WO2022007312A1 - 一种基于epid的光栅位置标定和验证方法 - Google Patents

一种基于epid的光栅位置标定和验证方法 Download PDF

Info

Publication number
WO2022007312A1
WO2022007312A1 PCT/CN2020/131312 CN2020131312W WO2022007312A1 WO 2022007312 A1 WO2022007312 A1 WO 2022007312A1 CN 2020131312 W CN2020131312 W CN 2020131312W WO 2022007312 A1 WO2022007312 A1 WO 2022007312A1
Authority
WO
WIPO (PCT)
Prior art keywords
grating
epid
image
plane
calculate
Prior art date
Application number
PCT/CN2020/131312
Other languages
English (en)
French (fr)
Inventor
王忠淼
文虎儿
姚毅
Original Assignee
苏州雷泰医疗科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州雷泰医疗科技有限公司 filed Critical 苏州雷泰医疗科技有限公司
Publication of WO2022007312A1 publication Critical patent/WO2022007312A1/zh

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61NELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00Radiation therapy
    • A61N5/10X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/1048Monitoring, verifying, controlling systems and methods
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61NELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00Radiation therapy
    • A61N5/10X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/1042X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy with spatial modulation of the radiation beam within the treatment head
    • A61N5/1045X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy with spatial modulation of the radiation beam within the treatment head using a multi-leaf collimator, e.g. for intensity modulated radiation therapy or IMRT
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61NELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00Radiation therapy
    • A61N5/10X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/1048Monitoring, verifying, controlling systems and methods
    • A61N5/1049Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61NELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00Radiation therapy
    • A61N5/10X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/1048Monitoring, verifying, controlling systems and methods
    • A61N5/1049Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam
    • A61N2005/1054Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam using a portal imaging system

