Automated QA method for accelerator system based on combined motifs

Abstract

The invention provides an automated QA method for an accelerator system based on a combined phantom. The combined phantom comprises a tungsten ball center point module, a dose monitoring module and an image analysis module which are integrated into one. The method comprises: placing the tungsten ball center point module at a theoretical isocenter position, controlling the mechanical parts of an accelerator to move to a specific position by importing a prefabricated plan, acquiring an image while beaming, and calculating a corresponding mechanical deviation by acquiring a field center, a tungsten ball position and a field boundary; importing the prefabricated plan into an accelerator system, realizing the positioning of a dose detection module by moving a treatment bed, and monitoring the change of a dose rate over time and a final beam-out dose during the beam-out process of the accelerator; and an image analysis module comprising a Las Vegas aperture and line pairs representing different resolutions, importing the prefabricated plan into an accelerator system, realizing the positioning of the image analysis module by moving the treatment bed, controlling the beam out of the accelerator and acquiring an image, and acquiring an image comprising the aperture or the line pair.

Description

Technical Field

[0001] The present invention relates to the field of radiotherapy, and in particular to an automated QA method for an accelerator system based on a combined phantom. Background Art

[0002] As the main equipment for radiotherapy, the accelerator and its components need to maintain extremely high stability during use after leaving the factory. The operation of its components needs to be tested regularly to ensure that they meet the treatment requirements.

[0003] Before the accelerator leaves the factory, it needs to be inspected and verified for specific items to ensure that it meets the requirements for use. In daily use in hospitals, the accelerator also needs to be inspected regularly and if there is a problem, it needs to be stopped immediately for maintenance to ensure that it can be used later after it has been restored to its ideal state.

[0004] The items that need to be tested include machinery, dosage, grating, imaging, etc. There are slight changes in the actual implementation process according to the specific requirements.

[0005] The inspection of the accelerator system is one of the most important tasks before the accelerator leaves the factory, and it is also something that hospitals need to perform regularly during actual use. Although the technical solutions and coverage provided by each company are slightly different, they generally include the verification of the frame, small head, bed board and other mechanical parts; the verification of dose linearity and stability; the verification of grating positioning accuracy, positioning repeatability, and transmission, and the verification of image contrast, resolution, and artifacts.

[0006] The verification of the mechanical part usually uses the front pointer to first determine the physical isocenter, and then uses this as the standard center to control the movement of each part of the machine to compare the degree of overlap between the mechanical movement center and the physical isocenter. For the verification of the dose part, it is necessary to use a dosimeter or ionization chamber to record the beam-emitting process of the accelerator, obtain the dose data by executing a specific test plan, and calculate to see if the calculation result meets the set threshold. For the verification of the grating, the light field is generally used for verification. By forming light field shapes of different sizes, it is supplemented by naked eye observation to confirm whether its boundaries and sizes are consistent with the set values. The verification of the image will directly use the image plate to capture the image, and observe the specific area with the naked eye to confirm whether the image quality meets the requirements.

[0007] Whether it is pre-shipment inspection or regular inspection after shipment, it involves many items and is a systematic and complex process. Various auxiliary equipment is required, and the operator has high quality requirements. It takes a lot of time to complete the entire inspection process, and the operator needs to concentrate, and it is necessary to replace various equipment and record various execution data. Summary of the invention

[0008] In order to solve the above technical problems, the present invention discloses an automated QA method for an accelerator system based on a combined phantom, which integrates the detection contents of mechanical, dosage, and accessory parts into a fully automated process, reduces the use of auxiliary equipment and the work of operators, thereby shortening the overall execution time.

[0009] To achieve the above object, the technical solution of the present invention is: an automated QA method for an accelerator system based on a combined phantom, wherein the combined phantom includes a tungsten ball center point module, a dose monitoring module, and an image analysis module integrated into one, and the method includes the following steps:

[0010] When the tungsten ball center point module is used, the tungsten ball center point module is placed at a theoretical isocenter position, and the mechanical parts of the accelerator are controlled to move to a specific position by importing a prefabricated plan of the accelerator, and an image is acquired through an imaging board while the accelerator emits a beam, and the corresponding mechanical deviation is calculated by acquiring the field center, the tungsten ball position, and the field boundary;

[0011] When the dose monitoring module is used, the corresponding prefabricated plan is imported into the accelerator system according to the detection items. The accelerator system automatically realizes the positioning of the dose detection module by moving the treatment bed, and automatically controls the accelerator beam output, and monitors the change of dose rate over time and the final beam output dose during the accelerator beam output process;

[0012] The image analysis module includes Las Vegas holes and line pairs representing different resolutions. When using the image analysis module, the corresponding prefabricated plan is imported into the accelerator system according to the detection project. The accelerator system automatically realizes the positioning of the image analysis module by moving the treatment bed, and then obtains the image including the hole or line pair by controlling the accelerator to emit beams and collecting images through the image board.

