Method and device for controlling tilt of medical equipment rack, computer device and CT system

Abstract

The application relates to a control method and device for the inclination of a medical equipment gantry, computer equipment and a CT system. The method comprises the following steps: acquiring the actual inclination angle of the gantry of the medical equipment at the current time, and calculating the error angle of the gantry according to the target inclination angle and the actual inclination angle of the gantry; updating the target inclination angular velocity of the position loop control of the controller according to the error angle; obtaining the inclination angular velocity error according to the updated target inclination angular velocity and the actual inclination angular velocity of the gantry; obtaining the pulse width modulation duty cycle of the motor according to the inclination angular velocity error; updating the speed loop control parameter of the controller, obtaining the updated pulse width modulation duty cycle according to the pulse width modulation duty cycle and the updated speed loop control parameter of the controller; and adjusting the inclination angle of the gantry according to the updated pulse width modulation duty cycle. The method can improve the control precision and stability of the inclination movement of the gantry.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a method, apparatus, computer equipment, and CT system for controlling the tilting of a medical equipment gantry. Background Technology

[0002] Medical equipment such as CT and PET-CT typically employs traditional angle measurement devices based on photoelectric encoders in its gantry tilt control system. For example, CT gantry tilt control is mostly based on segmented duty cycle or fixed duty cycle control methods using photoelectric encoders. The segmented duty cycle tilt control method controls the gantry with a preset duty cycle when it tilts to a fixed angle position, while the fixed duty cycle tilt control method drives the gantry to tilt with a fixed duty cycle in all postures.

[0003] While existing technologies include methods to adjust the torque of the DC motor built into the CT gantry by adjusting the PWM duty cycle, the gantry tilting motion is a time-varying nonlinear system. When the tilt angle changes, the system is subjected to gravitational and inertial disturbances. Traditional controllers with segmented duty cycles and pre-set parameters cannot guarantee precise control of the tilt angle and speed. Traditional methods also produce significant jitter when controlling the gantry to stop tilting, resulting in poor accuracy and timeliness of gantry tilt adjustments. Summary of the Invention

[0004] Based on this, in order to solve the technical problem of poor accuracy and timeliness of gantry tilt adjustment and to achieve the purpose of precise control of gantry tilt speed and tilt angle, this application provides a method, device, computer equipment and CT system for controlling the tilt of medical equipment gantry.

[0005] The method for controlling the tilt of a medical equipment rack includes: acquiring the actual tilt angle of the medical equipment rack at the current moment, and calculating the rack error angle based on the target tilt angle and the actual tilt angle; updating the target tilt angular velocity controlled by the position loop of the controller based on the error angle; obtaining the tilt angular velocity error based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack; obtaining the pulse width modulation duty cycle of the motor based on the tilt angular velocity error; updating the speed loop control parameters of the controller, and obtaining the updated pulse width modulation duty cycle based on the pulse width modulation duty cycle and the updated speed loop control parameters of the controller; and adjusting the tilt angle of the rack based on the updated pulse width modulation duty cycle.

[0006] Preferably, the speed loop control parameters of the controller are updated, and the updated pulse width modulation duty cycle is obtained based on the duty cycle and the updated speed loop control parameters of the controller; this includes: obtaining the speed loop control parameters of the controller, the parameters including at least one of proportional, integral, and derivative; obtaining the predicted tilt angular velocity error based on the speed loop control parameters of the controller and the tilt motion prediction model; and inputting the predicted tilt angular velocity error into a BP neural network to obtain the updated pulse width modulation duty cycle.

[0007] Furthermore, based on the speed loop control parameters of the controller and the tilt motion prediction model, the predicted tilt angular velocity error is obtained, including: acquiring the current state parameters of the rack; inputting the current state parameters into the tilt motion prediction model to obtain the tilt prediction angle of the rack at the next moment; and calculating the predicted tilt angular velocity error based on the tilt prediction angle and the tilt target angle at the next moment.

[0008] Furthermore, the training process of the tilt motion prediction model includes: constructing a state parameter dataset of the rack; the state parameters include at least one of tilt angle, target angle, starting angle, stopping angle, timestamp, duty cycle, motion direction, and stop motion signal; building a neural network model of rack tilt; and using the state parameter dataset to train the neural network model to obtain the tilt motion prediction model; and validating the tilt motion prediction model.

[0009] Furthermore, the tilt motion prediction model is trained to obtain the tilt motion prediction model, including: calculating the physical information loss of the tilt motion model based on the physical information loss function; calculating the data information loss of the tilt motion model based on the data information loss function; obtaining the total loss of the tilt motion model based on the physical information loss and the data information loss; and training the rack tilt motion model based on the total loss.

[0010] Furthermore, updating the target tilt angular velocity of the controller's position loop control based on the error angle includes: inputting the error angle into the controller's position loop control to obtain a preliminary angular velocity control quantity; comparing the preliminary angular velocity control quantity with a preset tilt angular velocity threshold; if the preliminary angular velocity control quantity is less than the preset threshold, then the preliminary angular velocity control quantity is the target tilt angular velocity; if the preliminary angular velocity control quantity is greater than the preset tilt angular velocity threshold, then the preset tilt angular velocity threshold is used as the target tilt angular velocity.

[0011] Furthermore, the method also includes: obtaining the adjusted tilt angle and the target position of the frame, calculating the difference between the adjusted tilt angle and the target tilt angle; if the difference is less than or equal to a preset angle threshold, the frame stops tilting; if the difference is greater than the preset angle threshold, the method continues to return to the step of updating the target tilt angular velocity controlled by the position loop of the controller based on the error angle.

