A control method, device, equipment and medium of a C-arm

Abstract

The embodiment of the present specification provides a control method, device, equipment and medium of a C-arm. The embodiment of the present specification determines the target body thickness of a patient, inputs a trained machine learning model, generates a plurality of target shooting positions suitable for the individual patient, and then generates a motion curve in line with the device dynamics constraint based on the target shooting positions, thereby realizing the automatic construction of the motion trajectory of the C-arm, effectively overcoming the limitations of relying on limited preset curves and experience selection, and ensuring the stability of the image quality, the scanning operation efficiency and the reliability of the C-arm control.

Description

Technical Field

[0001] This specification relates to the field of medical imaging technology, and in particular to a control method, device, equipment, and medium for a C-arm. Background Technology

[0002] In angiography, to obtain multi-angle diagnostic images of body parts such as the heart, the C-arm of the imaging device needs to perform complex multi-axis movements around the patient. This multi-axis movement, often referred to as a swing or pendulum motion, allows the C-arm to project and acquire images of the target area from a series of continuously changing body angles, thereby generating multiple images of body parts that can aid in diagnosis.

[0003] To achieve the aforementioned multi-axis motion, imaging equipment manufacturers typically pre-calculate and embed several standard Swing motion curves based on standard phantoms or population average data. In actual clinical practice, when multi-position scanning is required, operators need to select one of these limited preset curves to match the current patient's body shape characteristics. Due to differences in patient body shape, for example, when acquiring multi-angle diagnostic images of the heart, patients exhibit wide individual differences in trunk thickness. Fixed preset curves are difficult to accurately adapt to all situations, especially at certain key angles of the motion trajectory. The relationship between the X-ray path and organ structure may not be optimal, resulting in image quality that fails to meet requirements, such as unclear blood vessels or incomplete diagnostic information, potentially necessitating repeated scanning procedures. Repeated scanning not only prolongs surgical time and increases the workload of medical staff but also means that patients will receive unnecessary additional radiation exposure and contrast agent doses. Summary of the Invention

[0004] To overcome the aforementioned problems in the prior art, this disclosure provides a control method, device, equipment, and medium for a C-arm. Specifically, this disclosure is achieved through the following technical solution: According to a first aspect of the embodiments of this specification, a control method for a C-arm is provided, the C-arm being used to acquire medical images based on X-rays penetrating a target human body region, the control method comprising: The thickness information of the target human body region along the path of the X-ray is determined as the target body thickness; The target body thickness is input into a trained machine learning model to obtain multiple target shooting positions for the target human body region; the trained machine learning model is trained through a sample set set for the target human body region, the sample set including multiple samples; each sample includes thickness information and a target human body region bound to the thickness information; the target human body region has multiple labels, each label is used to identify a shooting position corresponding to the target human body region; The motion curve of the C-arm is generated based on the multiple target shooting positions, and the C-arm is controlled according to the motion curve.

[0005] According to a second aspect of the embodiments of this specification, a control device for a C-arm is provided, the C-arm being used to acquire medical images based on X-rays penetrating a target human body region, the control device comprising: The body thickness determination module is used to determine the thickness information of the target human body region along the path of the X-ray, as the target body thickness; The location acquisition module is used to input the target body thickness into a trained machine learning model to obtain multiple target shooting positions for the target human body region; the trained machine learning model is trained by several sample sets set for the target human body region, and the samples in the sample sets include: several thickness information, and a labeled target human body region bound to each thickness information, the label being multiple shooting positions corresponding to the target human body region; The curve fitting module is used to generate the motion curve of the C-arm based on the multiple target shooting positions, and to control the C-arm according to the motion curve.

[0006] According to a third aspect of the embodiments of this specification, an electronic device is provided, including a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being prompted by the machine-executable instructions to perform the method as described in the first aspect.

[0007] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, wherein the storage medium stores a computer program that, when executed by a processor, implements the method described in the first aspect.