[0013] In a further technical solution, the tungsten ball center point module includes a tungsten ball mounting seat, a connecting rod inlaid with a tungsten ball, and a three-dimensional manual fine-tuning slide. The objects detected by the tungsten ball center point module include at least one of the following: the deviation between the mechanical center and the isocenter of the field, the deviation between the rotation center of the frame and the isocenter, the deviation between the rotation center of the treatment bed and the isocenter, the positioning accuracy of the grating, and the positioning repeatability of the grating.

[0014] In a further technical solution, when detecting the deviation between the mechanical center and the field isocenter, the prefabricated plan includes 8 fields, and the angle positions of the gantry and the treatment bed of each field are different, namely, gantry: 60°, treatment bed: 0°, gantry: 60°, treatment bed: 338°, gantry: 130°, treatment bed: 0°, gantry: 130°, treatment bed: 338°, gantry: 230°, treatment bed: 0°, gantry: 230°, treatment bed: 22°, gantry: 300°, treatment bed: 0°, gantry: 300°, treatment bed: 22°;

[0015] Among them, the method of obtaining the field center and the method of obtaining the tungsten ball position are used to obtain the field center and tungsten ball center on the image returned by the image plate corresponding to each field, and the corresponding deviations are calculated. According to the gantry angle, the treatment bed angle, and the deviation between the two centers, the three-dimensional deviation between the mechanical center and the field isocenter is calculated by the WinstonLutz method to determine whether the threshold is met.

[0016] In a further technical solution, when detecting the deviation between the gantry rotation center and the isocenter, the prefabricated plan includes 6 fields, each of which has a different gantry angle position, namely: 0°, 45°, 90°, 200°, 245°, and 290°;

[0017] Among them, the method of obtaining the field center and the method of obtaining the tungsten ball position are used to obtain the field center and the tungsten ball center on the image returned by the image plate corresponding to each field, and the corresponding deviations are calculated. According to the frame angle and the deviation of the two centers, the three-dimensional deviation of the deviation between the frame rotation center and the isocenter is calculated by the WinstonLutz method to determine whether the threshold is met.

[0018] In a further technical solution, when detecting the deviation between the rotation center of the treatment couch and the isocenter, the prefabricated plan includes 5 radiation fields, each of which has a different angle position of the treatment couch, namely: 0°, 10°, 20°, 350°, and 340°;

[0019] By obtaining the position of the tungsten ball, the center of the tungsten ball on the image returned by the image plate corresponding to each field is obtained respectively. A square area with the image center as the center is taken, and each pixel in this area is taken as the center point. The distance between this center point and the center of different tungsten balls is calculated, and the variance of all 5 distances is calculated to obtain the center point with the minimum calculated variance. At this time, the maximum value of the distance between the center point and the center of the tungsten ball is the deviation, and it is judged whether the deviation meets the threshold.

[0020] In a further technical solution, when calculating the grating in-place accuracy, the prefabricated plan includes 7 fields, and the grating isocenter positions of each field are different, namely: LeafA=-70mm, LeafB=-50mm, LeafA=-50mm, LeafB=-30mm, LeafA=-30mm, LeafB=-10mm, LeafA=-10mm, LeafB=10mm, LeafA=10mm, LeafB=30mm, LeafA=30mm, LeafB=50mm, LeafA=50mm, LeafB=70mm;

[0021] The left and right or upper and lower boundary positions of the field of view are obtained by obtaining the field of view boundary, the deviation between the theoretical blade position and the obtained boundary position is calculated, and it is determined whether the deviation meets the threshold.