[0012] Furthermore, the control device for tilting the medical equipment rack includes: an acquisition module for acquiring the actual tilt angle of the medical equipment rack at the current moment, and calculating the error angle of the rack based on the target tilt angle and the actual tilt angle; a first update module for updating the target tilt angular velocity controlled by the position loop of the controller based on the error angle; a calculation module for obtaining the tilt angular velocity error based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack; an adjustment module for obtaining the pulse width modulation duty cycle of the motor based on the tilt angular velocity error; a second update module for updating the speed loop control parameters of the controller, and obtaining the updated pulse width modulation duty cycle based on the duty cycle and the updated speed loop control parameters of the controller; and an adjustment module for adjusting the tilt angle of the rack based on the updated pulse width modulation duty cycle.

[0013] Furthermore, the computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the aforementioned method. In one embodiment, it further includes a CT system, characterized in that the CT system comprises:

[0014] The system comprises: a gantry; an X-ray tube, mounted on the gantry for generating X-rays; a detector, mounted on the gantry for receiving X-rays passing through the object being detected; the X-ray tube and the detector being drivable to rotate circumferentially along the gantry; a motor for driving the gantry to tilt forward or backward from its own axis of rotation; an inclinometer, mounted on the gantry for determining the gantry tilt angle; and a controller, including a position loop control and a speed loop control. In the position loop control, the controller acquires the actual tilt angle of the gantry at the current moment and calculates the error angle of the gantry based on the target tilt angle and the actual tilt angle, and determines the target tilt angular velocity based on the error angle. In the speed loop control, the controller acquires the actual tilt angular velocity of the gantry and obtains the tilt angular velocity error based on the target tilt angular velocity and the actual tilt angular velocity of the gantry.

[0015] The pulse width modulation duty cycle of the motor is obtained based on the tilt angular velocity error.

[0016] The motor is controlled to drive the frame's movement based on the pulse width modulation duty cycle of the motor.

[0017] The beneficial effects of this invention are as follows: by acquiring the actual tilt angle of the medical device's gantry at the current moment, and calculating the error angle of the gantry based on the target tilt angle and the actual tilt angle; updating the target tilt angular velocity of the controller's position loop control based on the error angle; obtaining the tilt angular velocity error based on the updated target tilt angular velocity and the actual tilt angular velocity of the gantry; obtaining the pulse width modulation duty cycle of the motor based on the tilt angular velocity error; updating the controller's speed loop control parameters, obtaining the updated pulse width modulation duty cycle based on the duty cycle and the updated controller's speed loop control parameters; and adjusting the gantry tilt angle based on the updated pulse width modulation duty cycle; thereby enabling advance prediction of the tilt angle of the gantry response under the current controller output and tilt state, allowing the controller to adapt to changes in the gantry tilt state in advance, thus achieving precise control of the gantry tilt speed and tilt angle, and improving the control accuracy and stability of the gantry tilt movement. Attached Figure Description

[0018] Figure 1 This is an application environment diagram of a method for controlling the tilting of a medical equipment rack in one embodiment;

[0019] Figure 2 This is a flowchart illustrating a method for controlling the tilting of a medical device rack in one embodiment;

[0020] Figure 3 This is a flowchart illustrating a method for controlling the tilting of a medical device rack in another embodiment;

[0021] Figure 4 This is a flowchart illustrating a method for controlling the tilting of a medical device rack in another embodiment;

[0022] Figure 5 Here is a structural diagram of the neural network model in another embodiment;

[0023] Figure 6A This is a flowchart illustrating a method for controlling the tilting of a medical device rack in another embodiment;

[0024] Figure 6B This is a schematic diagram of the workflow for controlling the tilting of a medical device rack in another embodiment;

[0025] Figure 7 This is a structural diagram of the control device for tilting the medical device rack in another embodiment;

[0026] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] The medical equipment in this case includes, but is not limited to, computed tomography (CT) and positron emission tomography-CT (PET-CT).

[0029] Taking CT equipment as an example, a CT scanner mainly consists of three core parts: the scanning section, the computer system, and the image display and storage system. The scanning section mainly includes the X-ray tube, detectors, and gantry. As an important function of medical equipment, the gantry effectively solves the problem of patients needing to assume special positions during imaging examinations of certain organs. During the imaging process, the gantry will tilt at a certain angle in specific scenarios to achieve better imaging.

[0030] The medical equipment rack tilt control method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, medical device 102 communicates with server 104 via a network. Medical device 102 can be, but is not limited to, various computed tomography (CT) systems, positron emission tomography-X-ray computed tomography (PET-CT) systems, etc., and server 104 can be a standalone server or a server cluster consisting of multiple servers.

[0031] like Figure 2 As shown, a method for controlling the tilt of a medical device rack is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0032] S201, obtain the current actual tilt angle of the medical equipment rack, and calculate the rack error angle based on the target tilt angle and the actual tilt angle.