Definitions

  • the invention belongs to the field of medical technology, and in particular relates to an EPID-based grating position calibration and verification method.
  • Grating is an indispensable collimating device in modern radiotherapy equipment. Through the movement of the grating blades, it can have a very good conformal effect on the treatment target area; the multi-leaf collimator (MLC) is a kind of clinical radiotherapy.
  • MLC multi-leaf collimator
  • TPS treatment planning system
  • the traditional method is to lay a graph paper on the isocenter plane, irradiate the MLC with a light field, and obtain the position of the MLC on the isocenter plane by reading the projected position of the MLC on the graph paper. Error or deviation is large.
  • a commonly used method is to use an electronic portal imaging system (EPID) to record the image of the ray in the isocenter plane after passing through the MLC, and then use the image processing technology to detect the position of the MLC.
  • EPID electronic portal imaging system
  • the commonly used method to determine the coordinate origin of the isocenter plane is to confirm the laser light through the front pointer on the accelerator head, and the laser light confirms a position of the isocenter plane (O 3 as shown in Figure 1), and then the tungsten ball It is placed in this position, and the EPID is used to record the image at this time when the accelerator is out of the beam.
  • the image processing technology is used to extract the center of the tungsten sphere, which is considered to be the coordinate origin of the isocenter plane.
  • the edge of the tungsten sphere image recorded by EPID is blurred, and the center of the tungsten sphere extracted by image processing technology is not accurate enough.
  • the projection of the origin of the coordinates of the MLC on the isocenter plane should coincide with the center of the tungsten sphere.
  • a common method is to open a pair of left and right blades at the center of the grating, and let the left and right blades at other positions There is also a relatively small interval, forming a cross-shaped field, as shown in Figure 6. Keep the position of the tungsten ball still, and use EPID to record the image at this time when the accelerator is out of the beam, as shown in Figure 7.
  • the center of the tungsten sphere extracted by the image processing technology is required to be accurate enough, and after the grating opens the cross-shaped field, the origin of the grating coordinates and the calculated center of the tungsten sphere are determined by visual observation. Whether it is coincident or not, this method is not accurate and objective enough.
  • the present invention provides a new EPID-based grating position calibration and verification method.
  • An EPID-based grating position calibration and verification method which specifically includes the following steps:
  • S1.2 Use image processing technology to extract the two boundary positions of the upper grating blade and the lower grating blade along the moving direction, and calculate the average value of the two boundary positions to obtain the coordinate origin of the isocenter plane;
  • step S1.1 As a preferred solution, in step S1.1,
  • the blade at the center of the upper grating or the lower grating is only one blade
  • the number of blades of the upper grating or the lower grating is an even number, the blades at the center of the upper grating or the lower grating are two adjacent blades.
  • step S2 specifically includes the following content:
  • SCD 1 is the distance from the radiation source to the center of the upper grating
  • SCD 2 is the distance from the radiation source to the center of the lower grating.
  • the detection method also includes:
  • step S5 the function is obtained by one of the following methods: using polynomial fitting to obtain the correlation coefficient, or using a piecewise linear interpolation method.
  • the detection method also includes:
  • the relational expression is considered unreliable, and steps S1-S5 need to be repeated; or, the relational expression between the physical position x of the MLC on the grating plane and the position y of the isocenter plane is re-established.
  • the image processing technology is: by searching for the position with the maximum slope in each row and/or column on the EPID image, and then calculating the average value of all the positions with the maximum slope to obtain the boundary position of the grating.
  • the image processing technology in step S1.2 specifically includes the following contents:
  • step S4 specifically includes the following steps:
  • SAD is the distance from the radiation source to the isocenter plane.
  • the image processing technology also includes the first step: performing smooth filtering and thresholding on the image, and setting all pixels in the image whose pixel value is greater than k (0.5 ⁇ k ⁇ 1) times the maximum pixel value as the maximum pixel value. k times.
  • the invention provides a new grating position calibration and verification method based on EPID, which can effectively solve the problem of position detection of double-layer gratings in the isocenter plane, is easy to operate, and can improve the detection accuracy, which is extremely important for clinical practice. significance.
  • Fig. 1 is the spatial schematic diagram and coordinate system definition of grating installation
  • Fig. 2 is an example diagram after the ray recorded by EPID passes through the grating
  • Fig. 3 is the horizontal section line of the example diagram after the ray recorded by EPID passes through the grating
  • Fig. 5 is the algorithm flow chart of the image processing technique that detects the position of MLC proposed by the embodiment of the present invention.
  • FIG. 6 is a schematic diagram of the position of the MLC cruciform blade
  • Fig. 7 is the EPID image of MLC opening the cross-shaped field and irradiating the tungsten ball
  • FIG. 8 is an example diagram of a grating position detected on an EPID image proposed by an embodiment of the present invention (white lines are the detected grating positions);
  • FIG. 9 is a schematic diagram of the position of the double-layer grating cross-shaped blade according to an embodiment of the present invention.
  • the equipment based on the present invention is an electronic portal imaging system (EPID) and an orthogonal double-layer grating.
  • EPID can measure the ray intensity emitted from the radiation source and penetrate the patient in radiotherapy, and convert it into an electrical signal to form a digital image. It is an important tool to verify the accuracy of radiotherapy setup; the orthogonal double-layer grating is divided into upper and lower layers, and the motion directions of the two layers of gratings are orthogonal to each other and perpendicular to the ray direction.