[0022] In a further technical solution, when calculating the repeatability of the grating in place, the prefabricated plan includes 12 fields, with a total of 4 different grating isocenter positions, namely: LeafA = -70mm, LeafB = -50mm, LeafA = -30mm, LeafB = -10mm, LeafA = 10mm, LeafB = 30mm, LeafA = 50mm, LeafB = 70mm, and each position is executed 3 times;

[0023] The left and right or upper and lower boundary positions of the field are obtained by obtaining the field boundary, and the standard deviation of the boundaries obtained three times for the grating at the same theoretical position is calculated to determine whether the standard deviation meets the threshold.

[0024] In a further technical solution, the dose monitoring module includes a three-dimensional water phantom and a dose detector inserted into the center of the three-dimensional water phantom, and the detection items of the dose monitoring module include one of dose repeatability and dose linearity.

[0025] In a further technical solution, when testing dose repeatability, the pre-made plan includes 10 radiation fields, each of which is the same, and the beam dose is 10MU. The dose accumulation during the beam discharge process is recorded by a dose detector, and the actual final accumulation of each radiation field is returned. The method for determining dose repeatability is to calculate the coefficient of variation: coefficient of variation = standard deviation of measured dose / mean of measured dose*100%;

[0026] When testing dose linearity, the pre-made plan contains 25 radiation fields with 5 different beam doses, namely: 100MU, 300MU, 500MU, 700MU, and 900MU. Each dose is executed 5 times. The dose accumulation of the beam process will be recorded by the dose detector, and the actual final accumulation of each radiation field will be returned. The method to determine the dose repeatability is to calculate the deviation: deviation = ABS (measured dose - theoretical dose) / beam dose * 100%.

[0027] In a further technical solution, the detection item of the image analysis module includes one of detecting image contrast and detecting image resolution;

[0028] When detecting the image contrast, the coverage area of ​​each hole on the image is obtained according to the relative physical position and aperture of the center of the image analysis module and each hole, and the average grayscale value in the coverage area of ​​each hole is calculated respectively, and the average grayscale value of the square coverage area circumscribed by each hole is calculated, and then the ratio of the two is calculated to determine whether the threshold requirement is met;

[0029] When detecting image resolution, the coverage area of ​​each line pair on the image is obtained based on the relative physical position between the center of the image analysis module and each line pair area and the size of the line pair area, and the standard deviation of the grayscale in each area is calculated to determine whether the threshold requirement is met.

[0030] The present invention summarizes and integrates the detection items required by the accelerator, and integrates the auxiliary equipment that may be needed in the detection process into a single phantom, reducing the time for replacing auxiliary equipment in the original detection process. Since the execution plans of different detection items are built-in in advance, the entire execution process is completely realized by the accelerator's automatic execution plan, reducing the original time for manually controlling the accelerator's movement to emit beams. In addition, the switching of each module in the phantom is also achieved by planning to control the bed to move to different positions. After the entire detection execution process begins, the operator no longer needs to enter and exit the accelerator room, shortening the execution time. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a structural schematic diagram of the combined motif of the present invention;

[0032] Figure 2 It is a side view of the combined mold body of the present invention.

[0033] Marked in the figure: 1-three-dimensional water model, 2-positioning pin, 3-impact analysis module, 4-positioning plate, 5-three-dimensional manual fine-tuning slide, 6-bed plate, 7-tungsten ball, 8-connecting rod, 9-tungsten ball mounting seat. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described below in conjunction with specific embodiments, but the present invention is not limited to these embodiments.

[0035] In order to realize the automated QA of the accelerator system, the present invention designs an integrated phantom to be used together with the imaging board. By using different parts of the phantom for dose collection and image acquisition, and using a specific algorithm to analyze the data, the system inspection results are finally obtained, such as Figure 1 and 2 As shown, the combined motif of the present invention includes:

[0036] Image analysis module: Figure 1 The far left is a general imaging phantom containing the Las Vegas holes and lines of different resolutions;

[0037] Dose monitoring module: Figure 1 The three-dimensional water phantom in the middle position where the probe can be inserted;

[0038] Tungsten ball center point module: Figure 1 The connecting rod with a tungsten ball embedded in the far right.

[0039] The above three parts are combined into a mold body through the base groove and other mechanical parts, and it is ensured that when the mold body moves as a whole, the relative positions of the three parts of the mold body do not change, and are physically defined fixed relative positions.