[0008] The embodiments in this specification determine the patient's target body thickness and input it into a trained machine learning model to generate multiple target imaging positions adapted to the individual patient. Then, based on these target imaging positions, motion curves that conform to the device's dynamic constraints are generated, realizing the automated construction of the C-arm motion trajectory. This effectively overcomes the limitations of relying on limited preset curves and experience-based selection, ensuring the stability of image quality, scanning operation efficiency, and the reliability of C-arm control. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of a C-arm as exemplarily shown in the embodiments of this specification; Figure 2 This is a flowchart illustrating a control method for a C-arm as shown in an embodiment of this specification; Figure 3aThis is an exemplary embodiment of the present specification showing the rate of change of the first mapping relationship after parameter adjustment based on the second constraint relationship; Figure 3b This is an exemplary embodiment of the present specification showing the change acceleration curve of the first mapping relationship after parameter adjustment based on the second constraint relationship; Figure 4a This is an exemplary embodiment of the embodiments in this specification, showing the rate of change of the first mapping relationship after smoothing. Figure 4b This is an exemplary embodiment of the present specification showing the change acceleration curve of the first mapping relationship after smoothing; Figure 5 The motion curve of the C-arm is shown as an exemplary embodiment in this specification. Figure 6 This is a schematic diagram of a control device for a C-arm, exemplarily shown in an embodiment of this specification. Figure 7 This is a schematic diagram illustrating the structure of an electronic device as exemplified in an embodiment of this specification. Detailed Implementation

[0010] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this description. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this description as detailed in the appended claims.

[0011] The terminology used in this description is for the purpose of describing particular embodiments only and is not intended to be limiting of this description. The singular forms “a,” “the,” and “the” used in this description and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0012] It should be understood that although the terms first, second, third, etc., may be used in this description to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this description, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0013] In angiography, to obtain multi-angle diagnostic images of body parts such as the heart, the C-arm of the imaging device needs to perform complex multi-axis movements around the patient. This multi-axis movement, often referred to as a swing or pendulum motion, allows the C-arm to project and acquire images of the target area from a series of continuously changing body angles, thereby generating multiple images of body parts that can aid in diagnosis.

[0014] To achieve the aforementioned multi-axis motion, imaging equipment manufacturers typically pre-calculate and embed several standard Swing motion curves based on standard phantoms or population average data. In actual clinical practice, when multi-position scanning is required, the operator needs to select one of these limited preset curves to match the current patient's body shape characteristics. For example, a standard motion curve can be fitted for a patient with a trunk thickness of 23cm in the heart region, and then matched to the same motion curve for patients with trunk thicknesses in the range of [20cm, 25cm] in actual application. It can be seen that for patients with trunk thicknesses other than 23cm, the motion curve is not an precisely fitting curve.

[0015] Therefore, due to variations in patient body shape, such as the wide individual differences in trunk thickness when acquiring multi-angle diagnostic images of the heart, fixed preset curves are difficult to accurately adapt to all situations. Especially at certain key angles of the motion trajectory, the relationship between the X-ray path and organ structure may not be optimal, leading to unsatisfactory image quality, such as unclear blood vessels or incomplete diagnostic information. This may necessitate repeating the scanning process. Repeated scans not only prolong surgical time and increase the workload for medical staff, but also mean that patients will receive unnecessary additional radiation exposure and contrast agent doses.

[0016] Therefore, the current limitation of C-arm control in practice lies in the contradiction between the limited preset curves and the diversity of patient body shapes, making it difficult for existing systems to ensure the accuracy and reliability of image quality in all situations.

[0017] To address the aforementioned problems, embodiments of this specification provide a control method for a C-arm, which is used to obtain medical images based on X-rays penetrating a target human body region.

[0018] like Figure 1 As shown, Figure 1 This is a schematic diagram of a C-arm as illustrated in the embodiments of this specification. The principle of C-arm imaging is that the X-ray beam generated by the X-ray tube 102 penetrates the target human body area to be examined. The X-ray tube 102 has a focal spot, which emits a cone-shaped beam of X-rays after being excited by electricity.

[0019] Detector 101 is located on the opposite side of X-ray tube 102, and together with X-ray tube 102, they constitute the imaging system. When X-rays penetrate the target area of ​​the human body, due to the different attenuation characteristics of different tissues and structures within the human body, the penetrated X-ray beam carries information about the spatial distribution of the internal structures of the human body. Detector 101 receives this penetrated X-ray and converts it into electrical signals. These electrical signals are then processed and reconstructed to ultimately form a medical image of the target human body area.

[0020] Figure 2 This is a flowchart illustrating a control method for a C-arm as shown in an embodiment of this specification. The control method includes: 200: Determine the thickness information of the target human body region along the path of the X-ray, as the target body thickness.

