at least one other feature, integral, step, operation, component, part, and / or combination thereof.

[0035] It will be understood that the terms “system,” “engine,” “unit,” “module,” and / or “block” used herein are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels in ascending order. However, these terms may be replaced with other expressions if they serve the same purpose.

[0036] Generally, the terms “module,” “unit,” or “block” as used herein refer to a collection of logical or software instructions embodied in hardware or firmware. The modules, units, or blocks described herein may be implemented as software and / or hardware and may be stored in any type of non-transitory computer-readable medium or other storage device. In some embodiments, software modules / units / blocks may be compiled and linked into an executable program. It will be appreciated that software modules may be invoked from other modules / units / blocks or from themselves, and / or may be invoked in response to detected events or interrupts. Software modules / units / blocks configured to execute on a computing device (e.g., such as…) Figure 8-9 The processor (shown herein) may be located on a computer-readable medium, such as an optical disc, digital video disc, flash drive, hard disk, or any other tangible media, or as a digital download (and may be initially stored in a compressed or installable format that requires installation, decompression, or decryption prior to execution). Such software code may be stored, partially or wholly, on the storage device of the executing computing device and applied to the operation of the computing device. Software instructions may be embedded in firmware, such as EPROM. It will also be appreciated that hardware modules / units / blocks may be included in connected logical components, such as gates and flip-flops, and / or may be included in programmable units, such as programmable gate arrays or processors. The modules / units / blocks or computing device functions described herein may be implemented as software modules / units / blocks, or may be represented in hardware or firmware. Typically, the modules / units / blocks described herein may be combined with other modules / units / blocks, or, although physically organized or stored, may be divided into submodules / subunits / subblocks. This description may apply to a system, an engine, or a portion thereof.

[0037] It will be understood that when a unit, engine, module, or block is referred to as being "on," "connected," or "coupled to" another unit, engine, module, or block, it may be directly on, connected to, coupled to, or communicate with the other unit, engine, module, or block, or there may be an intermediate unit, engine, module, or block, unless the context clearly indicates otherwise. In this application, the term "and / or" may include any one or more of the relevant listed items or a combination thereof.

[0038] This application provides a collision detection algorithm for a medical system, which can be applied to, for example... Figure 1In the application environment shown, terminal 150, server 120, medical system 110, and storage device 140 establish communication and data connections with each other via a network. Medical system 110 acquires real-time spatial pose data of the component to be detected and sends the data to server 120 for processing. Server 120 calls the pre-stored reference coordinate system parameters and component geometric model library in storage device 140 to execute coordinate transformation and a two-level collision detection algorithm: first, a coordinate extreme value comparison method is used to quickly screen for collision risks; then, a separation axis projection method is triggered to accurately verify potential collision combinations. Storage device 140 simultaneously records device running time, component wear data, and historical collision events, and improves algorithm efficiency by caching the separation axis optimal direction set. Finally, the collision conclusion is returned by server 120 to control operation terminal 150, which generates three-level control commands and overlays them with medical image data for visualization. In this process, storage device 140 acts as the core data warehouse, ensuring the persistence of reference parameters, the continuity of historical learning, and the system's fault recovery capabilities, forming a closed-loop technical solution from data storage and algorithm execution to decision output. Among them, storage device 140 is a disk array or distributed cloud storage system, control and operation terminal 150 includes but is not limited to mobile devices such as computers and smartphones, and server 120 is implemented by independent server or cluster.

[0039] It should be noted that in some embodiments, the collision detection algorithm can be executed independently by the medical system 110, that is, the medical system 110 completes all the processing by itself, but the data can be synchronized to the server 120 and the terminal 150, or the medical system 110 and the server 120 can be executed collaboratively.

[0040] In some embodiments, such as Figure 2 As shown, a collision detection algorithm for a medical system is provided, including the following steps:

[0041] S201, Obtain the real-time spatial pose data of the component to be tested; S202, convert the real-time spatial pose data of the component to be detected into a reference coordinate system; S203, perform preliminary collision detection on the component to be detected based on the first collision detection algorithm, and determine the preliminary collision detection result; S204, Based on the preliminary collision detection results, and combined with the second collision detection algorithm, perform secondary collision detection to determine the secondary collision detection results; S205, Based on the preliminary collision detection results and the secondary collision detection results, output the collision detection conclusion.

