An automatic patient repositioning system and method based on 3D marker points

Abstract

The present invention relates to a patient automatic resetting system and a resetting method based on 3D marking points. The present invention comprises a resetting platform module, a patient transport module, a visualization module, an image acquisition module, an image processing module and a TCS host computer control system; the image acquisition realizes the image acquisition function; the visualization module comprises two cross laser lights and one straight laser light, one straight laser light is arranged on the top to project a vector line, and the two cross laser lights are arranged on the left and right of the equipment or the resetting platform to project a transverse line and a crown line for assisting a physicist to adjust the position and posture of a treatment bed; the image processing module is composed of three-dimensional point cloud software and a computer dedicated to images, and the position and posture resetting information is transmitted to the host computer in real time for display to assist the physicist to adjust the positioning bed; the patient resetting platform module is composed of a ZONDA six-axis parallel platform, and the resetting information is transmitted to the TCS host computer control system, and X, Y, Z, Rx, Ry and Rz are input to complete the control and adjustment of the positioning bed; the patient transport device is composed of an AGV trolley, and is used to transfer the patient between a resetting room and a treatment room.

Description

Technical Field

[0001] The present invention relates to the medical field, and in particular to a patient automatic repositioning system and method based on 3D marking points. Background Art

[0002] Gamma Knife is a large-scale radiotherapy device that integrates stereotactic technology and radiosurgery technology and is mainly used to treat tumors. It cross-superimposes the radiation beams so that the lesion area receives a high dose, thereby killing cancer cells. In current clinical applications, the patient's repositioning operation process first requires a positioning CT scan before treatment. The positioning CT scan is used to associate the position relationship between the patient's lesion site and the gamma knife positioning bed. Before treatment, the patient completes the position repositioning on the positioning bed in the treatment room before treatment. The consistency between the patient's repositioning position on the positioning bed and the position during the CT scan is a key factor in precision radiotherapy. The following two methods are commonly used to ensure the patient's repositioning accuracy: (1) the coordinated use of negative pressure bags, positioning films, repositioning rulers, and marking points to ensure the patient's repositioning; (2) the coordinated use of negative pressure bags, positioning films, laser lights, and marking points to ensure the patient's repositioning. Both repositioning methods are completed in the treatment room and require more technicians to operate. Therefore, the current repositioning method requires too much time in the treatment room for repositioning, which affects the efficiency of the treatment room. In addition, the repositioning error caused by manual operation is large, which affects the treatment effect.

[0003] Therefore, in order to ensure the accuracy of treatment and improve the efficiency of equipment use, changing the resetting method and improving automation and information technology are the current urgent clinical needs. The present invention provides a patient automatic resetting system based on 3D markers for the resetting operation of patients, which can avoid the errors caused by manual resetting and improve the treatment effect of gamma knife. Summary of the invention

[0004] In order to solve the above-mentioned technical problems existing in the background technology, the present invention provides a patient automatic repositioning system and repositioning method based on 3D markers, which can more effectively avoid the errors caused by the physiotherapist's repositioning, reduce the difficulty of operation, and improve the treatment effect and efficiency of the equipment.

[0005] The technical solution of the present invention is: the present invention is a patient automatic repositioning system based on 3D markers, which is special in that: the patient automatic repositioning system based on 3D markers includes a repositioning platform module, a patient transfer module, a visualization module, an image acquisition module, an image processing module and a TCS host computer control system; the image acquisition module includes a binocular camera and a 3D camera with blue structured light, the binocular camera and the 3D camera with blue structured light realize the image acquisition function, and send the point cloud data to the end of the buffer zone of the image processing module in real time through the network port for storage, and the buffer zone head of the image processing module reads the data in real time to calculate the position and posture relationship and feedback; the visualization module includes two cross laser lights and one straight laser light, one straight laser light is placed on the top to project the vector line, and the two cross laser lights are placed on the left and right of the equipment or the repositioning platform to project the transverse line and the coronal line to assist the physicist in adjusting the position and posture of the treatment bed; the image processing module is composed of three-dimensional point cloud software and a computer dedicated to image, It can process millions of points of information in parallel, and transmit the posture reset information to the TCS host control system in real time for display, to assist the physicist in adjusting the positioning bed; the patient reset platform module is composed of the ZONDA six-axis parallel platform, and the reset information is transmitted to the TCS host control system, and the input is X, Y, Z, Rx, Ry, and Rz, where X, Y, Z, Rx, Ry, and Rz are the patient's axial displacement and posture deflection angle in the X, Y, and Z directions, to complete the control and adjustment of the positioning bed; the patient transfer device is composed of an AGV trolley, on which the positioning bed adapter and gripping device are fixed to transfer the patient between the reset room and the treatment room. The patient automatic reset system based on 3D markers uses reflective marker balls and human markers as two reference points, and the 3D camera with blue structured light can accurately identify the markers as posture information; the image processing module is used to feedback the contour information of the patient and the positioning bed in real time, and the visualization module is combined to accurately complete the reset operation of the patient and the positioning bed.