  • an embodiment of the present invention provides an EPID-based grating position calibration and verification method, which specifically includes the following steps:
  • S1.2 Use image processing technology to extract the two boundary positions of the upper grating blade and the lower grating blade along the moving direction, and calculate the average value of the two boundary positions to obtain the coordinate origin of the isocenter plane;
  • the present invention provides an EPID-based grating position calibration and verification method, which is simpler and more accurate compared to the method of using a tungsten ball to determine the coordinate origin of the isocenter plane mentioned in the background art.
  • step S1.1 the difference is that in step S1.1,
  • the blade at the center of the upper grating or the lower grating is only one blade
  • the number of blades of the upper grating or the lower grating is an even number, the blades at the center of the upper grating or the lower grating are two adjacent blades.
  • step S2 specifically includes the following content:
  • SCD 1 is the distance from the radiation source to the center of the upper grating
  • SCD 2 is the distance from the radiation source to the center of the lower grating.
  • step S2 the existing method can be used to obtain the installation height of the EPID, and the method of the present invention can also be used more optimally.
  • the current common method is to place a phantom with a known width d ISO on the isocenter plane, and then let the accelerator exit the beam and use EPID to collect images, and obtain the width d EPID of the phantom through image processing technology. Use the following formula to calculate the installation height of the EPID
  • the detection method further includes:
  • step S5 the function is obtained by one of the following methods: using polynomial fitting to obtain the correlation coefficient, or using a piecewise linear interpolation method.
  • the detection method also includes:
  • the relational expression is considered unreliable, and steps S1-S5 need to be repeated; or, the relational expression between the physical position x of the MLC on the grating plane and the position y of the isocenter plane is re-established.
  • the remaining feature technologies are the same, the difference is that the above-mentioned image processing technology is: by searching the position of the maximum slope of each row and/or each column on the EPID image , and then averaged over all the locations of the largest slopes to obtain the boundary locations of the raster.
  • step S1.2 specifically includes the following contents:
  • step S4 specifically includes the following steps:
  • SAD is the distance from the radiation source to the isocenter plane.
  • the image processing technology in the above-mentioned steps S2, S4, and S6 also includes the first step: smooth filtering and threshold processing are performed on the image, and all pixel values in the image are greater than k (0.5 ⁇ k ⁇ 1) Pixels that times the maximum pixel value are set to k times the maximum value.
  • the threshold processing step since the pixel value of the EPID image reflects the ray intensity it receives, all the pixels in the image whose pixel value is greater than k (0.5 ⁇ k ⁇ 1) times the maximum pixel value are set to k times the maximum value, Often the boundary locations of MLCs are not in these high-intensity regions.
  • FIG. 8 it is an example diagram of the grating position detected on the EPID image using the image processing technology of the present invention, and the white line is the detected grating position.
  • steps S1, S4 and S6 of the present application can also be processed in the existing way to extract the boundary of the motif, but the existing detection MLC position will be analyzed in combination with the characteristics of the EPID image below.
  • the shortcomings of the image processing technology can also be processed in the existing way to extract the boundary of the motif, but the existing detection MLC position will be analyzed in combination with the characteristics of the EPID image below.
  • the dark area in Figure 2 is the MLC blocking area, and the bright area is the area that is not blocked by MLC. Due to the influence of the penumbra, there is also a light-dark transition area; the section line shown in Figure 3 can be more intuitive. Reflects the light intensity relationship in different regions.
  • a commonly used image processing technology for detecting the position of MLC adopts the threshold method, that is, the area with the intensity lower than a certain threshold (generally the average value of the light intensity) is considered to be the area blocked by the MLC, and the area greater than or equal to the threshold is The non-MLC blocks the area, and the junction of the two areas is judged as the boundary position of the MLC on the isocenter plane.
  • a certain threshold generally the average value of the light intensity
  • Threshold-based boundary detection technology different thresholds will detect different positions; and the noise in the image will also interfere with the selection of the threshold.
  • the invention provides a new grating position calibration and verification method based on EPID, the core of the method is to make full use of the characteristics of the orthogonal double-layer grating to determine the coordinate origin of the isocenter plane and the installation height of the EPID, and the operation is relatively simple; and By searching for the position of the maximum slope in each row or column on the EPID image, and then calculating the average value of all the positions with the maximum slope, the boundary position of the raster is obtained.
  • the invention also establishes the relational expression between the physical position of the grating and the grating position of the isocenter plane, and proposes an experimental scheme to verify the rationality of the established relational expression.
  • the invention provides a new grating position calibration and verification method based on EPID, which can effectively solve the problem of position detection of double-layer gratings in the isocenter plane, is easy to operate, and can improve the detection accuracy, which is extremely important for clinical practice. significance.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Pathology (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