[0021] This step aims to quantify the total radial thickness of the body tissue traversed by the X-ray beam as it penetrates the target human body region, i.e., the target body thickness. It should be noted that the left and right coronary arteries of the heart are the primary target human body regions addressed in the embodiments of this specification. This is because the anatomy of the coronary arteries is particularly complex in angiography applications. To clearly and non-overlappingly visualize each vessel of the coronary artery, a specific projection angle must be found so that the X-ray is precisely tangential to that vessel segment. The ideal projection angle differs significantly between different segments of the left and right coronary arteries. This requirement of "one location, multiple specific angles" necessitates that the C-arm project and acquire data from the target region from a series of continuously changing projection angles.

[0022] When imaging the left or right coronary artery of the heart, target thickness refers to the equivalent thickness of the thoracic tissue that X-rays must penetrate to obtain a clear image at a specific projection angle. It includes not only the thickness of the heart itself but also the cumulative thickness of all the tissues covered by the X-ray. Understandably, the target thickness obtained for an obese patient is greater than that for a thin patient. Accurately obtaining this thickness information is a prerequisite for subsequently obtaining the optimal imaging angle, generating personalized Swing motion curves, and ultimately obtaining high-quality medical images.

[0023] As one or more embodiments of the present invention, determining the thickness information of the target human body region along the path of the X-ray includes: The thickness information of the target human body region along the path of the X-ray is matched with the thickness information of the calibration object, and the thickness information of the calibration object is determined by the current working state of the C-arm.

[0024] Specifically, this embodiment uses the quantitative relationship between the known thickness of a calibration object and the grayscale information of an image to accurately estimate the thickness information of the target human body region. In this embodiment, the calibration object serves as a reference, and its material is typically selected from materials with stable physical properties and known X-ray attenuation characteristics, such as aluminum plates or water models of a specific thickness. By performing imaging analysis on these calibration objects, a correspondence between thickness and image grayscale information can be established.

[0025] In practical applications, establishing a quantitative relationship between a known calibration object and image grayscale values ​​requires first obtaining imaging data from the C-arm under different operating conditions. These operating conditions refer to the combination of control parameters of the C-arm that affect the quality and characteristics of X-ray imaging, specifically including but not limited to: tube voltage, tube current, beam filtering conditions, exposure pulse width, and the distance between the X-ray tube and the detector.

[0026] Under any given combination of control parameters, a definite correspondence exists between the image grayscale information of the calibration object in the imaging system and its actual physical thickness. To establish and apply this definite relationship, this embodiment provides two technical approaches to achieve accurate estimation of the thickness information of the target human body region.

[0027] The first technical approach employs the Monte Carlo simulation method. This is a numerical calculation technique based on random sampling statistics, which simulates the microscopic process of the interaction between X-ray photons and a calibration object through random sampling. By running simulations of a large number of photons, the attenuation of the X-ray beam caused by calibration objects of different thicknesses under different combinations of imaging control parameters can be systematically calculated, thereby predicting the corresponding image grayscale value, i.e., pre-determining the correspondence between grayscale information and the thickness information of the calibration object. When executing step S200, after obtaining the image grayscale value of the target human body region, the thickness information corresponding to the image grayscale value can be directly determined through the pre-determined correspondence under the current combination of control parameters. It should be noted that the thickness information determined by this modeling method is not the actual geometric thickness of the human tissue, but rather an equivalent thickness value of the calibration object. Its physical meaning is: under the current imaging parameters, the attenuation effect of the target human body region on X-rays is equivalent to the attenuation effect caused by a calibration object of a certain thickness.

[0028] The second technical approach employs a specially trained neural network model. This model can be configured to take the operating state (i.e., the combination of control parameters) and the grayscale value of the calibration object as input, and the thickness of the calibration object as output. Trained using a large amount of experimental data, the neural network model learns the complex mapping relationship between multidimensional combinations of control parameters and the thickness of the calibration object to the output thickness. During step S200, the trained neural network model can directly output the corresponding thickness information of the target human body region based on the current combination of control parameters and the grayscale value of the target human body region. Similarly, based on the model's training mechanism, this output value is also an equivalent thickness value of the calibration object.