[0042] In this application, the medical system includes at least one or more of the following: magnetic resonance imaging (MRI) equipment, computed tomography (CT) equipment, positron emission tomography (PET) equipment, radiation therapy (RT) equipment, and single-photon emission computed tomography (SPECT) equipment. The component to be detected includes at least one or more of the following: a moving component, a stationary component, and an environmental entity.

[0043] In some embodiments, in S201, acquiring the real-time spatial pose data of the component to be detected includes: during the collision detection initiation phase, firstly, acquiring the spatial pose data of the component to be detected in real time through a multi-source sensor system deployed in the medical device and / or environment (the environment may include the environmental space where the medical device is located, such as a scanning room or a workroom).

[0044] In some embodiments, the moving parts may be parts that can move relative to the medical device, including but not limited to rotating gantry, movable treatment head / probe, robotic arm and other movable parts mounted on the medical device, as well as patient support devices that can move relative to the medical device; the fixed parts may be parts that do not need to move relative to the medical device, including at least fixed gantry, device base, bracket and other immovable or non-movable parts mounted on the medical device; the environmental entities may be other objects that are not part of the medical device but are located near the medical device, including, for example, walls, device operators, patients and other people or objects that are not part of the medical system itself but are present in the diagnosis and treatment process in the scanning environment of the medical system.

[0045] In some embodiments, the spatial pose data can characterize the position and orientation information of the component to be detected in three-dimensional space. Real-time characterization of this data involves continuous acquisition or continuous updating. The spatial pose data includes at least: spatial position coordinates and rotational attitude angles. The spatial position coordinates are three-dimensional coordinate values ​​(X, Y, Z) describing the component to be detected in three-dimensional space, and can be denoted as P. The rotational attitude angles characterize the rotational state of the component to be detected, and can be denoted as R. The rotational state includes at least: a principal rotation angle θ and a rotation angle γ, wherein the principal rotation angle θ characterizes the amount of rotation of the component to be detected around the Z-axis of the coordinate system, and the rotation angle γ characterizes the amount of deflection of the component to be detected's own rotation.

[0046] See Figure 4a and Figure 4b Taking the movement of probe 111 on the gantry in a SPECT device as an example, see [link to relevant documentation]. Figure 4a X, Y, Z are the initial coordinate system; S, T, Z are the coordinate system after probe 111 has rotated θ degrees around the Z-axis (also known as the axis in medical equipment) on the gantry; X1, Y1, Z1 are the coordinate system of the first probe itself; X2, Y2, Z2 are the coordinate system of the second probe itself; and γ is the deflection angle of the probe's rotation in its own coordinate system. See also Figure 4b S, T', and Z form the coordinate system after probe 111 rotates θ degrees around the Z-axis on the frame and then translates along the S-axis. T represents the translation distance. This can be understood as... Figure 4b The translation direction is merely an example; the probe can translate in any direction. Therefore, T represents the radial offset. It can be understood that the radial offset represents the displacement of the component under test in the reference coordinate system. In other words, during coordinate transformation, it represents the displacement of the component's own coordinate system origin relative to the reference coordinate system origin in the X / Y / Z directions.

[0047] In this embodiment, the moving parts in the medical system can be equipped with high-precision encoders or sensors to provide real-time and continuous feedback of their spatial position coordinates P and rotational attitude angle R. For example, encoders or sensors can be used to continuously provide feedback of the spatial pose data of the moving parts at a frequency of 100Hz.

[0048] In some embodiments, the fixed components can obtain static pose data through a preset pose parameter library. Specifically, during the installation phase, a laser tracker can be used to measure the precise coordinates of feature points (e.g., frame corners, base positioning points, etc.) of each fixed component within the medical device. Based on these feature point coordinates, a three-dimensional geometric model of each fixed component can be constructed, forming a preset pose parameter library. In subsequent collision detection, the corresponding geometric model of the fixed component can be directly retrieved for collision detection calculations, eliminating the need for real-time sensing and monitoring to obtain the static pose of the fixed components.

[0049] The set of key vertex coordinates representing the geometric shape of the component's outline, as described by the feature points, can be denoted as: {V1,V2,…,Vn} represents the coordinates of the four corner points of a rectangular component.