[0040] 1. Use the tungsten ball center point module to place the entire phantom horizontally on the treatment bed. By manually adjusting the position of the treatment bed and the phantom, three mutually perpendicular laser light lines intersect on the tungsten ball (that is, place the tungsten ball at the theoretical isocenter position of the accelerator determined by the laser light). According to the different mechanical parts to be detected and the different project contents (the deviation between the mechanical center and the isocenter of the field, the deviation between the gantry rotation center and the isocenter, the deviation between the treatment bed rotation center and the isocenter, the grating in-place accuracy, and the grating in-place repeatability), manually select and import the corresponding pre-made plan file into the accelerator system. The accelerator system will automatically control the movement of the mechanical parts to be detected to the corresponding positions according to the contents of the plan file, and automatically open the image acquisition position of the image board (the place where the relative position with the phantom is fixed, located below the phantom and parallel to the phantom), and then control the accelerator to emit beams. After receiving radiation from the accelerator, the image board will return an image. According to the different radiation doses received by the image board, each pixel on the image will present different grayscales (the more radiation, the greater the grayscale).

[0041] 2. When detecting the deviation between the mechanical center and the field isocenter, the prefabricated plan includes 8 fields, and the angle positions of the gantry and the treatment bed of each field are different, namely, gantry: 60°, treatment bed: 0°; gantry: 60°, treatment bed: 338°; gantry: 130°, treatment bed: 0°; gantry: 130°, treatment bed: 338°; gantry: 230°, treatment bed: 0°; gantry: 230°, treatment bed: 22°; gantry: 300°, treatment bed: 0°; gantry: 300°, treatment bed: 22°.

[0042] By obtaining the field center and the tungsten ball position, the field center and tungsten ball center on the image returned by the image plate corresponding to each field are obtained, and their deviations are calculated. According to the gantry angle, the treatment bed angle, and the deviation between the two centers, the three-dimensional deviation between the mechanical center and the field isocenter calculated by the WinstonLutz method is judged whether it meets the threshold.

[0043] 3. When detecting the deviation between the gantry rotation center and the isocenter, the pre-made plan includes 6 fields, and the gantry angle position of each field is different, namely 0°; 45°; 90°; 200°; 245°; 290°.

[0044] By obtaining the field center and the tungsten ball position, the field center and the tungsten ball center on the image returned by the image board corresponding to each field are obtained, and their deviations are calculated. According to the gantry angle and the deviation of the two centers, the three-dimensional deviation of the deviation between the gantry rotation center and the isocenter is calculated by the WinstonLutz method to determine whether it meets the threshold.

[0045] 4. When detecting the deviation between the rotation center of the treatment bed and the isocenter, the pre-made plan includes 5 radiation fields, and the angle position of the treatment bed in each radiation field is different, namely 0°; 10°; 20°; 350°; 340°.

[0046] By obtaining the position of the tungsten ball, the center of the tungsten ball on the image returned by the image board corresponding to each field is obtained. A square area with a side length of 3mm (can be modified according to actual conditions) is taken as the center of the image center. Each pixel in this area is taken as the center point, and the distance between this center point and the center of different tungsten balls is calculated. The variance of all 5 distances is calculated to obtain the center point with the smallest calculated variance. At this time, the maximum value of the distance between the center point and the center of the tungsten ball is the deviation, and it is judged whether it meets the threshold.

[0047] 5. When calculating the grating accuracy, the pre-made plan includes 7 fields, and the grating center position of each field is different, namely LeafA = -70mm, LeafB = -50mm; LeafA = -50mm, LeafB = -30mm;

[0048] LeafA=-30mm, LeafB=-10mm; LeafA=-10mm, LeafB=10mm;

[0049] LeafA=10mm, LeafB=30mm; LeafA=30mm, LeafB=50mm;

[0050] LeafA=50mm, LeafB=70mm.

[0051] By obtaining the field boundary method, the left and right (or upper and lower) boundary positions of the field (the field is a square field) can be obtained, and the deviation between the theoretical leaf position and the obtained boundary position can be calculated to determine whether it meets the threshold

[0052] 6. When calculating the repeatability of the grating in place, the prefabricated plan contains 12 fields, with a total of 4 different grating isocenter positions, namely LeafA = -70mm, LeafB = -50mm; LeafA = -30mm, LeafB = -10mm;

[0053] LeafA=10mm, LeafB=30mm; LeafA=50mm, LeafB=70mm, execute 3 times for each position.