[0029] Of the two technical approaches mentioned above, Monte Carlo simulation, based on first principles, has a clearer physical meaning; while neural network models possess greater adaptability and computational efficiency. In practical applications, the appropriate technical approach can be selected based on specific needs, or the two methods can be combined to achieve high-precision estimation of the equivalent thickness of the human body.

[0030] Furthermore, the step of determining the thickness information of the target human body region along the path of the X-ray in this embodiment can be integrated into the automatic brightness control process of the C-arm. The purpose of automatic brightness control is to maintain the brightness of the image signal received and converted by the detector within a preset ideal range. Specifically, the automatic brightness control process may include: real-time monitoring of the image brightness signal output by the detector; when the brightness deviates from the target value, the system automatically adjusts the combination of control parameters to correct this brightness deviation. For example, when the detected image brightness is lower than the target value (indicating that the X-rays penetrating the human body are too weak), the tube voltage can be increased to increase the X-ray penetration ability, or the tube current can be increased to enhance the X-ray intensity, thereby bringing the image brightness back to the preset ideal range.

[0031] By integrating the thickness information determination step S200 into the automatic brightness control process, body thickness estimation can be performed "incidentally" during the regular automatic brightness control process. Specifically, the system uses the combination of control parameters determined by the automatic brightness control, as well as the stable image acquired under that combination of control parameters, to calculate the thickness information of the target human body region. This avoids additional scanning or calculations for separate body thickness measurement, optimizes the workflow, and improves overall efficiency.

[0032] As one or more embodiments of the present invention, determining the thickness information of the target human body region along the path of the X-ray includes: The height information of the upper surface of the target human body region is obtained through sensors; Based on the height information of the upper surface and the height information of the lower surface of the target human body region obtained in advance, the thickness information of the target human body region along the path of the X-ray is determined.

[0033] Compared to the aforementioned method of obtaining thickness information through correspondence, this embodiment determines the thickness information of the target human body region along the X-ray path by directly measuring the spatial geometric dimensions.

[0034] This embodiment uses a pre-configured depth sensor to non-contactly acquire the position data, i.e., height information, of the upper surface of the target human body region in three-dimensional space. The depth sensor can be integrated into a C-arm. After acquiring the height information of the upper surface, the thickness can be calculated by combining it with the height information of the lower surface of the target human body region, which is usually directly obtainable. In a typical application scenario, the height information of the lower surface is the height of the support plane of the examination bed that supports the patient. The height of this support plane is known and can be accurately measured and stored in advance during the calibration stage. After simultaneously knowing the heights of the upper and lower surfaces of the target region, the process of determining the thickness information is simply the difference between these two height values ​​in the direction perpendicular to the plane of the examination bed.

[0035] 202: Input the target body thickness into a trained machine learning model to obtain multiple target shooting positions for the target human body region.

[0036] The trained machine learning model is trained using a sample set set for the target human body region. The sample set includes multiple samples. Each sample includes thickness information and a target human body region bound to the thickness information. The target human body region has multiple labels, and each label is used to identify a shooting position corresponding to the target human body region.

[0037] Step 202 takes the previously determined target body thickness as input and passes it to the trained machine learning model. This machine learning model, which can be implemented using regressive neural networks, support vector machines, multilayer perceptrons, etc., can output multiple target imaging positions for the target human body region based on the input target body thickness. These target imaging positions represent the optimal combination of gantry angles recommended for achieving high-quality medical imaging under specific body thickness conditions.

[0038] The training process of machine learning models relies on a sample set constructed for a specific target human body region (such as the left or right coronary artery). This sample set contains multiple samples, each of which includes two parts of data: first, thickness information, which is the actual geometric thickness of human tissue or the thickness value of an equivalent calibration object obtained through measurement or calculation; and second, the target human body region bound to this thickness information, which is labeled with multiple tags, each tag identifying an ideal imaging position corresponding to the target human body region under specific thickness conditions.

[0039] This sample structure design enables the model to learn the complex mapping relationship between thickness parameters and a series of discrete but related imaging positions. In terms of model architecture, this embodiment can train a dedicated prediction model for each target human body region, achieving highly specialized mapping learning; alternatively, a unified comprehensive model can be constructed by using type identifiers of the target human body region (such as identifiers representing the left and right coronary arteries, respectively) as additional input features, enabling the model to distinguish different regions and output corresponding imaging positions. In this way, the machine learning model can not only understand the influence of thickness parameters on imaging positions but also grasp the imaging requirements specific to different anatomical sites, ultimately achieving accurate prediction of personalized target imaging positions from individualized thickness information for each patient.