[0050] In this embodiment, the origin position of the reference coordinate system can be periodically checked when performing feature point matching. For example, the laser ranging device is triggered to check the origin position every 5 seconds. When the actual measurement deviates from the initial origin position or the theoretical origin position, the transformation matrix parameters in the feature point matching process are automatically updated.

[0051] For example, to simplify subsequent calculations, after obtaining the 3D geometric model of the fixed components, the minimum bounding box parameters and spatial orientation of each fixed component can be calculated. Then, the maximum boundary values ​​(Xmax, Ymax, Zmax) and minimum boundary values ​​(Xmin, Ymin, Zmin) of their bounding boxes along the 3D coordinate axes, along with the orientation vectors of each fixed component, are pre-stored to form a preset pose parameter library. In subsequent collision detection, the parameters in this database can be directly called for detection calculations, saving real-time computing resources.

[0052] In some embodiments, environmental entities can be captured in real time using, for example, laser rangefinders and / or vision sensors. In some embodiments, optical reflective markers can be placed on the surface of the environmental entity, and real-time spatial pose data can be obtained by photographing or capturing these optical marker points at a fixed frequency using an infrared optical tracking system deployed in the environment. In some embodiments, vision sensors such as 3D cameras or depth cameras can also be used to photograph the environmental entity in real time at a fixed frequency to obtain real-time point cloud data, and then perform model matching calculations using the ICP algorithm to obtain spatial pose data.

[0053] For example, continuing with SPECT, spatial pose data of the rotating probe, moving bed, patient, and scanning room walls can be acquired. After filtering and noise reduction, all spatial pose data is transmitted to the processing core to form a pose dataset containing spatial coordinates P=(X, Y, Z) and rotation angles R=[θ, γ], providing complete dynamic input for subsequent coordinate system normalization. In some embodiments, see... Figure 3 In S202, the real-time spatial pose data of the component to be detected is converted to a reference coordinate system, including: S2021, Determine the reference coordinate system; S2022, the real-time spatial pose data of the component to be tested is converted to the reference coordinate system to obtain the reference spatial pose data of each component to be tested in the reference coordinate system.

[0054] In this embodiment, to achieve effective fusion and comparison of multi-source data, the coordinate systems of all components to be detected can be transformed into a unified reference coordinate system. The reference coordinate system refers to a fixed three-dimensional Cartesian coordinate system used to uniformly represent the spatial pose data of all components to be detected.

[0055] In some embodiments, in S2021, the determination of the reference coordinate system can, for example, use the isocenter point of the medical device as the origin of the reference coordinate system, i.e., the coordinate system of the medical device as the reference coordinate system, to reduce the computational load during coordinate system transformation. In some embodiments, physical marker points in the scanning area (e.g., corners, positioning points, etc.) can be used as the origin of the reference coordinate system, i.e., the coordinate system of the scanning area as the reference coordinate system. In some embodiments, the geometric center of the hospital bed can be used as the origin of the reference coordinate system, i.e., the coordinate system where the hospital bed is located as the reference coordinate system. In some embodiments, an independent coordinate system can be established through a calibration object, i.e., a self-defined virtual coordinate system can be used as the reference coordinate system. It is understood that the establishment of the reference coordinate system can be adjusted and selected according to implementation requirements, and is not limited here.

[0056] In this embodiment, in S2022, the real-time spatial pose data of the component to be detected is transformed into a reference coordinate system to obtain the reference spatial pose data of each component to be detected in the reference coordinate system. The specific transformation method of the coordinate system can be dynamically adapted according to the type of the component to be detected and the selection of the base coordinate system. For example, for moving components, the effects of rotation and principal rotation angles need to be compensated; based on the selected base coordinate system, the transformation process includes at least rotation compensation and displacement offset compensation. For fixed components, the static pose is obtained based on a preset pose parameter library; based on the selected base coordinate system, the transformation process includes at least coordinate mapping. For environmental entities, the transformation process is based on real-time measured spatial pose data, and its correlation with the sensing method used to measure the spatial pose data.

[0057] In some embodiments, when the reference coordinate system is the coordinate system of the medical device, for moving parts, coordinate system transformation can be achieved through rotation matrix operations, including at least principal rotation angle compensation, rotation angle compensation, and radial offset superposition compensation. The superposition compensation can be implemented using vector addition. For environmental entities, a laser measurement point matching algorithm combined with least squares optimization can be used to obtain spatial transformation parameters and achieve coordinate system transformation. For fixed parts, since fixed parts are usually located on the medical device, when the coordinate system of the medical device is selected as the reference coordinate system, the static pose data of the fixed parts in a preset pose parameter library can be directly retrieved. After the above-described coordinate system transformations, the reference spatial pose data of each component to be detected in the reference coordinate system can be obtained.