[0054] By obtaining the field boundary method, the left and right (or upper and lower) boundary positions of the field (the field is a square field) can be obtained, and the standard deviation of the boundaries obtained three times for the grating at the same theoretical position can be calculated to determine whether it meets the threshold

[0055] 7. Method for obtaining the field boundary: Sort the grayscale values ​​of all pixels on the image from large to small, calculate the average grayscale value of the first 10% (can be modified according to actual conditions) pixels, and binarize the image with 50% of the average value (can be modified according to actual conditions) as the threshold. The boundary of the binary bright area is the field boundary.

[0056] Method for obtaining the field center: Calculate the geometric center of the field area defined by the field boundary, which is the field center

[0057] Method for obtaining the position of the tungsten ball: Take the center of the field as the center of the circle and 3mm (can be modified according to actual conditions) as the diameter of the area. Take each pixel point in this area as the center of the circle and 3mm as the radius, calculate the variance of the grayscale of all pixels in the new circular area, and the center position with the smallest variance is the center of the tungsten ball.

[0058] 8. The dose detection module in the phantom contains a 30cmX30cmX30cm water cube (density equal to that of a water cube), and a dose detector (for receiving radiation) inserted into the center of the cube, and connected to the dose monitoring display outside the machine room via a connecting line (which can reflect the radiation dose received by the detector in real time). When using this module, place the phantom in the same way as the tungsten ball center point module, and then import the corresponding prefabricated plan execution file into the accelerator system according to the different test items (which can detect dose repeatability and dose linearity). The accelerator system will automatically position the dose detection module by moving the treatment bed, move the dose detector to the isocenter position, and then close the image board, automatically control the frame to move to 0°, the grating to move to 10cmX10cm open field, and automatically control the accelerator beam output.

[0059] 9. When testing dose repeatability, the pre-made plan contains 10 radiation fields, each of which is the same, and the beam dose is 10MU. The dose accumulation during the beam process will be recorded by the dose detector, and the actual final accumulation of each radiation field will be returned. The method to determine the dose repeatability is to calculate its coefficient of variation to determine whether it meets the threshold requirement. Coefficient of variation = standard deviation of measured dose / mean of measured dose*100%.

[0060] 10. When testing dose linearity, the pre-made plan contains 25 radiation fields, with a total of 5 different beam doses of 100MU; 300MU; 500MU; 700MU; 900MU, and each dose is executed 5 times. The dose accumulation of the beam process will be recorded by the dose detector, and the actual final accumulation of each radiation field will be returned. The method to determine the dose repeatability is to calculate its deviation and determine whether it meets the threshold. First calculate the average value of the measured dose of the radiation field with the same beam dose 5 times, with the beam dose as X and the average value as Y, and use the least squares method to calculate the fitting line. Substitute the 5 beam doses into the fitting line to obtain 5 theoretical doses. Deviation = ABS (measured dose - theoretical dose) / beam dose * 100%.

[0061] 11. The image analysis module of the phantom contains a set of (28) Las Vegas holes (with different combinations of apertures and depths) and different (12) densities of line pairs. When using this module, place the phantom in the same way as the tungsten ball center point module, and then import the corresponding prefabricated plan execution file into the accelerator system according to the different detection items (image contrast and image resolution can be detected). The accelerator system will automatically realize the positioning of the image analysis module by moving the treatment bed, so that the center position of the image analysis module moves to the isocenter, and open the image board to the image acquisition position, and then automatically control the gantry to move to 0°, the grating to 26cmX26cm open field, and automatically control the accelerator to emit beams. By collecting images on the image board, images containing Las Vegas holes and line pairs will be obtained. The number of holes and line pair density that are considered clear can be identified through algorithm analysis, and it can be determined whether the image contrast and resolution meet the requirements.

[0062] 12. When testing the contrast of an image, the coverage area of ​​each hole on the image will be obtained based on the relative physical position and aperture of the module center (i.e., image center) and each hole, and the average grayscale value in the coverage area of ​​each hole will be calculated, and the average grayscale value of the square coverage area circumscribed by each hole will be calculated. The ratio of the two will be calculated to determine whether it meets the threshold requirement.

[0063] 13. When detecting the resolution of an image, the coverage area of ​​each line pair on the image is obtained based on the relative physical position between the center of the module (i.e. the center of the image) and each line pair area and the size of the line pair area. The standard deviation of the grayscale in each area is calculated to determine whether it meets the threshold requirement.