[0040] Furthermore, when the target human body region is the left coronary artery, the target imaging position is used to characterize the imaging position of the left coronary artery in the head-on, left shoulder, spider, foot, liver, or right shoulder position.

[0041] When the target human body region is the right coronary artery, the target imaging position is used to characterize the imaging position of the right coronary artery in the left lateral, head-on, or right lateral views.

[0042] This embodiment defines multiple designated target imaging positions for the left and right coronary arteries. These target imaging positions are standard projection positions widely recognized and used in clinical coronary angiography. These positions have been validated through long-term practice, enabling clear visualization of the main trunks and major branches (such as the left anterior descending artery and the circumflex artery) of the left and right coronary arteries from different angles, minimizing vessel shortening and overlap. Medical images acquired based on these target imaging positions ensure the completeness of information required for clinical diagnosis and avoid unnecessary additional radiation exposure and contrast agent doses.

[0043] By training the model using these samples, the machine learning algorithm gradually learns the complex nonlinear mapping relationship between thickness information and the set of imaging positions. After training, the model possesses personalized prediction capabilities. Therefore, when performing step 202, only the target body thickness of the current patient needs to be input, and the model can automatically output a set of target imaging positions, thus providing a data foundation for the subsequent generation of accurate motion curves.

[0044] 204: Generate the motion curve of the C-arm based on the multiple target shooting positions, and control the C-arm according to the motion curve.

[0045] Step 204, based on the multiple discrete target imaging positions obtained in the previous step, generates a continuous and smooth C-arm motion trajectory curve through motion planning. This motion curve ensures that the C-arm can accurately pass through all preset key imaging positions in sequence, and that the entire motion process meets the device dynamics requirements and clinical imaging needs.

[0046] Specifically, generating motion curves requires transforming a discrete spatial position sequence into a continuous time-space function. For example, polynomial interpolation, Bézier curves, and B-spline curves can be used to generate motion curves. The process of generating motion curves needs to comprehensively consider the physical constraints of the C-arm's motion, including velocity and acceleration limits for each motion axis, to avoid excessive impact during start-up, shutdown, or turning. Simultaneously, the motion trajectory must remain smooth and continuous to prevent image acquisition quality from being affected by frame vibration or excessive motion amplitude.

[0047] The resulting motion curve contains complete positional information of the C-arm in the time dimension. This curve is sent to the underlying motion control system of the C-arm, driving the X-ray tube and detector to accurately complete the multi-angle scanning task according to the planned path, speed, and acceleration.

[0048] In one or more preferred embodiments of the present invention, the target imaging position includes the left-right tilt angle and the head-to-foot tilt angle of the C-arm. The left-right tilt angle refers to the angle of rotation of the C-arm around the longitudinal axis of the patient's body (head-to-foot direction). It determines whether the X-ray projection direction is from the patient's left side to the right side or from the right side to the left side, used for observing the target structure from different sides. The head-to-foot tilt angle refers to the angle of rotation of the C-arm around the patient's left-right axis. It determines whether the X-ray projection direction is from the patient's head to the feet or from the feet to the head, used for observing the target structure from a vertical direction.

[0049] The target shooting position satisfies the first constraint condition, which is used to constrain the range of values ​​for the left and right tilt angles and the head-to-foot tilt angles.

[0050] To ensure that the generated imaging position meets both clinical diagnostic requirements and equipment and patient safety standards, this embodiment introduces a first constraint to limit the range of values ​​for the left and right tilt angles and the head-foot tilt angle. For example, in the target imaging position for the "spider position" of the left coronary artery, this constraint can set the standard for the left and right tilt angles to "45 degrees left anterior oblique," with a range of variation limited to "±5 degrees," meaning the range of the left and right tilt angles is from 40 degrees to 50 degrees left anterior oblique. Simultaneously, the range of the head-foot tilt angle is limited to 25 degrees to 35 degrees head-lateral. This provides clear boundaries for the output of the machine learning model, making the generated target position itself feasible and optimized.