一种基于EPID的光栅位置标定和验证方法,包括以下步骤:S1:确定光栅平面的物理坐标系原点在等中心平面的投影位置;S2:确定EPID的安装高度:S3:让光栅平面的上层光栅和下层光栅运动到n个不同位置,每个位置均用EPID记录一幅图像;S4:采用图像处理技术检测MLC在等中心平面的位置。利用正交双层光栅的特性来确定等中心平面的坐标原点,操作较为简便,且能够提升检测精度,对于临床有着极其重要的意义。

Description

一种基于EPID的光栅位置标定和验证方法 技术领域
本发明属于医疗技术领域,具体涉及一种基于EPID的光栅位置标定和验证方法。
背景技术
光栅是现代放疗设备中所必不可少的一种准直设备,通过光栅叶片的运动,可以对治疗靶区有着非常好的适形效果;多叶准直器(MLC)是一种放疗临床中常用的光栅,一般情况下MLC的运动控制越精确,越能接近治疗计划系统(TPS)计算的治疗效果。由于TPS设计的计划中光栅运动位置均在等中心平面(ISO),而MLC是安装在等中心平面之上,如图1所示。故需要将MLC在光栅平面的物理位置与等中心平面的位置建立对应关系,这其中就需要检测MLC在等中心平面的边界位置。
传统的方法是在等中心平面铺设一张坐标纸,用光野照射MLC,通过读取MLC投影在坐标纸上的位置来获得MLC在等中心平面的位置,该方法不仅操作繁琐而且人工肉眼读数易出错或偏差较大。目前 一种常用的方法是采用电子射野影像系统(EPID)记录射线经过MLC后在等中心平面的图像,然后通过图像处理技术来检测MLC的位置。下面介绍该方案的一些细节并分析现有方法的不足之处:
为了建立MLC平面的物理坐标系和等中心(ISO)平面的坐标系之间的关系,如图1所示。首先需要使得MLC坐标原点在等中心平面的投影位置与等中心平面的坐标原点重合。目前常用的确定等中心平面的坐标原点的方法是通过加速器机头上的前指针确认激光灯,激光灯确认等中心平面的一个位置(如图1所示中的O 3),然后将钨球摆放在该位置,并在加速器出束的情况下利用EPID记录下此时的图像,最后运用图像处理技术来提取钨球的中心,认为该中心正是等中心平面的坐标原点。而由于半影的影响,EPID记录的钨球图像边缘较模糊,再通过图像处理技术提取的钨球中心是不够准确的。
此外,还要使得MLC的坐标原点在等中心平面的投影与钨球中心重合,目前一种常用的方法是将光栅中心处的一对左、右叶片打开,并让其它位置的左、右叶片也有一个相对较小的间隔,形成一个十字形射野,如图6所示。保持钨球位置不动,在加速器出束的情况下利用EPID记录下此时的图像,如图7所示。然后观察钨球中心是否在十字形的中心,如果不在则调节MLC位置,直到钨球中心与十字中 心重合,这样就保证了MLC的物理坐标系原点与等中心平面的坐标原点对应上了。
综上,以上方案会面临如下问题;
在确定MLC坐标原点在等中心平面的投影位置时,要求通过图像处理技术提取的钨球中心足够准确,且光栅打开十字形射野后,是通过肉眼观察确定光栅坐标原点与计算的钨球中心是否重合,因此该方法不够准确客观。
发明内容
为了解决上述现有的基于光野和基于EPID的光栅位置标定和验证方法的检测结果不够准确客观,且操作较为繁琐的技术问题,本发明提供一种新的基于EPID的光栅位置标定和验证方法。
为了达到上述目的,本发明的技术方案如下:
一种基于EPID的光栅位置标定和验证方法,具体包括以下步骤:
S1:确定光栅平面的物理坐标系原点在等中心平面的投影位置,具体包括以下步骤:
S1.1:仅伸出上层光栅和下层光栅中心处的叶片,并运动超过中心位置一段距离,形成十字形;在加速器出束的情况下,利用EPID 记录下此时的图像;
S1.2:运用图像处理技术提取上层光栅叶片和下层光栅叶片各自沿运动方向的两个边界位置,并计算这两个边界位置的平均值,即可得到等中心平面的坐标原点;
S2:确定EPID的安装高度;
S3:让光栅平面的上层光栅和下层光栅运动到n个不同的位置记为x=[x 1,x 2,...x n] T,并在加速器出束的情况下,每一个位置均用EPID记录一幅图像,记采集的图像组为f 1,f 2,...f n
S4:对每一幅EPID记录的图像f i,i=1,2,...n,采用图像处理技术检测MLC在等中心平面的位置,得到y=[y 1,y 2,...y n] T
在上述技术方案的基础上,还可做如下改进:
作为优选的方案,在步骤S1.1中,
若上层光栅或下层光栅的叶片数量为奇数时,则上层光栅或下层光栅中心处的叶片仅为一个叶片;
若上层光栅或下层光栅的叶片数量为偶数时,则上层光栅或下层光栅中心处的叶片则为相邻的两个叶片。
作为优选的方案,步骤S2具体包括以下内容:
利用步骤S1得到的上层光栅沿运动方向的两个边界位置计算出 叶片在EPID图像上的宽度
Figure PCTCN2020131312-appb-000001
记上层光栅栅中心叶片实际物理宽度为
Figure PCTCN2020131312-appb-000002
通过下式计算出EPID的安装高度SDD:
Figure PCTCN2020131312-appb-000003
其中:SCD 1为从放射源到上层光栅中心的距离;
或,
利用步骤S1得到的下层光栅沿运动方向的两个边界位置计算出叶片在EPID图像上的宽度
Figure PCTCN2020131312-appb-000004
记下层光栅中心叶片实际物理宽度为
Figure PCTCN2020131312-appb-000005
通过下式计算出EPID的安装高度SDD:
Figure PCTCN2020131312-appb-000006
其中:SCD 2为从放射源到下层光栅中心的距离。
作为优选的方案,检测方法还包括:
S5:对x=[x 1,x 2,...x n] T和y=[y 1,y 2,...y n] T,建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式,记为函数x=g(y)。
作为优选的方案,步骤S5中,通过以下一种方式得到函数:采 用多项式拟合得到相关系数,或采用分段线性插值的方法。
作为优选的方案,检测方法还包括:
S6:验证建立的关系式的合理性,具体包括以下步骤:
S6.1:给定一组等中心平面的位置为y v=[y 1,y 2,...y s] T,通过步骤S5得到的函数计算出对应的光栅物理运动位置x v=[x 1,x 2,...x s] T