[0058] In this embodiment, the specific steps for coordinate transformation of the moving part are as follows: The acquired real-time spatial pose data P is first compensated for the rotation angle γ by the rotation matrix R(γ), then the radial offset vector T is superimposed, and finally transformed to the reference coordinate system by the main rotation matrix R(θ) to obtain the transformed reference spatial pose data P' of the moving part, that is: P'=Rθ×Rγ×P+T.

[0059] For example, regarding the coordinate transformation of a fixed component, since the fixed component is usually located on a medical device, when the coordinate system of the medical device is selected as the reference coordinate system, the spatial pose data P of the fixed component itself can be considered as the reference spatial pose data P' in the reference coordinate system, that is: P=P'.

[0060] For example, coordinate transformation of an environmental entity can be performed using a feature point matching algorithm, which is a method for aligning the measured pose data with a reference coordinate system. For example, firstly, feature matching is performed by aligning the laser measurement points with the reference coordinate system to obtain the spatial pose data P of the environmental entity to be measured in the current coordinate system. Then, the transformation matrix M between the environmental entity and the reference coordinate system is obtained using the least squares method. Based on the transformation matrix M, the spatial pose data P is transformed into reference spatial pose data P' in the reference coordinate system.

[0061] In some embodiments, when the reference coordinate system is the scanning room coordinate system, the corner of the scanning room can be selected as the origin. For moving parts, it is necessary to calculate the three-dimensional spatial offset of the center point of the medical device relative to the origin of the scanning room coordinate system. The position of the moving part in its own coordinate system is compensated for by rotation angle and principal rotation angle, and then the three-dimensional spatial offset is added to obtain the reference spatial pose data of the moving part in the scanning room coordinate system. For fixed parts, the static pose data of the fixed parts in the pre-stored pose parameter library can be directly read, and then the three-dimensional spatial offset of the center point of the medical device relative to the origin of the scanning room coordinate system is added to obtain the reference spatial pose data of the fixed parts in the scanning room coordinate system. For environmental entities, their three-dimensional coordinates in the scanning room coordinate system can be directly obtained through sensor devices installed in the scanning room.

[0062] In some embodiments, when the reference coordinate system is the bed coordinate system, it is established with the intersection of the drive axes at the initial position of the bed as the origin. The moving component obtains its spatial pose data relative to the current actual position of the bed by subtracting the current position offset of the bed from its coordinates in the device coordinate system; the fixed component maps its spatial pose data in the device coordinate system to the bed coordinate system through a preset spatial transformation matrix, wherein the preset spatial transformation matrix can be determined by measuring the relative position of the medical device and the bed during the initial installation of the device; the environmental entities can be directly acquired by sensors installed on the bed or other fixed positions.

[0063] In some embodiments, the coordinate system of the medical device is typically used as the reference coordinate system to reduce computational load.

[0064] See also Figure 5 In S203, the preliminary collision detection of the component to be detected based on the first collision detection algorithm, and the determination of the preliminary collision detection result, may include the following steps: Based on the coordinate extreme value comparison method, preliminary collision detection results are determined through rapid boundary screening.

[0065] In this embodiment, determining the preliminary collision detection results through rapid boundary screening based on the coordinate extreme value comparison method may include the following steps: S2031, the reference spatial pose data of each component to be tested in the reference coordinate system are combined into a set; S2032, Based on the set, determine the maximum and minimum boundary values ​​of each component to be detected in each coordinate axis direction in the reference coordinate system, so as to determine the bounding box of each component to be detected in the reference coordinate axis direction; S2033, Perform full-combination collision detection on all components to be tested to determine the preliminary collision detection results.

[0066] In some embodiments, in S2033, the full combination collision detection is a strategy of performing pairwise pairwise detection on all moving parts, stationary parts, and environmental entities in the medical system. The full combination collision detection on all parts to be detected involves dividing the parts to be detected into pairs, traversing all possible combinations of the parts to be detected, performing collision detection on any pair of parts to be detected, and determining whether the two parts will collide or if there is a potential risk of collision.