[0033] The actual tilt angle is the current tilt angle of the rack, which can be obtained in real time by an angle sensor mounted on the rack. This angle sensor can be an inclinometer or a gyroscope, etc. The angle sensor can be mounted on a fixed fixture on the left side of the rack, with the inclinometer mounting surface parallel to the surface of the object being measured. Additionally, the sensing line should be parallel to the axis of the surface being measured, i.e., the inclinometer should be mounted parallel to the direction where the rack tilt angle is 0°. The rack tilt status data is then sent to a fixed address in the RAM (Random Access Memory) of the SIB (Stator Interface Board) control board. The GSW reads the specified DMA (Direct Memory Access) address of the inclinometer status data from the SIB control board via the PCIE interface on the SIB, retrieving 64 bits of valid data to obtain the actual tilt angle of the rack. In addition, sensor status and communication status can also be obtained.

[0034] For the inclinometer data acquisition system, the specific implementation is to define the inclinometer status register 0xD60 and the CT gantry tilt control register 0xD98 in the SIB Firmware. The SIB control board actively reads the current status data of the inclinometer every 30ms and refreshes the RAM area designated at address 0xD60 of the inclinometer status data in the SIB. The GSW reads the data in the SIB RAM area corresponding to address 0xD60 through the PCIE interface to obtain the current sampling angle, status and other data of the inclinometer, and controls the tilt of the gantry by writing parameters such as the target angle and DC motor duty cycle to address 0xD98.

[0035] The target tilt angle is the preset tilt angle of the rack in the control system of the medical equipment. The target tilt angle of the rack can be set within a certain tilt angle range as needed. For example, the tilt angle range can be -30 degrees to 30 degrees, or it can be set to 35 degrees, 20 degrees, etc.

[0036] Once the actual tilt angle of the medical equipment rack at the current moment and the preset target tilt angle are obtained, the error angle can be calculated based on these two values. Typically, the difference between the two is calculated to obtain the error angle. For example, if the target tilt angle is 30 degrees and the current actual tilt angle is 25 degrees, then the error angle is 5 degrees.

[0037] S202, based on the error angle, update the target tilt angular velocity controlled by the position loop of the controller.

[0038] The position loop control of a controller uses the error angle as input and generates the target tilt angular velocity through proportional, integral, or composite control algorithms in the outer loop. Generally, a PID control algorithm is used; a PID (Proportional-Integral-Derivative) controller optimizes system performance by adjusting control parameters. Adaptive PID control algorithms combine adaptive control concepts with conventional PID controllers, forming adaptive PID control or self-tuning PID control technology. PID controller basics: Proportional (P): Adjusts the control output based on the current error magnitude. Integral (I): Adjusts the control output based on the accumulated error to eliminate steady-state error. Derivative (D): Adjusts the control output based on the rate of error change to predict future error changes.

[0039] The target tilt angular velocity is the tilt angular velocity preset by the frame's control system.

[0040] Specifically, updating the target tilt angular velocity of the controller's position loop control based on the error angle includes: inputting the error angle into the controller's position loop control to obtain a preliminary angular velocity control value; comparing the preliminary angular velocity control value with a preset tilt angular velocity threshold; if the preliminary angular velocity control value is less than the preset threshold, then the preliminary angular velocity control value is the target tilt angular velocity; if the preliminary angular velocity control value is greater than the preset tilt angular velocity threshold, then the preset tilt angular velocity threshold is used as the target tilt angular velocity. For example, if the calculated preliminary angular velocity control value is 5 degrees / second, and the preset tilt angular velocity threshold is 3 degrees / second, then the target tilt angular velocity is taken as 3 degrees / second. This design can avoid instability or overshoot caused by excessively fast rack movement.

[0041] S203, based on the updated target tilt angular velocity and the actual tilt angular velocity of the frame, the tilt angular velocity error is obtained;

[0042] The actual tilt angular velocity of the rack is calculated by the angle difference and time difference between at least two samplings. That is, the angle sensor obtains the angle at least twice, the angle difference between the two samplings is calculated, and then the angle difference is divided by the time difference to obtain the actual tilt angular velocity.

[0043] Once the actual tilt angular velocity of the medical equipment rack at the current moment and the preset target tilt angular velocity are obtained, the tilt angular velocity error can be calculated based on these two values. Typically, the difference between the two values ​​is calculated to obtain the tilt angular velocity error. For example, if the actual tilt angular velocity ω_actual is 4 degrees / second and the updated target tilt angular velocity ω_target is 3 degrees / second, the tilt angular velocity error Δω = ω_target - ω_actual is calculated, resulting in a tilt angular velocity error of 1 degree / second.

[0044] S204, based on the tilt angular velocity error, obtain the pulse width modulation duty cycle of the motor.

[0045] The speed loop control of the controller takes the tilt angular velocity error Δω between the actual tilt angular velocity ω_actual and the target tilt angular velocity ω_target as input, and generates a pulse width modulation duty cycle through a PID algorithm or adaptive control. The pulse width modulation duty cycle is the duty cycle value of the PWM (pulse width modulation) signal output by the speed loop control of the controller, which is used to drive the motor to adjust the tilt angle of the frame.

[0046] S205, update the speed loop control parameters of the controller, and obtain the updated pulse width modulation duty cycle based on the pulse width modulation duty cycle and the updated speed loop control parameters of the controller;

[0047] The speed loop control parameters of the controller mainly include proportional (P), integral (I), and derivative (D). In this embodiment, the speed loop control parameters of the controller need to be updated to obtain the updated speed loop control parameters. There are several ways to update the speed loop control parameters of the controller, including: anti-integral saturation: limiting the accumulation of the integral term when Δω is continuously less than a threshold; disturbance observation compensation: estimating external disturbances through a load torque observer and correcting the duty cycle output; adaptive gain adjustment: adaptively adjusting the proportional coefficient Kp_v according to the error change rate. This scheme uses adaptive adjustment of the proportional coefficient Kp_v based on the error change rate. Alternatively, the parameters can also be updated using deep learning.