[0064] When testing different test items, it is necessary to import the corresponding pre-made plan. After the operator places the phantom and imports the plan of the test item into the accelerator system, the accelerator will automatically open or close the imaging board and automatically execute the corresponding plan to obtain the corresponding image or monitor the beam dose. Then, based on the obtained image or measured dose, the corresponding algorithm is used to calculate whether the result meets the threshold. If the test item includes the use of different modules, the switching between different modules of the phantom will be achieved through the automatic control movement of the bed board.

[0065] The present invention achieves the following beneficial technical effects by designing the above-mentioned integrated model and designing an automated execution process:

[0066] The present invention summarizes and integrates the detection items required for the accelerator, and integrates the auxiliary equipment that may be needed in the detection process into a single phantom, thereby reducing the time for replacing the auxiliary equipment in the original detection process.

[0067] The execution plans for different detection projects are pre-built in, and the entire execution process is completely realized by the accelerator's automatic execution plan, reducing the time originally required for manual control of the accelerator's movement and beam emission.

[0068] The switching of modules within the phantom is also achieved by planning and controlling the bed to move to different positions. After the entire detection execution process begins, the operator no longer needs to enter and exit the accelerator room, shortening the execution time.

[0069] The above is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several modifications and improvements without departing from the inventive concept of the present invention, which all fall within the protection scope of the present invention.

Claims

1. An automated QA method for an accelerator system based on a combined motif, characterized in that: The combined phantom includes a tungsten ball center point module, a dose monitoring module and an image analysis module integrated into one body, wherein when the combined phantom moves as a whole, the relative positions of the three modules do not change, and are physically defined fixed relative positions, so that the switching of each module is achieved by controlling the treatment bed to move to different positions. The method includes the following steps: The tungsten ball center point module includes a tungsten ball mounting seat, a connecting rod in which a tungsten ball is embedded, and a three-dimensional manual fine-tuning slide. The objects detected by the tungsten ball center point module include at least one of the following: the deviation between the mechanical center and the isocenter of the field, the deviation between the gantry rotation center and the isocenter, the deviation between the treatment bed rotation center and the isocenter, the positioning accuracy of the grating, and the positioning repeatability of the grating. When using the tungsten ball center point module, the tungsten ball center point module is placed at the theoretical isocenter position, and the mechanical parts of the accelerator are controlled to move to a specific position by importing the prefabricated plan of the accelerator, and an image is obtained through an imaging plate while the accelerator emits a beam, and the corresponding mechanical deviation is calculated by obtaining the field center, the tungsten ball position, and the field boundary. The dose monitoring module includes a three-dimensional water phantom and a dose detector inserted into the center of the three-dimensional water phantom. The detection items of the dose monitoring module include one of dose repeatability and dose linearity. When the dose monitoring module is used, the corresponding prefabricated plan is imported into the accelerator system according to the detection items. The accelerator system automatically realizes the positioning of the dose detection module by moving the treatment bed, and automatically controls the accelerator beam output, and monitors the change of the dose rate over time and the final beam output dose during the accelerator beam output process; The detection items of the image analysis module include detecting one of image contrast and detecting image resolution. The image analysis module contains Las Vegas holes and line pairs representing different resolutions. When using the image analysis module, the corresponding prefabricated plan is imported into the accelerator system according to the detection items. The accelerator system automatically realizes the positioning of the image analysis module by moving the treatment bed, and then obtains the image containing the hole or the line pair by controlling the accelerator to emit beams and collecting images through the image board.

2. The method according to claim 1, characterized in that When detecting the deviation between the mechanical center and the field isocenter, the prefabricated plan included 8 fields, and the angle positions of the gantry and the treatment bed of each field were different, namely, gantry: 60°, treatment bed: 0°, gantry: 60°, treatment bed: 338°, gantry: 130°, treatment bed: 0°, gantry: 130°, treatment bed: 338°, gantry: 230°, treatment bed: 0°, gantry: 230°, treatment bed: 22°, gantry: 300°, treatment bed: 0°, gantry: 300°, treatment bed: 22°; Among them, the method of obtaining the field center and the method of obtaining the tungsten ball position are used to obtain the field center and tungsten ball center on the image returned by the image plate corresponding to each field, and the corresponding deviations are calculated. According to the gantry angle, the treatment bed angle, and the deviation between the two centers, the three-dimensional deviation between the mechanical center and the field isocenter is calculated by the WinstonLutz method to determine whether the threshold is met.