[0006] Furthermore, the graphics-dedicated computer is an RTX 2080Ti GPU module.

[0007] Furthermore, in the ZONDA six-axis parallel platform, the X and Y travels are 400mm, the Z travel is 300mm, the Rx, Ry, and Rz travels are ±20°, and the load capacity is not less than 300Kg.

[0008] A method for realizing the above-mentioned patient automatic repositioning system based on 3D markers is special in that the method comprises the following steps:

[0009] 1) Use medical solution to mark cross lines on the patient's body surface, such as the sternum and ribs, as the human body reference position; place 3D reflective marker balls on the corners of the positioning bed as reference points; and fix them to the positioning bed with a negative pressure belt;

[0010] 2) Turn on the laser light based on the fixed position to mark the checkpoints; and use the coordinate calibration plate to calibrate the world coordinate system of the working area;

[0011] 3) Perform CT scan to determine the relative position of the patient and the positioning bed ori , and form an initial image;

[0012] 4) The positioning bed adapter and grabbing device based on the AGV trolley transfer the patient to the reduction room;

[0013] 5) The physicist fixes the patient on the reset platform according to the process, turns on the image acquisition module, adjusts the exposure parameters, gain parameters, etc. to adapt to the current ambient light to ensure the quality of the acquired image is complete, and then uses the image acquisition module with an acquisition frequency of 60 Hz to confirm the acquisition of the reset image;

[0014] 6) The image is transmitted to the cache area of ​​the image processing module in real time based on the TCP protocol. The image processing module reads the cache data in real time, and calculates the relative posture relationship between the patient's marker points and the positioning bed in parallel through the image-specific computer, and compares and analyzes the image with the initial image collected by the CT room in step 3) to obtain the reset deviation information D = [X, Y, Z, Rx, Ry, Rz]; wherein X, Y, Z, Rx, Ry, Rz are the patient's X, Y, Z axial displacement and posture deflection angle;

[0015] 7) The TCS host computer control system inputs the reset deviation information D, adjusts the axial displacement and posture deviation of the positioning bed, and adjusts the patient to the initial state in the positioning room to achieve the reset purpose by combining the feedback information of the visualization module and the three-dimensional contour displayed in real time by the three-dimensional point cloud software and image-specific calculation;

[0016] 8) Use a cube with matte surface treatment or titanium powder spraying to calibrate the camera's coordinate system accuracy.

[0017] Furthermore, the specific steps of step 2) are as follows: regard the upper left corner of the calibration plate as the coordinate origin, align the single axis with the calibration plate, align the plane with the depth data, manually set the camera position and clear the calibration data of the current camera's working range to complete the calibration, adjust the device exposure parameters, gain parameters, etc. to adapt to the ambient light, and then use the image acquisition module to capture the initial image.

[0018] Furthermore, the specific steps of step 8) are as follows: the cube is placed in the camera field of view, four corner points are collected respectively, the volume of the cube is calculated, and the statistical deviation is compared with the theoretical volume, so as to calibrate and correct the world coordinate system to ensure the collection accuracy.

[0019] The patient automatic repositioning system and repositioning method based on 3D markers provided by the present invention can perform automatic repositioning during treatment, reduce manual operations, improve treatment accuracy, reduce subjective errors caused by physicists' repositioning through automatic correction of repositioning errors, and improve the level of precision radiotherapy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a system block diagram of the present invention;

[0021] Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0022] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] See also Figure 1 The structure of the specific embodiment of the present invention includes a patient resetting platform module, a patient transfer module, a visualization module, an image acquisition module, an image processing module and a TCS host computer control system.

[0024] The present invention adds a reset room in the radiotherapy area, and each of the reset room and the CT room is equipped with a set of image acquisition devices based on 3D cameras; marking points are added to the treatment positioning bed to locate the relationship between the patient and the positioning bed during image acquisition and processing; the patient is transferred from the reset room to the treatment room using the patient transfer module, and the surface contour data of the patient and the positioning bed are collected and transmitted to the image processing module in real time, and the image processing module calculates the posture relationship of the patient relative to the positioning bed. The physicist adjusts the positioning bed according to the relationship data to keep the patient in the relative posture relationship in the reset room;

[0025] In the specific scheme of the present invention, a cross line is marked with medical solution at the shoulder blade or rib of the patient's body surface as the human body reference position. A 3D reflective marker ball is placed at the corner of the positioning bed as a reference point. In the image acquisition module, a binocular camera plus a 3D camera with blue structured light realizes the image acquisition function, and sends the point cloud data to the end of the image processing module buffer zone through the network port in real time for storage. The head of the image processing module buffer zone reads the data in real time to calculate the position and posture relationship and feedback.