[0051] As one or more preferred embodiments of the present invention, generating the motion curve of the C-arm based on the plurality of target shooting positions includes: S2040: Based on the left and right tilt angles of the multiple target shooting positions, a first mapping relationship between the left and right tilt angles of the C-arm and the motion time is obtained by fitting.

[0052] S2042: Based on the head-foot tilt angles of the multiple target shooting positions, a second mapping relationship between the head-foot tilt angles of the C-arm and the motion time is obtained by fitting.

[0053] To generate the motion curve of the C-arm, the motion of the two degrees of freedom, left and right tilt angles and head-to-foot tilt angles, can be modeled independently first. Based on the left and right tilt angle data from all target shooting positions, a function describing how the left and right tilt angles change continuously with motion time is constructed using a curve fitting algorithm; this is the first mapping relationship. Similarly, based on all head-to-foot tilt angle data, a function describing how the head-to-foot tilt angles change with time is constructed independently; this is the second mapping relationship.

[0054] S2044: Generate the motion curve of the C-arm based on the preset second constraint, the first mapping relationship and the second mapping relationship; the second constraint is used to constrain the range of values ​​of the change rate and change acceleration of the left and right tilt angles and the head-foot tilt angle.

[0055] The two independent angle-time curves initially fitted cannot guarantee that the corresponding motion process meets the dynamic requirements of the C-arm and the clinical imaging needs. Therefore, a second constraint is introduced to constrain the range of values ​​that the rate of change (first derivative) and the acceleration (second derivative) of the left and right tilt angles and the head-foot tilt angle must adhere to during the motion. Based on these ranges of velocity and acceleration, the first and second mapping relationships are adjusted and optimized.

[0056] Further, generating the motion curve of the C-arm based on the preset second constraint, the first mapping relationship, and the second mapping relationship includes: S20440: Based on the second constraint, adjust the parameters in the first mapping relationship used to characterize the rate of change and acceleration of the left and right tilt angles, and the parameters in the second mapping relationship used to characterize the rate of change and acceleration of the head-foot tilt angles.

[0057] For example, the first mapping relationship obtained by fitting through cubic spline interpolation can be expressed as: , Where i represents the number of the target imaging position, such as the imaging positions of the left coronary artery in the head view, left shoulder view, spider view, foot view, liver view and right shoulder view, which are numbered from 1 to 6 respectively; This refers to the time point when the C-arm is adjusted to the i-th target shooting position. This represents the starting time point of the C-arm's movement, and is arbitrary. and The time intervals between them are the same; t is a time variable, representing the motion time of the C-arm, and the range of values ​​for t is... ; This represents the left and right tilt angles of the C-arm at time point t; Indicates the C-arm at time point The angle values ​​of the left and right tilt angles; ; Indicates the C-arm at time point The acceleration due to the change in the left and right tilt angles; Indicates the C-arm at time point The rate of change of acceleration due to the change in the left and right tilt angles.

[0058] Based on the first mapping relationship obtained through fitting, the rates of change and accelerations of the left and right tilt angles at any given time point can be obtained using the forward difference formulas for velocity and acceleration. Then, according to the value ranges defined in the second constraint conditions for the rates of change and accelerations of the left and right tilt angles, the parameters characterizing the rates of change and accelerations are adjusted so that the first mapping relationship falls completely within the safe range specified by the second constraint conditions at any given time point. In the above example, the parameters used to characterize the rates of change include... , and Parameters used to characterize changing acceleration include and .

[0059] The parameter adjustment method for the second mapping relationship is the same as that for the first mapping relationship, and will not be elaborated here.

[0060] S20442: Smooth the first mapping relationship and the second mapping relationship.

[0061] After the parameter adjustment is completed, the smoothing process of the mapping relationship is then performed. Figure 3a This is an exemplary embodiment of the embodiments in this specification, showing the rate of change of the first mapping relationship after parameter adjustment based on the second constraint relationship. Figure 3b This is an exemplary embodiment of the embodiments in this specification, showing the change acceleration curve of the first mapping relationship after parameter adjustment based on the second constraint relationship. Figure 3a and Figure 3b It can be seen that the forced parameter adjustments in the previous stage, while satisfying the dynamic constraints, may cause discontinuities or abrupt curvature changes in the velocity and acceleration curves, resulting in jitter or impacts in the C-arm's motion. To eliminate these defects, the adjusted first and second mapping relationships need to be smoothed, for example, by using a moving average filter with boundary processing to smooth out high-frequency fluctuations and abrupt changes in the curves.