S6.2:让光栅分别运动到x v,并在加速器出束的情况下,用EPID记录下光栅在不同位置时的图像,采用图像处理技术对这些图像检测出光栅在等中心平面的位置y measure=[y 1,y 2,...y s] T
S6.3:比较y v和y measure,若两者误差在合理范围内,则认为检测的位置、建立的光栅平面的物理位置与等中心平面的位置关系式是可靠的;
若两者误差不在合理范围内,则认为其关系式不可靠,需重新重复步骤S1-S5;或,重新建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式。
作为优选的方案,图像处理技术为:通过搜索EPID图像上每一行和/每一列的斜率最大的位置,然后对所有最大斜率的位置计算平均值,得到光栅的边界位置。
作为优选的方案,步骤S1.2中的图像处理技术具体包括以下内容:
(A)对步骤S1.1得到的图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m 1,m 1为图像的总行数;
(B)对所有的r k计算平均值row c,即:
Figure PCTCN2020131312-appb-000007
(C)对步骤S1.1得到的图像上的每一列计算斜率,并找出斜率最大的位置c k;其中,k=1,2,...m 2,m 2为图像的总列数;
(D)对所有的c k计算平均值col c,即:
Figure PCTCN2020131312-appb-000008
(E)得到等中心平面的坐标原点(row c,col c)。
作为优选的方案,步骤S4具体包括以下步骤:
(a)对每幅图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m,m为图像的总行数;
(b)对所有的r k计算平均值row i,即:
Figure PCTCN2020131312-appb-000009
(d)记EPID的像素尺寸为pd×pd大小、根据中心位置(row c,col c),可按如下公式计算得到MLC在等中心平面的位置;
Figure PCTCN2020131312-appb-000010
其中:SAD为从放射源到等中心平面的距离。
作为优选的方案,图像处理技术还包括首步内容:对图像进行光滑滤波及阈值处理,将图像中所有像素值大于k(0.5<k<1)倍最大像素值的像素均设置为最大值的k倍。
本发明具有以下有益效果:
本发明提供一种新的基于EPID的光栅位置标定和验证方法,其能够有效地解决双层光栅在等中心平面的位置检测问题,操作较为简便,且能够提升检测精度,对于临床有着极其重要的意义。
附图说明
为了更清楚地说明本发明实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,应当理解,以下附图仅示出了本发明的某些实施例,因此不应被看作是对范围的限定,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他相关的附图。
图1为光栅安装的空间示意图及坐标系定义;
图2为EPID记录的射线经过光栅后的示例图;
图3为EPID记录的射线经过光栅后的示例图的水平剖面线;
图4为本发明实施例提出的基于EPID的光栅位置标定和验证方法的流程图;
图5为本发明实施例提出的检测MLC的位置的图像处理技术的 算法流程图;
图6为MLC十字形叶片位置示意图;
图7为MLC开十字形射野并照射钨球的EPID图像;
图8为本发明实施例提出的在EPID图像上检测出的光栅位置示例图(白线为检测出的光栅位置);
图9为本发明实施例提出的双层光栅十字形叶片位置示意图。
具体实施方式
下面结合附图详细说明本发明的优选实施方式。
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
本发明基于的设备为电子射野影像系统(EPID)和正交双层光栅,EPID可以测量放射治疗中从辐射源发出并穿透患者后的射线强度,将其转换为电信号形成数字图像,是一种验证放射治疗摆位准确性的重要工具;而正交双层光栅分为上下两层,这两层光栅的运动方向相互正交且垂直于射线方向。
如图4所示,本发明实施例提供一种基于EPID的光栅位置标定和验证方法,具体包括以下步骤:
S1:确定光栅平面的物理坐标系原点在等中心平面的投影位置,具体包括以下步骤:
S1.1:仅伸出上层光栅和下层光栅中心处的叶片,并运动超过中心位置一段距离,形成十字形,如图9所示;在加速器出束的情况下,利用EPID记录下此时的图像;
S1.2:运用图像处理技术提取上层光栅叶片和下层光栅叶片各自沿运动方向的两个边界位置,并计算这两个边界位置的平均值,即可得到等中心平面的坐标原点;
S2:确定EPID的安装高度;
S3:让光栅平面的上层光栅和下层光栅运动到n个不同的位置记为x=[x 1,x 2,...x n] T,并在加速器出束的情况下,每一个位置均用EPID记录一幅图像,记采集的图像组为f 1,f 2,...f n
S4:对每一幅EPID记录的图像f i,i=1,2,...n,采用图像处理技术检测MLC在等中心平面的位置,得到y=[y 1,y 2,...y n] T
本发明提供一种基于EPID的光栅位置标定和验证方法,相较于背景技术内提到的利用钨球来确定等中心平面的坐标原点的方法,本方法更简单,且更准确。
为了进一步地优化本发明的实施效果,在另外一些实施方式中,其余特征技术相同,不同之处在于,在步骤S1.1中,
若上层光栅或下层光栅的叶片数量为奇数时,则上层光栅或下层 光栅中心处的叶片仅为一个叶片;
若上层光栅或下层光栅的叶片数量为偶数时,则上层光栅或下层光栅中心处的叶片则为相邻的两个叶片。
为了进一步地优化本发明的实施效果,在另外一些实施方式中,其余特征技术相同,不同之处在于,步骤S2具体包括以下内容:
利用步骤S1得到的上层光栅沿运动方向的两个边界位置计算出叶片在EPID图像上的宽度
Figure PCTCN2020131312-appb-000011
记上层光栅栅中心叶片实际物理宽度为
Figure PCTCN2020131312-appb-000012
通过下式计算出EPID的安装高度SDD:
Figure PCTCN2020131312-appb-000013
其中:SCD 1为从放射源到上层光栅中心的距离;
或,
利用步骤S1得到的下层光栅沿运动方向的两个边界位置计算出叶片在EPID图像上的宽度
Figure PCTCN2020131312-appb-000014
记下层光栅中心叶片实际物理宽度为