[0026] The visualization module consists of two cross laser lights, one straight laser light and laser light fixing accessories. One straight laser light is placed on the top to project the vector line, and two cross laser lights are placed on the left and right of the equipment or reset platform to project the transverse line and the coronal line to assist the physiotherapist in adjusting the position and posture of the treatment bed; the laser light fixing accessories are used to fix the two cross laser lights and one straight laser light.

[0027] The image processing module consists of 3D point cloud software and a dedicated image computer. The dedicated image computer is an RTX 2080Ti GPU module, which can process millions of points in parallel and transmit the posture reset information to the TCS host control system in real time for display, assisting the physicist in adjusting the positioning bed.

[0028] The patient repositioning platform module is composed of the ZONDA six-axis parallel platform of Dongzhi Precision, with X and Y strokes of 400mm, Z stroke of 300mm, Rx, Ry, Rz strokes of ±20°, and a load capacity of not less than 300Kg.

[0029] The reset information is transmitted to the TCS host control system, and the axial displacement and attitude angle of X, Y, Z, Rx, Ry, and Rz are input to complete the control and adjustment of the positioning bed.

[0030] The patient transfer device consists of an AGV trolley with a load capacity greater than 300Kg and shock-absorbing capabilities. The AGV trolley is fixed with a positioning bed adapter and a grabbing device to transfer patients between the reduction room and the treatment room.

[0031] See also Figure 2 The present invention provides a patient automatic repositioning method based on 3D markers, and the specific implementation method is as follows:

[0032] 1) Use medical solution to mark cross lines on the patient's body surface, such as the sternum and ribs, as the human body reference position;

[0033] Place 3D reflective marker balls on the corners of the positioning bed as reference points and fix them to the positioning bed with fixing devices such as negative pressure belts;

[0034] 2) Based on the fixed position, turn on the laser light to mark the checkpoints; and use the coordinate calibration plate to calibrate the world coordinate system of the working area, regard the upper left corner of the calibration plate as the coordinate origin, use the calibration plate to align the single axis, use the depth data to align the plane, manually set the camera position and clear the calibration data of the current camera's working range to complete the calibration, adjust the device exposure parameters, gain parameters, etc. to adapt to the ambient light, and then use the image acquisition module to capture the initial image;

[0035] 3) Perform CT scan to determine the relative position of the patient and the positioning bed ori , and form an initial image;

[0036] 4) The positioning bed adapter and grabbing device based on the AGV trolley transfer the patient to the reduction room;

[0037] 5) The physicist fixes the patient on the reset platform according to the process, turns on the image acquisition module, adjusts the exposure parameters, gain parameters, etc. to adapt to the current ambient light to ensure the quality of the acquired image is complete, and then uses the image acquisition module with an acquisition frequency of 60 Hz to confirm the acquisition of the reset image;

[0038] 6) The image is transmitted to the cache area of ​​the image processing module in real time based on the TCP protocol. The image processing module reads the cache data in real time, and calculates the relative posture relationship between the patient's marker points and the positioning bed in parallel through the RTX 2080Ti GPU module, and compares and analyzes the image with the initial image collected in the CT room in step 3) to obtain the reset deviation information D = [X, Y, Z, Rx, Ry, Rz]; wherein X, Y, Z, Rx, Ry, Rz are the patient's X, Y, Z axial displacement and posture deflection angle;

[0039] 7) The TCS host control system inputs the reset deviation information D, adjusts the axial displacement and posture deviation of the positioning bed, and combines the feedback information of the visualization module and the three-dimensional contour displayed in real time by the three-dimensional point cloud software and image-specific calculation to adjust the patient to the initial state in the positioning room to achieve the purpose of reset;

[0040] 8) Use a cube with matte surface treatment or titanium powder spraying to calibrate the camera coordinate system. Specifically, place the cube in the camera field of view, collect four corner points respectively, calculate the volume of the cube, compare the statistical deviation with the theoretical volume, and calibrate the world coordinate system to ensure the collection accuracy.

[0041] The content of the present invention and the technical content not specifically described in the above embodiments are the same as the prior art.