[0067] In some embodiments, in S2031, a set of reference spatial pose data of each component to be detected in the reference coordinate system is obtained, consisting of P1', P2', ..., Pn', where Pn' represents the reference spatial pose data of the nth component to be detected in the reference coordinate system.

[0068] In some embodiments, in S2032, based on the reference spatial pose data of each component to be detected in the reference coordinate system, a set P1', P2', ..., Pn' is formed, and the maximum boundary values ​​Xmax, Ymax, Zmax and minimum boundary values ​​Xmin, Ymin, Zmin of each component to be detected in each coordinate axis direction are determined respectively, and then a bounding box of each component to be detected in the reference coordinate axis direction is constructed.

[0069] In some embodiments, in S2033, determining the preliminary collision detection result may include: if a component to be detected satisfies the following condition in at least one coordinate axis direction: the minimum boundary value of one axis is greater than the maximum boundary value of the other axis, then it is determined that there is no collision risk or no collision has occurred between the components to be detected. If all coordinate axes of a component to be detected do not satisfy the above conditions, then it can be determined that the component to be detected may collide or has a potential collision risk. Thus, preliminary collision detection results can be obtained after preliminary collision detection between all components to be detected.

[0070] In some embodiments, components to be tested that do not pose a collision risk or have not been involved in a collision in the preliminary collision detection results can be marked as safe components, and the results can be recorded and stored to form a list of safe components; components to be tested that may be involved in a collision or have a potential collision risk can be marked as risky components to form a list of collision components.

[0071] In some embodiments, in S204, the step of performing secondary collision detection based on the preliminary collision detection result and in combination with the second collision detection algorithm to determine the secondary collision detection result may further include the following step: if the preliminary collision detection result indicates that there is a potential collision risk, secondary collision detection can be performed based on the second collision detection algorithm to determine the secondary collision detection result.

[0072] In other words, secondary collision detection can be performed based on the results of the initial collision detection. That is, the parts to be detected that are marked as potentially colliding or having potential collisions in the initial collision detection results (i.e., the parts to be detected in the collision part list) are obtained, and the second collision detection is performed to further determine the probability of collision between each part to be detected.

[0073] For ease of understanding and concise explanation, the component to be detected in secondary collision detection will be referred to as the secondary component to be detected in the following text.

[0074] In some embodiments, see Figure 6 The second collision detection algorithm can employ the split-axis projection method, determining the secondary collision result through precise geometric analysis of the secondary component to be detected. Further steps may include: S2041, Obtain the list of components to be inspected in the second test; S2042, calculate the projection range for the separation axis direction corresponding to each secondary component to be tested, and obtain the projection range of each secondary component to be tested in each separation axis direction.

[0075] S2043, Perform full combination collision detection on all secondary components to be tested, and obtain the secondary collision detection results.

[0076] In this embodiment, the full combination collision detection in S2043 has been described elsewhere in this document and will not be repeated here.

[0077] In some embodiments, in S2042, the projection interval is calculated for the separation axis direction corresponding to each secondary component to be detected, and the projection interval of each secondary component to be detected in each separation axis direction is obtained. Specific steps include: performing a dot product operation between the coordinates of all feature points of each secondary component to be detected and the unit vector of the separation axis to obtain a set of projection values, and then determining the projection interval. Further, specific judgment conditions may include: if a pair of secondary components to be detected do not overlap in the projection interval of any separation axis, it can be considered that the pair of secondary components to be detected has no collision risk or has not collided; conversely, when a pair of secondary components to be detected overlaps in the projection intervals of all separation axes, it can be considered that the pair of secondary components may collide or has a collision risk.

[0078] In some embodiments, the split-axis projection method is more effective for convex polygons. Therefore, before step S2042, the secondary detection components can be classified according to their geometric characteristics. For example, each secondary detection component can be classified into convex polygon components and non-convex polygon components based on its geometric characteristics. The convex polygon component is a geometric component characterized by all interior angles not exceeding 180 degrees and the line connecting any two points located inside the polygon. In this embodiment, when performing the second collision detection on the convex polygon component, the normal directions of all its edges can be directly selected as the split axis, and then the above calculations are performed to obtain the secondary collision results.