Figure PCTCN2020131312-appb-000015
通过下式计算出EPID的安装高度SDD:
Figure PCTCN2020131312-appb-000016
其中:SCD 2为从放射源到下层光栅中心的距离。
值得注意的是,步骤S2可采用现有方式进行获取EPID的安装高度,也可以更优的采用本发明的方法。
但是,目前现有的常用方法是在等中心平面摆放一个已知宽度d ISO的模体,然后让加速器出束并利用EPID采集图像,通过图像处理技术来获取该模体的宽度d EPID,利用下式计算EPID的安装高度
Figure PCTCN2020131312-appb-000017
其存在以下问题:在确定EPID的安装高度时,需要额外摆放模体。而本申请提出的确定EPID的安装高度的方法无需额外摆放模体,更简单。
为了进一步地优化本发明的实施效果,在另外一些实施方式中,其余特征技术相同,不同之处在于,检测方法还包括:
S5:对x=[x 1,x 2,...x n] T和y=[y 1,y 2,...y n] T,建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式,记为函数x=g(y)。
进一步,步骤S5中,通过以下一种方式得到函数:采用多项式拟合得到相关系数,或采用分段线性插值的方法。
进一步,检测方法还包括:
S6:验证建立的关系式的合理性,具体包括以下步骤:
S6.1:给定一组等中心平面的位置为y v=[y 1,y 2,...y s] T,通过步骤S5得到的函数计算出对应的光栅物理运动位置x v=[x 1,x 2,...x s] T
S6.2:让光栅分别运动到x v,并在加速器出束的情况下,用EPID记录下光栅在不同位置时的图像,采用图像处理技术对这些图像检测出光栅在等中心平面的位置y measure=[y 1,y 2,...y s] T
S6.3:比较y v和y measure,若两者误差在合理范围内,则认为检测的位置、建立的光栅平面的物理位置与等中心平面的位置关系式是可靠的;
若两者误差不在合理范围内,则认为其关系式不可靠,需重新重复步骤S1-S5;或,重新建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式。
为了进一步地优化本发明的实施效果,在另外一些实施方式中,其余特征技术相同,不同之处在于,上述的图像处理技术为:通过搜索EPID图像上每一行和/每一列的斜率最大的位置,然后对所有最大斜率的位置计算平均值,得到光栅的边界位置。
进一步对上述实施例进行补充,步骤S1.2中的图像处理技术具体包括以下内容:
(A)对步骤S1.1得到的图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m 1,m 1为图像的总行数;
(B)对所有的r k计算平均值row c,即:
Figure PCTCN2020131312-appb-000018
(C)对步骤S1.1得到的图像上的每一列计算斜率,并找出斜率最大的位置c k;其中,k=1,2,...m 2,m 2为图像的总列数;
(D)对所有的c k计算平均值col c,即:
Figure PCTCN2020131312-appb-000019
(E)得到等中心平面的坐标原点(row c,col c)。
进一步对上述实施例进行补充,如图5所示,步骤S4具体包括以下步骤:
(a)对每幅图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m,m为图像的总行数;
(b)对所有的r k计算平均值row i,即:
Figure PCTCN2020131312-appb-000020
(d)记EPID的像素尺寸为pd×pd大小、根据中心位置 (row c,col c),可按如下公式计算得到MLC在等中心平面的位置;
Figure PCTCN2020131312-appb-000021
其中:SAD为从放射源到等中心平面的距离。
进一步对上述实施例进行补充,上述步骤S2、S4、S6中的图像处理技术还包括首步内容:对图像进行光滑滤波及阈值处理,将图像中所有像素值大于k(0.5<k<1)倍最大像素值的像素均设置为最大值的k倍。
阈值处理步骤中,由于EPID图像的像素值反映了其接收到的射线强度,将图像中所有像素值大于k(0.5<k<1)倍最大像素值的像素均设置为最大值的k倍,通常MLC的边界位置不会在这些高强度区域。如图8所示,其为采用本发明的图像处理技术在EPID图像上检测出的光栅位置示例图,白线为检测出的光栅位置。
值得注意的是,本申请步骤S1、S4、S6中应用的图像处理技术也可采用现有方式进行处理,来提取模体的边界,但是下面将结合EPID图像的特点分析现有的检测MLC位置的图像处理技术的不足之处。
如图2所示,图2中暗区为MLC阻挡区域,亮区是未被MLC遮挡区域,由于半影的影响还产生了明暗过渡的区域;如图3所示的 剖面线能更加直观地反映不同区域的光强关系。
目前一种常用的检测MLC位置的图像处理技术采用了阈值方法,即认为强度低于某一个阈值(一般取光强的平均值)的区域均为MLC所遮挡的区域,大于等于阈值的区域为非MLC遮挡区域,而将两者区域的交界处判断为MLC在等中心平面的边界位置。
基于阈值的边界检测技术,不同的阈值会检测出不同的位置;而且图像中的噪声也会干扰阈值的选取。
综上述,因现有的基于光野和基于EPID的光栅位置标定和验证方法的检测结果不够准确客观,且操作较为繁琐。本发明提供一种新的基于EPID的光栅位置标定和验证方法,该方法的核心是充分利用正交双层光栅的特性来确定等中心平面的坐标原点及EPID的安装高度,操作较为简便;并通过搜索EPID图像上每一行或每一列的斜率最大的位置,然后对所有最大斜率的位置计算平均值,得到光栅的边界位置。同时本发明还建立了光栅的物理位置和等中心平面的光栅位置的关系式,并提出了一种实验方案来验证建立的关系式的合理性。
本发明具有以下有益效果:
本发明提供一种新的基于EPID的光栅位置标定和验证方法,其能够有效地解决双层光栅在等中心平面的位置检测问题,操作较为简 便,且能够提升检测精度,对于临床有着极其重要的意义。
以上多种实施方式可交叉并行实现。
上述实施例只为说明本发明的技术构思及特点,其目的在于让本领域普通技术人员能够了解本发明的内容并加以实施,并不能以此限制本发明的保护范围,凡根据本发明精神实质所作的等效变化或修饰,都应涵盖在本发明的保护范围内。