[0048] S206, adjusts the rack tilt angle according to the updated pulse width modulation duty cycle.

[0049] After obtaining the updated duty cycle, the updated pulse width modulation duty cycle is converted into a PWM signal to drive a DC motor or stepper motor, which in turn drives the frame to rotate through a reduction mechanism, thereby adjusting the tilt angle of the frame.

[0050] Further, the adjusted tilt angle and the target position of the frame are obtained, and the difference between the adjusted tilt angle and the target tilt angle is calculated. If the difference is less than or equal to a preset angle threshold, the frame decelerates, and the adjustment is considered complete. If the difference is greater than the preset angle threshold, the process returns to the step of updating the target tilt angular velocity controlled by the controller's position loop based on the error angle. This iterative adjustment is repeated until the difference between the adjusted tilt angle and the target tilt angle is less than or equal to the preset angle threshold. For example, when the angle error Δθ is less than a set angle threshold (e.g., 0.1°) and remains stable for a duration (e.g., 3 seconds), the adjustment is considered complete.

[0051] In the aforementioned method for controlling the tilt of a medical equipment rack, the actual tilt angle of the rack at the current moment is obtained, and the error angle of the rack is calculated based on the target tilt angle and the actual tilt angle. Based on the error angle, the output target tilt angular velocity of the controller's position loop is updated. The tilt angular velocity error is obtained based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack. Based on the tilt angular velocity error, the pulse width modulation (PWM) duty cycle is obtained through the controller's speed loop control. The controller's speed loop control parameters are updated, and the updated PWM duty cycle is obtained based on the duty cycle and the updated controller speed loop control parameters. The tilt angular velocity of the rack is adjusted based on the updated PWM duty cycle, thereby controlling the tilt angle of the rack per unit time. Through the dual-loop controller setup, the rack can adaptively adjust the duty cycle in real time according to the angular velocity error, thus precisely controlling the tilt angle of the rack.

[0052] In one embodiment, such as Figure 3 As shown, the speed loop control parameters of the controller are updated. Based on the duty cycle and the updated speed loop control parameters of the controller, the updated pulse width modulation duty cycle is obtained, including:

[0053] Step 302: Obtain the parameters of the speed loop control of the controller, including at least one of the proportional, integral, and derivative coefficients.

[0054] As mentioned above, the parameters of the controller's speed loop control include the proportional coefficient Kp, integral coefficient Ki, derivative coefficient Kd, and historical control data (such as the duty cycle output values ​​of the past 5 control cycles). Parameter acquisition methods include directly reading preset initial parameters from the controller's memory; and updating parameter values ​​in real time through online identification algorithms (such as recursive least squares method) to cope with the time-varying characteristics of the system.

[0055] Step 304: Based on the parameters of the speed loop control of the controller and the tilt motion prediction model, the predicted tilt angular velocity error is obtained.

[0056] The tilt motion prediction model is a neural network model that combines physical information loss and mathematical information loss to predict the trend of tilt angular velocity error changes over a future time period.

[0057] Specifically, the current state parameters of the rack are obtained; the current state parameters are input into the tilt motion prediction model to obtain the tilt prediction angle of the rack at the next moment; based on the tilt prediction angle and the tilt target angle at the next moment, the predicted tilt angular velocity error is calculated.

[0058] Step 306: Input the predicted tilt angular velocity error into the BP neural network to obtain the updated pulse width modulation duty cycle.

[0059] A backpropagation neural network (BP neural network) is a multi-layered feedforward neural network consisting of an input layer, hidden layers, and an output layer. It adjusts weights using a backpropagation algorithm to approximate nonlinear functions. Its core characteristics are forward propagation of the signal and backward propagation of the error. It minimizes the error between the actual output and the desired output by adjusting the weights and thresholds in the network. In this scheme, its role is to dynamically optimize the pulse width modulation duty cycle based on the prediction error.

[0060] In this embodiment, by introducing the parameters of the speed loop control of the controller and the tilt motion prediction model, and combining them with a BP neural network to output the updated pulse width modulation duty cycle, the high-precision prediction performance of the neural network motion model and the stability of the dual closed-loop control of the controller are appropriately integrated, thereby improving the accuracy of the rack adaptively adjusting the tilt angle.

[0061] In one embodiment, such as Figure 4 As shown, the implementation steps of the tilt motion prediction model include:

[0062] Step S401: Construct a rack state parameter dataset; the state parameters include at least one of tilt angle, target angle, starting angle, stopping angle, timestamp, duty cycle, direction of motion, and stop motion signal.

[0063] State parameters are parameters characterizing the rack state, including at least one of the following: tilt angle, target angle, starting angle, stopping angle, timestamp, duty cycle, direction of motion, and stop signal. The essence of constructing the dataset is to perform time-series data prediction. The input features are the state parameters from the previous N time steps, and the target output is the rack tilt angle at the next sampling time step.

[0064] Specifically, for example, the duty cycle can be incremented in 1% increments, from 0% to 100%. The initial and stopping angles during the tilting process are freely combined within the range [-30°, 30°] in 1° increments. By iterating through different combinations of starting angles, stopping angles, and duty cycles, this combination of state parameters is sent to the SIB (Stator Interface Board) via the rack software. The rack tilting is then performed, and tilting motion data is collected as the training set. Using random values ​​with a duty cycle updated every second from 0% to 100% as input, rack tilting motion data from -30° to 30° and from 30° to -30° are used as the test set.