[0079] See Figure 7 Let edges 1 to 7 be the edges of convex polygons A and B, and separation axes 1' to 7' be the axes corresponding to the normal directions of edges 1 to 7. Taking edge 1 as an example, the vector of edge 1 is calculated based on the coordinates of its two vertices, and then the normal vector of edge 1 is obtained as separation axis 1'. Then, the vectors formed by all vertices of polygons A and B and the origin are projected onto this separation axis, and the minimum and maximum values ​​(Pmin, Pmax) of the two polygon vertices projected onto the separation axis are recorded to form a projection line segment. It is then determined whether the projection line segments of the two parts on separation axis 1' overlap. If they do not overlap, then (PAmax)<PBmin)||(PAmin> PBmax indicates that there exists a straight line separating two polygons, and the two polygons do not intersect. If it is to determine overlap, it is necessary to continue traversing other separation axes. If the projections on any separation axis do not overlap, it means that A and B do not intersect. If the projection line segments on any separation axis overlap, it means that there is a risk of collision between the two components.

[0080] In some embodiments, non-convex polygonal components can be first subjected to convexification, which involves decomposing the non-convex polygon into multiple convex sub-components or constructing a minimum circumscribed convex hull. Then, the same processing method as for convex polygonal components is performed, i.e., selecting the normal directions of all edges as the separation axis.

[0081] For example, when performing convexification on a non-convex polygonal component, a decomposition method can be used, that is, dividing the non-convex polygonal component into multiple convex polygonal sub-components.

[0082] For example, when performing convexification on a non-convex polygon component, a compensation method can be used. That is, the minimum circumscribed convex hull of the non-convex polygon component is constructed, and the non-convex polygon component is compensated into a convex polygon component based on the minimum circumscribed convex hull. Here, the minimum circumscribed convex hull is a convex polygon with the smallest area that completely encloses the contour of the non-convex component, and its vertices are formed by the outermost feature points of the original component.

[0083] In some embodiments, the decomposition method and the compensation method can be combined to quickly transform non-convex polygonal parts into convex polygonal parts.

[0084] In some embodiments, when the number of convex polygonal sub-components after decomposition exceeds a preset number, the compensation method can be switched in real time for convexification processing.

[0085] In some embodiments, in S205, the safety component list and the collision component list can be updated based on the secondary collision detection results obtained above, to form a collision detection conclusion.

[0086] In some embodiments, a minimum safe space distance threshold δ between components can be preset as a collision judgment threshold to determine whether there is a collision risk between the components to be detected.

[0087] In this embodiment, the preset collision judgment threshold can be set by the safety standards of the device or component. Furthermore, the value can be dynamically adjusted based on at least one parameter among the device operating time, component wear data, and historical collision records. For example, the initial preset collision judgment threshold of the CT device is 9mm. After 600 hours of cumulative operation, the threshold can be appropriately lowered to 8.5mm based on the device wear condition to achieve dynamic optimization and balance sensitivity and false alarm rate.

[0088] In this embodiment, to accurately quantify the collision risk, a safety margin parameter can be introduced, which is the actual minimum spatial distance S between the components to be detected. When S≥δ, it indicates that the two components to be detected are in a safe state and there is no collision risk; when S<δ, it indicates that there is a collision risk between them, and a collision warning or braking command can be further triggered.

[0089] In the collision detection method of the above-mentioned medical system, a dual-algorithm collaborative detection mechanism is used to combine rapid boundary comparison with precise projection analysis. In the operation of medical equipment, the pose data of moving parts are acquired in real time and converted to a unified reference coordinate system to achieve millisecond-level collision risk warning.

[0090] It should be understood that, although Figure 2-6 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2-6 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0091] Based on the same inventive concept, this application also provides a collision detection system for implementing the aforementioned medical device collision detection method. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations of the medical device collision detection system embodiment provided below can be found in the limitations of the medical device collision detection method described above, and will not be repeated here.

[0092] In some embodiments, such as Figure 8 As shown, a medical device collision detection system 5 is provided, including: a data acquisition module 51, a coordinate system transformation module 52, a collision detection module 53, and a decision output module 54, wherein: The acquisition module 51 is used to acquire the real-time spatial pose data of the component to be tested.

[0093] The coordinate system transformation module 52 is used to transform the real-time spatial pose data to a unified reference coordinate system.

[0094] The collision detection module 53 is used to execute the first collision detection algorithm and / or the second collision detection algorithm to determine the preliminary collision detection result and / or the secondary collision detection result.