Claims (10)

  1. 一种基于EPID的光栅位置标定和验证方法,其特征在于,具体包括以下步骤:
    S1:确定光栅平面的物理坐标系原点在等中心平面的投影位置,具体包括以下步骤:
    S1.1:仅伸出上层光栅和下层光栅中心处的叶片,并运动超过中心位置一段距离,形成十字形;在加速器出束的情况下,利用EPID记录下此时的图像;
    S1.2:运用图像处理技术提取上层光栅叶片和下层光栅叶片各自沿运动方向的两个边界位置,并计算这两个边界位置的平均值,即可得到等中心平面的坐标原点;
    S2:确定EPID的安装高度;
    S3:让光栅平面的上层光栅和下层光栅运动到n个不同的位置记为x=[x 1,x 2,...x n] T,并在加速器出束的情况下,每一个位置均用EPID记录一幅图像,记采集的图像组为f 1,f 2,...f n
    S4:对每一幅EPID记录的图像f i,i=1,2,...n,采用图像处理技术检测MLC在等中心平面的位置,得到y=[y 1,y 2,...y n] T
  2. 根据权利要求1所述的基于EPID的光栅位置标定和验证方法, 其特征在于,在所述步骤S1.1中,
    若上层光栅或下层光栅的叶片数量为奇数时,则上层光栅或下层光栅中心处的叶片仅为一个叶片;
    若上层光栅或下层光栅的叶片数量为偶数时,则上层光栅或下层光栅中心处的叶片则为相邻的两个叶片。
  3. 根据权利要求1所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述步骤S2具体包括以下内容:
    利用步骤S1得到的上层光栅沿运动方向的两个边界位置计算出叶片在EPID图像上的宽度
    Figure PCTCN2020131312-appb-100001
    记上层光栅栅中心叶片实际物理宽度为
    Figure PCTCN2020131312-appb-100002
    通过下式计算出EPID的安装高度SDD:
    Figure PCTCN2020131312-appb-100003
    其中:SCD 1为从放射源到上层光栅中心的距离;
    或,
    利用步骤S1得到的下层光栅沿运动方向的两个边界位置计算出叶片在EPID图像上的宽度
    Figure PCTCN2020131312-appb-100004
    记下层光栅中心叶片实际物理宽度为
    Figure PCTCN2020131312-appb-100005
    通过下式计算出EPID的安装高度SDD:
    Figure PCTCN2020131312-appb-100006
    其中:SCD 2为从放射源到下层光栅中心的距离。
  4. 根据权利要求1所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述检测方法还包括:
    S5:对x=[x 1,x 2,...x n] T和y=[y 1,y 2,...y n] T,建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式,记为函数x=g(y)。
  5. 根据权利要求4所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述步骤S5中,通过以下一种方式得到函数:采用多项式拟合得到相关系数,或采用分段线性插值的方法。
  6. 根据权利要求4所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述检测方法还包括:
    S6:验证建立的关系式的合理性,具体包括以下步骤:
    S6.1:给定一组等中心平面的位置为y v=[y 1,y 2,...y s] T,通过所述步骤S5得到的函数计算出对应的光栅物理运动位置x v=[x 1,x 2,...x s] T
    S6.2:让光栅分别运动到x v,并在加速器出束的情况下,用EPID记录下光栅在不同位置时的图像,采用图像处理技术对这些图像检测出光栅在等中心平面的位置y measure=[y 1,y 2,...y s] T
    S6.3:比较y v和y measure,若两者误差在合理范围内,则认为检测的 位置、建立的光栅平面的物理位置与等中心平面的位置关系式是可靠的;
    若两者误差不在合理范围内,则认为其关系式不可靠,需重新重复步骤S1-S5;
    或,重新建立MLC在光栅平面的物理位置x与等中心平面的位置y之间的关系式。
  7. 根据权利要求1-6任一项所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述图像处理技术为:通过搜索EPID图像上每一行和/每一列的斜率最大的位置,然后对所有最大斜率的位置计算平均值,得到光栅的边界位置。
  8. 根据权利要求7所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述步骤S1.2中的图像处理技术具体包括以下内容:
    (A)对所述步骤S1.1得到的图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m 1,m 1为图像的总行数;
    (B)对所有的r k计算平均值row c,即:
    Figure PCTCN2020131312-appb-100007
    (C)对所述步骤S1.1得到的图像上的每一列计算斜率,并找出斜率最大的位置c k;其中,k=1,2,...m 2,m 2为图像的总列数;
    (D)对所有的c k计算平均值col c,即:
    Figure PCTCN2020131312-appb-100008
    (E)得到等中心平面的坐标原点(row c,col c)。
  9. 根据权利要求7所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述步骤S4具体包括以下步骤:
    (a)对每幅图像上的每一行计算斜率,并找出斜率最大的位置r k;其中,k=1,2,...m,m为图像的总行数;
    (b)对所有的r k计算平均值row i,即:
    Figure PCTCN2020131312-appb-100009
    (d)记EPID的像素尺寸为pd×pd大小、根据中心位置(row c,col c),可按如下公式计算得到MLC在等中心平面的位置;
    Figure PCTCN2020131312-appb-100010
    其中:SAD为从放射源到等中心平面的距离。
  10. 根据权利要求7所述的基于EPID的光栅位置标定和验证方法,其特征在于,所述图像处理技术还包括首步内容:对图像进行光滑滤波及阈值处理,将图像中所有像素值大于k(0.5<k<1)倍最大像素值的像素均设置为最大值的k倍。
PCT/CN2020/131312 2020-07-10 2020-11-25 一种基于epid的光栅位置标定和验证方法 WO2022007312A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010660774.2A CN111773560B (zh) 2020-07-10 2020-07-10 一种基于epid的光栅位置标定和验证方法
CN202010660774.2 2020-07-10

Publications (1)