[0065] Step S402: Build a neural network model for rack tilting; and train the neural network model using the state parameter dataset to obtain a tilting motion prediction model.

[0066] Neural network models for rack tilting can be used to learn mapping representations that conform to the laws of motion in the real physical world. They can extract deep features of rack motion by performing layer-by-layer linear and nonlinear transformations on simple representational feature data. This allows the model to analyze input features and fit motion patterns that conform to physical laws, thereby building a tilting motion prediction model. Types of neural network models that can be selected include: Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Generative Adversarial Networks (GAN), and Transformer Neural Networks.

[0067] Generally, internal factors influencing the motion of a rack tilt include the rack's center of gravity, the effective force of the tilting push rod, and the rack's mass. External factors include tilting speed, the response speed of the DC motor driver's duty cycle, and the input duty cycle. Fitting the motion of a rack tilt using conventional mathematical models (such as ARMA and linear regression) based solely on its characteristic features requires an enormous dataset, incurring significant human and material costs. Furthermore, these methods are not effective in exploring the physical laws governing rack motion. The neural network model in this embodiment is based on prior physical knowledge, thus constituting a physical neural network model. Compared to a typical convolutional neural network, it adds a physical information loss function. During training, it not only relies on data but also incorporates prior knowledge of physical laws. It can be used to learn a mapping representation that conforms to the motion laws of the real physical world and can extract deep features of rack motion by performing layer-by-layer linear and nonlinear transformations on simple characteristic data. This facilitates the model's analysis of input features and fitting motion laws that conform to physical laws, thereby improving the prediction accuracy of the motion model.

[0068] Physical neural network model structure as follows Figure 5 As shown. Figure 5 The diagram shows the structure of the neural network model. The left side represents the input feature set, which is the set of state parameters of the rack. These parameters are input into the neurons of the neural network model, resulting in the output of the predicted tilt angular velocity. The middle layer consists of hidden layer neurons.

[0069] In one embodiment, training a tilt motion prediction model to obtain a tilt motion prediction model specifically includes: calculating the physical information loss of the tilt motion model based on a physical information loss function; calculating the data information loss of the tilt motion model based on a data information loss function; obtaining the total loss of the tilt motion model based on the physical information loss and the data information loss; and training the rack tilt motion model based on the total loss.

[0070] During the training process, the tilt motion prediction model incorporates data information loss functions and physical information loss functions as constraints. Physical information loss functions include, but are not limited to, the kinematic equations of a rigid body rotating about a fixed axis, the rotational form of Newton's second law, and the torque calculation formula for a DC motor. Data information loss functions include, but are not limited to, mean squared error (MSE), mean absolute error (MAE), and the Huber loss function. For example, MSE, the dynamic equations of a rigid body rotating about an axis, and the Adam optimizer can be selected as the data information loss function, physical information loss function, and backpropagation optimizer for model training, respectively. A Bayesian optimization algorithm can then be used to search for the optimal training parameters to improve the model's generalization ability.

[0071] Specifically, firstly, by minimizing the mean squared error between the predicted and actual observations, the model learns the ability to represent input features into output predicted angles, and by minimizing the loss of physical information, the model learns the inherent physical laws during tilting motion. Secondly, the Adam optimizer is used to smooth the gradients during backpropagation, preventing excessive fluctuations in the neural network's parameter updates and facilitating the convergence of the loss function. Thirdly, a Bayesian optimization algorithm is used to search the hyperparameter space to find the optimal combination of network parameters and learning rate hyperparameters. Finally, the optimal hyperparameters are used to train the physical neural network model.

[0072] In this embodiment, the physical information loss function includes constructing a kinematic theory of the physical world for updating neural network parameters by combining the kinematic equations of rigid body rotation about a fixed axis, Newton's second law, and DC motor torque calculation formula, so that the predicted values ​​satisfy basic physical laws.

[0073] Combining the kinematic equations of a rigid body rotating about a fixed axis, specifically, the kinematic equations of a frame rotating about a fixed axis are expressed by formulas (1) and (2).

[0074] ω(t)=dθ(t) / dt (1)

[0075]

[0076] Where θ(t) represents the angle of rotation of the CT gantry around a fixed axis, in radians. ω(t) represents the rate of change of angular displacement with time (angular velocity), in radians per second. α(t) represents the rate of change of angular velocity with time (angular acceleration).

[0077] The formula for calculating Newton's second law in rotational motion is:

[0078] τ=I×α (3)

[0079] Where τ is the torque that pushes the CT gantry to tilt. I represents the moment of inertia of the CT gantry during its tilting motion around its axis, and α represents the angular acceleration. The moment of inertia of the CT gantry can be calculated using the following formula.

[0080] I=∫r 2 dm (4)

[0081] Where I is the moment of inertia of the rigid body about a fixed axis, and d m Let r be the mass distribution and r be the perpendicular distance from the mass point to the axis of rotation. Using the finite element method, the CT gantry is meshed. The mass distribution dm of each element is calculated based on the object density and unit volume. For objects with uneven density distribution, the mass of the region is roughly estimated using local material properties. The moment of inertia I is obtained by accumulating all elements. This parameter is a constant.

[0082] The output torque of a DC motor is determined by the DC motor input voltage V. avg The value is determined by the motor winding R and calculated using formulas (5), (6), and (7).