[0095] The decision output module 54 is used to generate and output collision detection conclusions.

[0096] In some embodiments, the acquisition module 51 may be a multi-source sensor system deployed in a medical device, such as a sensor system arranged on a moving part to acquire real-time spatial pose data of the moving part. For example, the sensor system may include a high-precision encoder, a distance sensor, a position sensor, etc.

[0097] In some embodiments, the acquisition module 51 may be a multi-source sensor system deployed in the environment, such as a sensor system arranged in a scanning room for real-time acquisition of surrounding information. The environmental information includes at least the spatial position information of fixed components, the spatial position information of environmental entities, and the spatial pose information of moving components. For example, the sensor system may include a laser rangefinder, a vision sensor, etc.

[0098] In some embodiments, when performing collision detection on moving parts in a medical device, in order to save costs and reduce the use of computing resources, only a multi-source sensor system deployed in the medical device can be used as the acquisition module 51.

[0099] In some embodiments, since the sensor system arranged in the scanning room for real-time acquisition of surrounding information can acquire spatial position information of medical devices, spatial position information of environmental entities, and spatial pose information of moving parts, a multi-source sensor system deployed in the environment can be used as the acquisition module 51 when performing collision detection between moving parts, between moving parts and fixed parts, and between moving parts and environmental entities.

[0100] In some embodiments, the multi-source sensor system in the environment is less accurate than the multi-source sensor system deployed in the medical device when collecting spatial pose information of moving parts due to issues such as setting distance and acquisition accuracy. Therefore, in some embodiments, corresponding multi-source sensor systems can be deployed in both the environment and the medical device, and then combined to serve as the acquisition module 51.

[0101] In some embodiments, see Figure 8 The coordinate system transformation module 52 includes at least: Reference coordinate system determination unit 521 is used to determine the reference coordinate system in collision detection calculation.

[0102] The coordinate transformation unit 522 is used to receive the real-time spatial pose data of each component to be detected transmitted by the acquisition module 51, and transform the real-time spatial pose data of each component to be detected to a reference coordinate system through corresponding coordinate system transformation calculations, thereby obtaining the reference spatial pose data corresponding to each component to be detected. It is understood that a detailed description of the coordinate system transformation calculations for the components to be detected can be found elsewhere in this document and will not be repeated here.

[0103] Furthermore, the coordinate system transformation module 52 may also include a dynamic calibration unit for periodically calibrating the origin of the reference coordinate system.

[0104] In some embodiments, see Figure 8 The collision detection module 53 includes at least: The fast detection unit 531 is used to execute the first collision detection algorithm to determine the preliminary collision detection result.

[0105] The precise verification unit 532 is used to execute the second collision detection algorithm and determine the secondary collision detection result.

[0106] Specifically, the introductions to the first and second collision detection algorithms can be found elsewhere in this article and will not be repeated here.

[0107] In some embodiments, see Figure 8 The decision output module 54 includes at least: Storage unit 541 is used to receive and / or acquire preliminary collision detection results and secondary collision detection results; Update unit 542 is used to update the preliminary collision detection results by combining the secondary collision detection results, and form a collision detection conclusion; Output unit 543 is used to output the collision detection conclusion.

[0108] Furthermore, in some embodiments, the medical device collision detection system 5 may also include a threshold adjustment module 55, which is used to dynamically adjust a preset collision judgment threshold based on at least one parameter among the medical device operating time, component wear data, and historical collision records.

[0109] For example, during the collision detection process, the threshold adjustment module collects the equipment running time in real time. When the cumulative time exceeds 500 hours, the preset collision judgment threshold is lowered from 10mm to 9mm. The maximum and minimum boundary values ​​of the bed (Ymax=1500mm, Ymin=1200mm) are calculated by the coordinate extreme value comparison method and compared with the probe boundary to complete the rapid boundary screening. The separation axis projection method is performed on potential collision components, and the safety margin parameter S=8mm is output. Since S<9mm, the system judges that there is a collision direction and can trigger a yellow warning.

[0110] For a detailed introduction to the medical device collision detection system 5, please refer to the above description of the medical device collision detection method; it will not be repeated here. Each module in the aforementioned medical device collision detection system 5 can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0111] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores relevant data for medical device collision detection. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a medical device collision detection method.