[0062] Figure 4a This is an exemplary embodiment of the present specification showing the rate of change of the first mapping relationship after smoothing. Figure 4b This is an exemplary embodiment of the embodiments in this specification illustrating the change acceleration curve of the first mapping relationship after smoothing. Based on Figure 4a and Figure 4b It can be seen that after parameter adjustment and smoothing, the first and second mapping relationships show that velocity and acceleration have become continuous without obvious abrupt changes.

[0063] S20444: Generate the motion curve of the C-arm based on the smoothed first and second mapping relationships.

[0064] After parameter adjustment and smoothing, the first and second mapping relationships, although showing continuous velocity and acceleration without significant abrupt changes, revealed during implementation that the relationship between position, velocity, and acceleration data in the smoothed first mapping relationship might no longer strictly adhere to basic kinematic laws. Therefore, based on the smoothed changing acceleration, an integral operation can be performed to deduce a velocity curve matching the acceleration curve. Subsequently, this newly obtained velocity curve is integrated to finally calculate a physically consistent position curve with the velocity and acceleration curves, thus regenerating the first and second mapping relationships that satisfy the second-order kinematic laws. Figure 5This is an example of the motion curve of a C-arm illustrated in the embodiments of this specification, where LAO / RAO represents the left and right tilt angles, and CAU / CRA represents the head-to-foot tilt angles. By fusing the first and second mapping relationships, the complete motion curve of the C-arm can be obtained. At any given moment, this motion curve corresponds to a position uniquely determined by the left and right tilt angles and the head-to-foot tilt angle.

[0065] The motion curves generated using the above methods ensure that the C-arm can accurately reach each target shooting position, and the motion process is smooth, efficient, and in line with the dynamic performance of the C-arm, thereby guaranteeing the quality of image acquisition and the safety of equipment operation.

[0066] The embodiments in this specification determine the patient's target body thickness and input it into a trained machine learning model to generate multiple target imaging positions adapted to the individual patient. Then, based on these target imaging positions, motion curves that conform to the device's dynamic constraints are generated, realizing the automated construction of the C-arm motion trajectory. This effectively overcomes the limitations of relying on limited preset curves and experience-based selection, ensuring the stability of image quality, scanning operation efficiency, and the reliability of C-arm control.

[0067] The embodiments in this specification also provide Figure 6 The exemplary embodiment provided is a control device for a C-arm, the C-arm being used to acquire medical images based on X-rays penetrating a target human body region, the control device comprising: The thickness determination module 600 is used to determine the thickness information of the target human body region along the path of the X-ray, as the target body thickness; The location acquisition module 602 is used to input the target body thickness into a trained machine learning model to obtain multiple target shooting positions for the target human body region; the trained machine learning model is trained by several sample sets set for the target human body region, and the samples in the sample sets include: several thickness information, and a labeled target human body region bound to each thickness information, the label being multiple shooting positions corresponding to the target human body region. The curve fitting module 604 is used to generate the motion curve of the C-arm based on the multiple target shooting positions, and to control the C-arm according to the motion curve.

[0068] The embodiments in this specification also provide Figure 7 The diagram illustrates the structure of an exemplary electronic device. The electronic device includes a processor 701 and a machine-readable storage medium 702; the machine-readable storage medium 702 stores machine-executable instructions that can be executed by the processor 701, which in turn cause the processor 701 to perform the method shown in any of the above embodiments.

[0069] like Figure 7 As shown, at the hardware level, the electronic device includes a processor 701, a system bus 703, a network interface, memory, and a machine-readable storage medium 702, and may also include other hardware required for business operations. The processor 701 reads the corresponding computer program from non-volatile memory into memory and then runs it to implement the method shown in any of the above embodiments. Of course, in addition to the software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0070] This specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method shown in any of the above embodiments.

[0071] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0072] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0073] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this illustrative solution according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0074] The embodiments of the subject matter and functional operation described in this specification can be implemented in the following ways: digital electronic circuits, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing apparatus or for controlling the operation of a data processing apparatus. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or combinations thereof.

[0075] The processing and logic described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output.