[0083] V avg =D×V supply (5)

[0084]

[0085] Where D represents the voltage duty cycle output by the CT gantry control software, V supply Kt represents the power supply voltage of the CT gantry, and I represents the torque constant of the DC motor controlling the tilt of the gantry. A R represents the drive current of the DC motor, R represents the resistance value of the DC motor winding, and T represents the torque of the DC motor at the duty cycle D.

[0086] For the physical information loss of the gantry tilting motion model, after ignoring the loss of the DC motor torque transmission device, the torque T of the DC motor under the duty cycle D can be approximately equal to the torque that pushes the CT gantry to tilt. Therefore, the physical information loss function of the neural network can be defined based on Newton's second law, as shown in formula (8).

[0087]

[0088] in, Let N represent the physical information loss, i represent the number of samples, and θ represent the i-th sample. pred (t i ) represents t i The tilt angle at time τ(t) i ) represents t i The torque at time t is calculated using a formula. I represents the moment of inertia of the CT gantry, a constant estimated through finite element analysis.

[0089] Data loss functions include mean squared error (MSE), mean absolute error (MAE), Huber loss function, etc. For example, the mean squared error can be obtained by calculating the model prediction value and the actual observation value. The calculation formula (9) is as follows:

[0090]

[0091] Where, θ true (t i ) represents t i The observed tilt angle of the gantry, sampled by the tilt sensor at any given moment. This indicates the loss of mathematical information.

[0092] Based on the physical information loss and the data information loss, the total loss of the tilt motion model is obtained; specifically, the total loss function used to train the tilt motion model of the rack can be obtained by weighted summation of the data information loss and the physical information loss, as shown in formula (10).

[0093]

[0094] Here, λ is the weight, used to adjust the relative importance of data information loss and physical information loss. λ can be obtained empirically, enabling the model to learn the mapping relationship between the input features such as the control quantity of CT gantry tilt and the current state to the angle at the next moment, and ensuring that the final prediction result conforms to the basic laws of motion physics.

[0095] Step S403: Validate the tilt motion prediction model.

[0096] To verify the reliability of the proposed model, the predictive performance can be evaluated on a real-world test dataset using metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Symmetric Mean Absolute Percentage Error (SMAPE), and Coefficient of Determination (R²).

[0097] In this implementation, a rack state parameter dataset is constructed; the state parameters include at least one of tilt angle, target angle, starting angle, stopping angle, timestamp, duty cycle, direction of motion, and stop motion signal; a neural network model of rack tilt is built; and the neural network model is trained using the state parameter dataset to obtain a tilt motion prediction model; the tilt motion prediction model is then validated; thereby improving the prediction accuracy of the neural network model.

[0098] Figure 6A Figure 6 shows a flowchart illustrating a method for controlling the tilt of a medical equipment gantry in another embodiment. Taking a CT system as an example, the method includes the following steps:

[0099] S601, Input the given target tilt angle of the CT gantry; Obtain the current actual tilt angle of the CT gantry;

[0100] S602, calculate the error angle of the CT gantry based on the target tilt angle and the actual tilt angle of the gantry;

[0101] S603, compare the error angle with the preset angle threshold. If the difference is less than or equal to the preset angle threshold, the frame stops tilting. If the difference is greater than the preset angle threshold, proceed to step S604.

[0102] S604 updates the target tilt angular velocity controlled by the position loop of the controller based on the error angle;

[0103] S605, based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack, the tilt angular velocity error is obtained;

[0104] S606, based on the tilt angular velocity error, obtains the pulse width modulation duty cycle of the motor;

[0105] S607 updates the speed loop control parameters of the controller, obtains the updated pulse width modulation duty cycle based on the pulse width modulation duty cycle and the updated speed loop control parameters of the controller, and adjusts the tilt angle of the frame based on the updated pulse width modulation duty cycle.

[0106] S608, obtain the adjusted tilt angle and the target position of the rack, and calculate the difference between the adjusted tilt angle and the target tilt angle;

[0107] S609, if the difference is less than or equal to the preset angle threshold, the rack stops tilting; if the difference is greater than the preset angle threshold, it continues to return to S604.

[0108] Figure 6B This is a schematic diagram of the workflow for controlling the tilt of a medical device gantry in another embodiment. The diagram specifically illustrates the process from initializing local variables to inputting a given target angle value, obtaining the actual tilt angle of the gantry based on an inclinometer, setting a threshold for the tilt angle, and achieving adaptive real-time control of the gantry tilt angle through position loop control and velocity loop control, thereby improving the accuracy and timeliness of the adjustment.

[0109] In the medical device rack tilt control method of this application embodiment, the actual tilt angle of the medical device rack at the current moment is obtained, and the error angle of the rack is calculated based on the target tilt angle and the actual tilt angle. Based on the error angle, the target tilt angular velocity controlled by the position loop of the controller is updated. Based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack, the tilt angular velocity error is obtained. Based on the tilt angular velocity error, the pulse width modulation duty cycle is obtained through the speed loop control of the controller. The speed loop control parameters of the controller are updated, and the updated pulse width modulation duty cycle is obtained based on the duty cycle and the updated speed loop control parameters of the controller. The tilt angle of the rack is adjusted based on the updated pulse width modulation duty cycle. Through the setting of the dual-loop control loop, the rack can more accurately adjust the duty cycle according to the angular velocity error, thereby adaptively adjusting the tilt angle of the rack in real time.