[0112] Those skilled in the art will understand that Figure 9 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0113] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0114] 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. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method of collision detection for a medical system, the method comprising: The method comprises: acquiring real-time spatial pose data of a component to be detected; converting the real-time spatial pose data of the component to be detected into a reference coordinate system; performing preliminary collision detection on the component to be detected based on a first collision detection algorithm to determine a preliminary collision detection result; performing secondary collision detection based on the preliminary collision detection result and in combination with a second collision detection algorithm to determine a secondary collision detection result; outputting a collision detection conclusion in combination with the preliminary collision detection result and / or the secondary collision detection result.

2. The method of claim 1, wherein: the preliminary collision detection on the component to be detected based on the first collision detection algorithm to determine the preliminary collision detection result comprises determining the preliminary collision detection result through fast boundary screening based on a coordinate extreme value comparison method.

3. The method of claim 1, wherein: the secondary collision detection based on the preliminary collision detection result and in combination with the second collision detection algorithm to determine the secondary collision detection result comprises performing accurate geometric analysis on the component to be detected in the preliminary collision detection result that has a collision risk to determine the secondary collision result by using a separate axis projection method.

4. The method of claim 1, wherein: the component to be detected comprises at least one or more of a moving component, a fixed component, and an environmental entity.

5. The method of claim 2 or 3, wherein: The first collision detection algorithm and / or the second collision detection algorithm further comprises performing full combination collision detection on all components to be detected, wherein the full combination collision detection is a strategy of performing collision detection on all components to be detected in pairs.

6. The method of claim 1, wherein The reference coordinate system is a coordinate system established with a medical device in the medical system as an origin.

7. The method of claim 1, wherein, Further comprising: determining a preset collision judgment threshold value based on a minimum safe spatial distance threshold value between the components to be detected.

8. The method of claim 7, wherein: The preset collision judgment threshold value can be adaptively adjusted based on at least one of a medical device operation time, component wear data, and historical collision records.

9. The method of claim 7, wherein, Further comprising: determining a safety margin parameter between the components to be detected based on the preliminary collision detection result and / or the secondary collision detection result; comparing the safety margin parameter with the preset collision judgment threshold value to determine the collision detection conclusion.

10. The method of claim 1, wherein: The component to be detected comprises a convex polygon component and a non-convex polygon component.

11. The method of claim 10, wherein, Further comprising: performing convexification processing on the non-convex polygon component, wherein the convexification processing comprises a decomposition method and / or a compensation method. The decomposition method comprises dividing the non-convex polygon component into a plurality of convex polygon sub-components. The compensation method comprises constructing a minimum circumscribed convex hull of the non-convex polygon component and compensating the non-convex polygon component into a convex polygon component based on the minimum circumscribed convex hull.

12. The method of claim 11, wherein: when the decomposition method is used, if the number of convex polygon sub-components after division exceeds a preset threshold value, the compensation method is switched to.

13. A medical device collision detection system characterized by Further comprising: a collection module (51) configured to acquire real-time spatial pose data of a component to be detected; a coordinate system conversion module (52) configured to convert the real-time spatial pose data into a unified reference coordinate system; a collision detection module (53) configured to execute the first collision detection algorithm and / or the second collision detection algorithm, and determine the preliminary collision detection result and / or the secondary collision detection result; a decision output module (54) configured to generate and output a collision detection conclusion.

14. The system of claim 13, wherein: the collision detection module (53) comprises: a fast detection unit (531) configured to execute the first collision detection algorithm and determine the preliminary collision detection result; a precise verification unit (532) configured to execute the second collision detection algorithm and determine the secondary collision detection result.

15. The system of claim 13, wherein, further comprising: a threshold adjustment module configured to dynamically adjust the preset collision judgment threshold according to at least one of the following parameters: device running time, component wear data, and historical collision records.

16. The system of claim 13, wherein: the coordinate system conversion module (52) further comprises a dynamic calibration unit configured to periodically verify the position of the origin of the reference coordinate system.

17. A computer readable storage medium characterized by a computer program stored in the computer readable storage medium and executable by a processor to implement the method of any one of claims 1-12.

18. A medical system, characterized by: comprising: a patient support device configured to support a patient; a medical device configured to treat / diagnose the patient; and the computer readable storage medium of claim 17, and / or the collision detection system of any one of claims 13-16. ​

Images