[0042] The above are only specific embodiments disclosed in the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A patient automatic repositioning system based on 3D markers, characterized in that: The patient automatic repositioning system based on 3D markers includes a patient repositioning platform module, a patient transfer module, a visualization module, an image acquisition module, an image processing module and a TCS host computer control system; the image acquisition module includes a binocular camera and a 3D camera with blue structured light, the binocular camera and the 3D camera with blue structured light realize the function of image acquisition, and send the point cloud data to the end of the buffer zone of the image processing module in real time through the network port for storage, and the buffer zone head of the image processing module reads the data in real time to calculate the position and posture relationship and feedback; the visualization module includes two cross laser lights and one straight laser light, one straight laser light is placed on the top to project the vector line, and the two cross laser lights are placed on the left and right of the equipment or patient repositioning platform to project the transverse line and the coronal line to assist the physicist in adjusting the position and posture of the treatment bed; the image processing module is composed of three-dimensional point cloud software and a dedicated image computer, which can perform parallel processing of millions of point information and convert the position and posture relationship into real-time data. The reset information is transmitted to the TCS host computer control system in real time for display, assisting the physicist to adjust the positioning bed; the patient reset platform module is composed of a ZONDA six-axis parallel platform, and the reset information is transmitted to the TCS host computer control system. X, Y, Z, Rx, Ry, and Rz are input to complete the control and adjustment of the positioning bed. X, Y, Z, Rx, Ry, and Rz are the patient's X, Y, and Z axial displacements and posture deflection angles; the patient transfer module is composed of an AGV trolley, and the AGV trolley is fixed with a positioning bed adapter and a gripping device to transfer the patient between the reset room and the treatment room. The patient automatic reset system based on 3D markers uses reflective marker balls and human markers as two reference reference points, and the blue structured light 3D camera can accurately identify the markers as posture information; the image processing module is used to feedback the contour information of the patient and the positioning bed in real time, and the visualization module is combined to accurately complete the reset operation of the patient and the positioning bed.

2. The patient automatic repositioning system based on 3D markers according to claim 1, characterized in that: The image-specific computer is an RTX 2080Ti GPU module.

3. The patient automatic repositioning system based on 3D markers according to claim 2, characterized in that: In the ZONDA six-axis parallel platform, the X and Y strokes are 400mm, the Z stroke is 300mm, the Rx, Ry, and Rz strokes are ±20°, and the load capacity is not less than 300Kg.

4. A method for implementing the patient automatic repositioning system based on 3D markers according to claim 1, characterized in that: The method comprises the following steps: 1) Use medical solution to mark cross lines on the rigid parts of the patient's body, including the sternum and ribs, as the human body reference position; place 3D reflective marker balls on the corners of the positioning bed as reference points; and fix them to the positioning bed with a negative pressure belt; 2) Turn on the laser light based on the fixed position to mark the checkpoints; and use the coordinate calibration plate to calibrate the world coordinate system of the working area; 3) Perform CT scan to determine the relative position Dori between the patient and the positioning bed and form an initial image; 4) The AGV-based positioning bed adapter and grabbing device transfer the patient to the reduction room; 5) The physicist fixes the patient to the patient resetting platform according to the process, turns on the image acquisition module, adjusts the exposure parameters and gain parameters to adapt to the current ambient light to ensure the quality of the acquired image is complete, and then uses the image acquisition module with an acquisition frequency of 60Hz to confirm the acquisition of the resetting image; 6) The image is transmitted to the cache area of ​​the image processing module in real time based on the TCP protocol. The image processing module reads the cache data in real time, and calculates the relative posture relationship between the patient's marker points and the positioning bed in parallel through the image-specific computer, and compares the image with the initial image collected in the CT room in step 3) to obtain the reset deviation information D=[X,Y,Z,Rx,Ry,Rz]; where X,Y,Z,Rx,Ry,Rz are the patient's X,Y,Z axial displacement and posture deflection angle; 7) The TCS host computer control system inputs the reset deviation information D, adjusts the axial displacement and posture deviation of the positioning bed, and adjusts the patient to the initial state in the positioning room to achieve the reset purpose by combining the feedback information of the visualization module and the 3D point cloud software and the 3D contour displayed in real time by the image-specific calculation; 8) Use a cube with matte surface treatment or titanium powder spraying to calibrate the camera's coordinate system accuracy.

5. The resetting method according to claim 4, characterized in that: The specific steps of step 2) are as follows: regard the upper left corner of the calibration plate as the coordinate origin, align the single axis with the calibration plate, align the plane with the depth data, manually set the camera position and clear the calibration data of the current camera's working range to complete the calibration, adjust the device exposure parameters and gain parameters to adapt to the ambient light, and then use the image acquisition module to acquire the initial image.

6. The resetting method according to claim 5, characterized in that: The specific steps of step 8) are as follows: place the cube under the camera field of view, collect four corner points respectively, calculate the volume of the cube, compare the statistical deviation with the theoretical volume, and calibrate and correct the world coordinate system to ensure the collection accuracy.

Images