3. The method according to claim 1, characterized in that When detecting the deviation between the gantry rotation center and the isocenter, the pre-made plan includes 6 fields, each with a different gantry angle position, namely: 0°, 45°, 90°, 200°, 245°, and 290°; Among them, the method of obtaining the field center and the method of obtaining the tungsten ball position are used to obtain the field center and the tungsten ball center on the image returned by the image plate corresponding to each field, and the corresponding deviations are calculated. According to the frame angle and the deviation of the two centers, the three-dimensional deviation of the deviation between the frame rotation center and the isocenter is calculated by the WinstonLutz method to determine whether the threshold is met.

4. The method according to claim 1, characterized in that: When detecting the deviation between the rotation center of the treatment bed and the isocenter, the pre-made plan includes 5 fields, and the angle position of the treatment bed in each field is different, namely: 0°, 10°, 20°, 350°, and 340°; By obtaining the position of the tungsten ball, the center of the tungsten ball on the image returned by the image plate corresponding to each field is obtained respectively. A square area with the image center as the center is taken, and each pixel in this area is taken as the center point. The distance between this center point and the center of different tungsten balls is calculated, and the variance of all 5 distances is calculated to obtain the center point with the minimum calculated variance. At this time, the maximum value of the distance between the center point and the center of the tungsten ball is the deviation, and it is judged whether the deviation meets the threshold.

5. The method according to claim 1, characterized in that When calculating the grating in-position accuracy, the prefabricated plan includes 7 fields, and the grating isocenter positions of each field are different, namely: LeafA=-70mm, LeafB=-50mm, LeafA=-50mm, LeafB=-30mm, LeafA=-30mm, LeafB=-10mm, LeafA=-10mm, LeafB=10mm, LeafA=10mm, LeafB=30mm, LeafA=30mm, LeafB=50mm, LeafA=50mm, LeafB=70mm; The left and right or upper and lower boundary positions of the field of view are obtained by obtaining the field of view boundary, the deviation between the theoretical blade position and the obtained boundary position is calculated, and it is determined whether the deviation meets the threshold.

6. The method according to claim 1, characterized in that When calculating the repeatability of the grating in place, the prefabricated plan contains 12 fields with a total of 4 different grating isocenter positions, namely: LeafA=-70mm, LeafB=-50mm, LeafA=-30mm, LeafB=-10mm, LeafA=10mm, LeafB=30mm, LeafA=50mm, LeafB=70mm, and each position is executed 3 times; The left and right or upper and lower boundary positions of the field are obtained by obtaining the field boundary, and the standard deviation of the boundaries obtained three times for the grating at the same theoretical position is calculated to determine whether the standard deviation meets the threshold.

7. The method according to claim 1, characterized in that When testing dose repeatability, the pre-made plan contains 10 radiation fields, each of which is the same, and the beam dose is 10MU. The dose accumulation during the beam process is recorded by the dose detector, and the actual final accumulation of each radiation field is returned. The method for determining dose repeatability is to calculate the coefficient of variation: coefficient of variation = standard deviation of measured dose / mean of measured dose*100%; When testing dose linearity, the pre-made plan contains 25 radiation fields with 5 different beam doses, namely: 100MU, 300MU, 500MU, 700MU, and 900MU. Each dose is executed 5 times. The dose accumulation of the beam process will be recorded by the dose detector, and the actual final accumulation of each radiation field will be returned. The method to determine the dose linearity is to calculate the deviation: deviation = ABS (measured dose - theoretical dose) / beam dose * 100%.

8. The method according to claim 1, characterized in that When detecting the image contrast, the coverage area of ​​each hole on the image is obtained according to the relative physical position and aperture of the center of the image analysis module and each hole, and the average grayscale value in the coverage area of ​​each hole is calculated respectively, and the average grayscale value of the square coverage area circumscribed by each hole is calculated, and then the ratio of the two is calculated to determine whether the threshold requirement is met; When detecting image resolution, the coverage area of ​​each line pair on the image is obtained based on the relative physical position between the center of the image analysis module and each line pair area and the size of the line pair area, and the standard deviation of the grayscale in each area is calculated to determine whether the threshold requirement is met.

Images