[0110] It should be understood that, although Figure 6A 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 6AAt 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.

[0111] Based on the same inventive concept, this application also provides a data transmission apparatus for implementing the data transmission method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, specific limitations in one or more data transmission apparatus embodiments provided below can be found in the limitations of the data transmission method described above, and will not be repeated here.

[0112] In one embodiment, such as Figure 7 As shown, a control device for tilting a medical equipment rack is provided, comprising:

[0113] The acquisition module is used to acquire the actual tilt angle of the rack of the medical equipment at the current moment, and calculate the error angle of the rack based on the target tilt angle and the actual tilt angle.

[0114] The first update module is used to update the target tilt angular velocity controlled by the position loop of the controller based on the error angle.

[0115] The calculation module is used to obtain the tilt angular velocity error based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack;

[0116] The adjustment module is used to obtain the pulse width modulation duty cycle based on the tilt angular velocity error through the speed loop control of the controller;

[0117] The second update module is used to update the speed loop control parameters of the controller, and obtains the updated pulse width modulation duty cycle based on the duty cycle and the updated speed loop control parameters of the controller.

[0118] The adjustment module is used to adjust the tilt angle of the rack according to the updated pulse width modulation duty cycle.

[0119] In one embodiment, the device further includes:

[0120] The second acquisition module is used to acquire the parameters of the speed loop control of the controller, including at least one of the proportional, integral, and derivative coefficients;

[0121] The second calculation module is used to obtain the predicted tilt angular velocity error based on the parameters of the controller's speed loop control and the tilt motion prediction model.

[0122] The third update module is used to input the predicted tilt angular velocity error into the BP neural network to obtain the updated pulse width modulation duty cycle.

[0123] In one embodiment, the device further includes:

[0124] The third acquisition module is used to acquire the current status parameters of the rack;

[0125] The prediction module is used to input the current state parameters into the tilt motion prediction model to obtain the tilt prediction angle of the frame at the next moment.

[0126] The third calculation module is used to calculate the predicted tilt angular velocity error based on the predicted tilt angle and the target tilt angle at the next moment. Specific limitations regarding the control device for medical equipment rack tilting can be found in the limitations of the control method for medical equipment rack tilting described above, and will not be repeated here. Each module in the aforementioned control device for medical equipment rack tilting can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0127] In one embodiment, a CT system is further included, characterized in that the CT system comprises:

[0128] frame;

[0129] The X-ray tube, mounted on the frame, is used to generate X-rays;

[0130] The detector, mounted on the frame, is used to receive X-rays passing through the object being inspected. The X-ray tube and detector can be driven to rotate circumferentially along the frame.

[0131] The motor is used to drive the frame to tilt forward of its own axis of rotation or to tilt backward of its own axis of rotation.

[0132] Inclinometer, mounted on the frame, is used to determine the frame tilt angle;

[0133] The controller includes position loop control and speed loop control. In position loop control, the controller is used to obtain the actual tilt angle of the rack at the current moment, and calculate the error angle of the rack based on the target tilt angle and the actual tilt angle. Based on the error angle, the target tilt angular velocity is determined.

[0134] In the speed loop control, the controller is used to obtain the actual tilt angular velocity of the rack and to obtain the tilt angular velocity error based on the target tilt angular velocity and the actual tilt angular velocity of the rack.

[0135] The pulse width modulation duty cycle of the motor is obtained based on the tilt angular velocity error.

[0136] The motor drives the frame's movement by adjusting the duty cycle of the motor's pulse width modulation.

[0137] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 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 data related to a control method for tilting a medical device rack. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a control method for tilting a medical device rack.

[0138] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0139] Obtain the current actual tilt angle of the medical equipment rack, and calculate the rack error angle based on the target tilt angle and the actual tilt angle.

[0140] Based on the error angle, the target tilt angular velocity controlled by the controller's position loop is updated;

[0141] The tilt angular velocity error is obtained based on the updated target tilt angular velocity and the actual tilt angular velocity of the rack.

[0142] The pulse width modulation duty cycle is obtained based on the tilt angular velocity error;

[0143] The speed loop control parameters of the controller are updated, and the updated pulse width modulation duty cycle is obtained based on the duty cycle and the updated speed loop control parameters of the controller.

[0144] Adjust the rack tilt angle based on the updated pulse width modulation duty cycle.

[0145] The processor executes computer programs to implement the method steps in the other embodiments described above.

[0146] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0147] Obtain the current actual tilt angle of the medical equipment rack, and calculate the rack error angle based on the target tilt angle and the actual tilt angle.

[0148] Based on the error angle, the target tilt angular velocity controlled by the controller's position loop is updated;

[0149] The tilt angular velocity error is obtained based on the updated target tilt angular velocity and the actual tilt angular velocity of the frame;

[0150] The pulse width modulation duty cycle is obtained based on the tilt angular velocity error;

[0151] The speed loop control parameters of the controller are updated, and the updated pulse width modulation duty cycle is obtained based on the duty cycle and the updated speed loop control parameters of the controller.

[0152] Adjust the rack tilt angle based on the updated pulse width modulation duty cycle.

[0153] Furthermore, when the computer program is executed by the processor, it implements the steps in the above-described method embodiments.

[0154] 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.

[0155] 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.