[0076] Suitable computers for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0077] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0078] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily intended to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation thereof.

[0079] Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0080] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0081] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A control method for a C-arm, characterized in that, The C-arm is used to obtain medical images based on X-rays penetrating a target human body region, and the control method includes: The thickness information of the target human body region along the path of the X-ray is determined as the target body thickness; The target body thickness is input into a trained machine learning model to obtain multiple target shooting positions for the target human body region. The trained machine learning model is trained using a sample set set for the target human body region, which includes multiple samples. Each sample includes thickness information and a target human body region bound to the thickness information. The target human body region has multiple labels, and each label is used to identify a shooting position corresponding to the target human body region. The target shooting position includes the left and right tilt angles and head-to-foot tilt angles of the C-arm. The motion curve of the C-arm is generated based on the multiple target shooting positions, and the C-arm is controlled to pass through the multiple target shooting positions in sequence according to the motion curve.

2. The control method according to claim 1, characterized in that, Determining the thickness information of the target human body region along the path of the X-ray includes: The thickness information of the target human body region along the path of the X-ray is matched with the thickness information of the calibration object, which is determined by the current working state of the C-arm.

3. The control method according to claim 1, characterized in that, Determining the thickness information of the target human body region along the path of the X-ray includes: The height information of the upper surface of the target human body region is obtained through sensors; Based on the height information of the upper surface and the height information of the lower surface of the target human body region obtained in advance, the thickness information of the target human body region along the path of the X-ray is determined.

4. The control method according to claim 1, characterized in that, The target shooting position satisfies the first constraint condition, which is used to constrain the range of values ​​for the left and right tilt angles and the head-to-foot tilt angles.

5. The control method according to claim 4, characterized in that, The generation of the motion curve of the C-arm based on the multiple target shooting positions includes: Based on the left and right tilt angles of the multiple target shooting positions, a first mapping relationship between the left and right tilt angles of the C-arm and the motion time is obtained by fitting. Based on the head-foot tilt angles of the multiple target shooting positions, a second mapping relationship between the head-foot tilt angles of the C-arm and the motion time is obtained by fitting. The motion curve of the C-arm is generated based on the preset second constraint, the first mapping relationship, and the second mapping relationship; the second constraint is used to constrain the range of values ​​for the change rate and change acceleration of the left and right tilt angles and the head-foot tilt angle.

6. The control method according to claim 5, characterized in that, The process of generating the motion curve of the C-arm based on the preset second constraint, the first mapping relationship, and the second mapping relationship includes: Based on the second constraint, adjust the parameters in the first mapping relationship that characterize the rate of change and acceleration of the left and right tilt angles, and the parameters in the second mapping relationship that characterize the rate of change and acceleration of the head-foot tilt angles; Smooth the first mapping relationship and the second mapping relationship; The motion curve of the C-arm is generated based on the smoothed first and second mapping relationships.

7. The control method according to claim 1, characterized in that, When the target human body region is the left coronary artery, the target imaging position is used to characterize the imaging position of the left coronary artery in the head-on, left shoulder, spider, foot, liver, or right shoulder position. When the target human body region is the right coronary artery, the target imaging position is used to characterize the imaging position of the right coronary artery in the left lateral, head-on, or right lateral views.

8. A control device for a C-arm, characterized in that, The C-arm is used to obtain medical images based on X-rays penetrating the target human body region, and the control device includes: The body thickness determination module is used to determine the thickness information of the target human body region along the path of the X-ray, as the target body thickness; The location acquisition module is used to input the target body thickness into a trained machine learning model to obtain multiple target shooting positions for the target human body region. The trained machine learning model is trained through several sample sets set for the target human body region. The samples in the sample sets include: several thickness information and a labeled target human body region bound to each thickness information. The label is multiple shooting positions corresponding to the target human body region. The target shooting positions include the left and right tilt angles and head-to-foot tilt angles of the C-arm. The curve fitting module is used to generate the motion curve of the C-arm based on the multiple target shooting positions, and control the C-arm to pass through the multiple target shooting positions sequentially according to the motion curve.

9. An electronic device, characterized in that, The device includes a processor and a machine-readable storage medium storing machine-executable instructions that can be executed by the processor, the processor being prompted by the machine-executable instructions to perform the control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the control method according to any one of claims 1 to 7.

Images