[0156] 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 for controlling the tilting of a medical equipment rack, wherein the medical equipment rack is driven by a motor, and the motor is controlled by a controller, characterized in that, The method includes: The actual tilt angle of the rack of the medical device at the current moment is obtained, and the error angle of the rack is calculated based on the target tilt angle of the rack and the actual tilt angle. Based on the error angle, the target tilt angular velocity controlled by the controller's position loop is updated; The tilt angular velocity error is obtained based on the updated target tilt angular velocity and the actual tilt angular velocity of the frame; The pulse width modulation duty cycle of the motor is obtained based on the tilt angular velocity error. The speed loop control parameters of the controller are updated, and the updated pulse width modulation duty cycle is obtained based on the pulse width modulation duty cycle and the updated speed loop control parameters of the controller. The driver is controlled to adjust the tilt angle of the rack according to the updated pulse width modulation duty cycle.

2. The method according to claim 1, characterized in that, The step of updating the speed loop control parameters of the controller, and obtaining the updated pulse width modulation duty cycle based on the pulse width modulation duty cycle and the updated speed loop control parameters of the controller, includes: Obtain the parameters of the speed loop control of the controller, wherein the parameters include at least one of proportional, integral, and derivative coefficients; Based on the parameters of the speed loop control of the controller and the tilt motion prediction model, the predicted tilt angular velocity error is obtained; The predicted tilt angular velocity error is input into the BP neural network to obtain the updated pulse width modulation duty cycle.

3. The method according to claim 2, characterized in that, The step of obtaining the predicted tilt angular velocity error based on the speed loop control parameters of the controller and the tilt motion prediction model includes: Obtain the current state parameters of the rack; The current state parameters are input into the tilt motion prediction model to obtain the tilt prediction angle of the rack at the next moment. The predicted tilt angular velocity error is calculated based on the predicted tilt angle at the next moment and the target tilt angle at the next moment.

4. The method according to claim 2, characterized in that, The training process of the tilt motion prediction model includes: Construct a rack status parameter dataset; the status parameters include at least one of tilt angle, target angle, start angle, stop angle, timestamp, duty cycle, direction of motion, and stop motion signal; A neural network model of the rack tilt is constructed; and the neural network model is trained using the state parameter dataset to obtain a tilt motion prediction model. The tilt motion prediction model was validated.

5. The method according to claim 4, characterized in that, The process of training the tilt motion prediction model to obtain the tilt motion prediction model includes: Calculate the physical information loss of the tilt motion model based on the physical information loss function; Calculate the data information loss of the tilt motion model based on the data information loss function; Based on the physical information loss and the data information loss, the total loss of the tilting motion model is obtained; The frame tilting motion model is trained based on the total loss.

6. The method according to claim 1, characterized in that, The step of updating the target tilt angular velocity of the controller's position loop control based on the error angle includes: The error angle is input into the position loop control of the controller to obtain the initial angular velocity control value; The initial angular velocity control value is compared with the preset tilt angular velocity threshold. If the initial angular velocity control value is less than the preset threshold, then the initial angular velocity control value is the target tilt angular velocity. If the initial angular velocity control value is greater than the preset tilt angular velocity threshold, then the preset tilt angular velocity threshold is used as the target tilt angular velocity.

7. The method according to claim 1, characterized in that, The method further includes: Obtain the adjusted tilt angle and the target position of the frame, and calculate the difference between the adjusted tilt angle and the target tilt angle; If the difference is less than or equal to a preset angle threshold, the frame stops tilting. If the difference is greater than a preset angle threshold, then return to the step of updating the target tilt angular velocity of the controller's position loop control based on the error angle.

8. A control device for tilting a medical equipment rack, characterized in that, The device includes: The acquisition module is used to acquire the actual tilt angle of the rack of the medical device at the current moment, and calculate the error angle of the rack based on the target tilt angle of the rack and the actual tilt angle. The first update module is used to update the target tilt angular velocity controlled by the position loop of the controller according to the error angle. The calculation module is used to obtain the tilt angular velocity error based on the updated target tilt angular velocity and the actual tilt angular velocity of the frame; The first adjustment module is used to obtain the pulse width modulation duty cycle of the motor based on the tilt angular velocity error; The second update module is used to update the speed loop control parameters of the controller, and obtains the updated pulse width modulation duty cycle based on the duty cycle and the updated speed loop control parameters of the controller. The second adjustment module is used to adjust the tilt angle of the rack according to the updated pulse width modulation duty cycle.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A CT system, characterized in that, The CT system includes: frame; An X-ray tube, mounted on the frame, is used to generate X-rays; A detector, mounted on the frame, is used to receive X-rays passing through the object being inspected; the X-ray tube and the detector can be driven to rotate circumferentially along the frame. An electric motor is used to drive the frame to tilt forward of its own rotation axis or to tilt the frame backward of its own rotation axis. An inclinometer, mounted on the frame, is used to determine the inclinometer angle of the frame; The controller includes a position loop control and a speed loop control. In the position loop control, the controller is used to obtain the actual tilt angle of the rack at the current moment, calculate the error angle of the rack based on the target tilt angle of the rack and the actual tilt angle, and determine the target tilt angular velocity based on the error angle. In the speed loop control, the controller is used to obtain the actual tilt angular velocity of the frame, and to obtain the tilt angular velocity error based on the target tilt angular velocity and the actual tilt angular velocity of the frame; The pulse width modulation duty cycle of the motor is obtained based on the tilt angular velocity error. The motor is controlled to drive the frame's movement based on the pulse width modulation duty cycle of the motor.

Images