Publication Number Publication Date
WO2022007312A1 true WO2022007312A1 (zh) 2022-01-13

Family

ID=72768277

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/131312 WO2022007312A1 (zh) 2020-07-10 2020-11-25 一种基于epid的光栅位置标定和验证方法

Country Status (2)

Country Link
CN (1) CN111773560B (zh)
WO (1) WO2022007312A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114887239A (zh) * 2022-04-28 2022-08-12 苏州雷泰医疗科技有限公司 基于epid的光栅全场自动标定方法、计算设备及存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111773560B (zh) * 2020-07-10 2022-01-14 苏州雷泰医疗科技有限公司 一种基于epid的光栅位置标定和验证方法
CN113521560B (zh) * 2021-06-22 2024-04-09 苏州雷泰医疗科技有限公司 一种基于epid的光栅全自动标定方法、装置及放射治疗设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6322249B1 (en) * 1999-07-26 2001-11-27 Siemens Medical Solutions Usa, Inc. System and method for automatic calibration of a multileaf collimator
CN103127623A (zh) * 2013-03-06 2013-06-05 中国科学院合肥物质科学研究院 一种放射治疗中在线验证加速器出束准确性的方法
CN105288870A (zh) * 2015-11-03 2016-02-03 上海联影医疗科技有限公司 一种多叶准直器的校准方法
CN205672359U (zh) * 2016-04-18 2016-11-09 苏州雷泰医疗科技有限公司 一种光栅叶片控制装置及加速器治疗装置
CN109589504A (zh) * 2018-11-20 2019-04-09 华中科技大学同济医学院附属协和医院 一种多叶光栅叶片到位精度验证系统及其实现方法
CN111773560A (zh) * 2020-07-10 2020-10-16 苏州雷泰医疗科技有限公司 一种基于epid的光栅位置标定和验证方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8716668B2 (en) * 2012-09-18 2014-05-06 Elekta Ab (Publ) Radiation detector and radiotherapy apparatus
US9089696B2 (en) * 2013-11-07 2015-07-28 Varian Medical Systems International Ag Time-resolved pre-treatment portal dosimetry systems, devices, and methods

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6322249B1 (en) * 1999-07-26 2001-11-27 Siemens Medical Solutions Usa, Inc. System and method for automatic calibration of a multileaf collimator
CN103127623A (zh) * 2013-03-06 2013-06-05 中国科学院合肥物质科学研究院 一种放射治疗中在线验证加速器出束准确性的方法
CN105288870A (zh) * 2015-11-03 2016-02-03 上海联影医疗科技有限公司 一种多叶准直器的校准方法
CN205672359U (zh) * 2016-04-18 2016-11-09 苏州雷泰医疗科技有限公司 一种光栅叶片控制装置及加速器治疗装置
CN109589504A (zh) * 2018-11-20 2019-04-09 华中科技大学同济医学院附属协和医院 一种多叶光栅叶片到位精度验证系统及其实现方法
CN111773560A (zh) * 2020-07-10 2020-10-16 苏州雷泰医疗科技有限公司 一种基于epid的光栅位置标定和验证方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114887239A (zh) * 2022-04-28 2022-08-12 苏州雷泰医疗科技有限公司 基于epid的光栅全场自动标定方法、计算设备及存储介质

Also Published As

Publication number Publication date
CN111773560B (zh) 2022-01-14
CN111773560A (zh) 2020-10-16

Similar Documents

Publication Publication Date Title
WO2022007312A1 (zh) 一种基于epid的光栅位置标定和验证方法
KR100987855B1 (ko) 방사선 촬상장치
US10702710B2 (en) System and methods for processing images to measure collimator leaf and collimator performance
US7024026B1 (en) Relative calibration for dosimetric devices
CN110553665B (zh) 激光测距器光轴偏差自动测量装置及测量方法
CN1857163B (zh) 校准计算机断层x射线光束跟踪回路的方法和设备
US8699659B2 (en) Systems and methods for focal spot motion correction
JP4799053B2 (ja) X線画像における画像乱れの補償方法およびx線装置
EP3062289B1 (en) Quality assurance system for radiation treatment machine and quality assurance method thereof
CN104635428B (zh) 一种基于图像处理的调焦调平测量装置和方法
Wang et al. Correlation between gamma passing rate and complexity of IMRT plan due to MLC position errors
US7801344B2 (en) Edge boundary definition for radiographic detector
KR101912715B1 (ko) 방사선이 방출된 위치의 분포를 추정하는 방법 및 장치
US8045806B2 (en) Method and device for identifying material boundaries of a test object
US10067238B2 (en) Method and apparatus for ion beam Bragg Peak measurement
US7399119B2 (en) Method and system for measuring an alignment of a detector
JP2014128662A (ja) Ctシステムで使用するためのコリメータ
CN107851391B (zh) 台阶检测装置及台阶检测方法
CN104337530A (zh) X光探测器的探测器模块的中心位置确定方法及系统
WO2013119887A1 (en) Method and statistical validation technique for detecting differences between radiation therapy images with application to the detection and control of radiation therapy treatment delivery errors
JPS63147281A (ja) パタ−ン検出方法
CN114113173B (zh) 一种x射线设备、应用于x射线设备中的散射校正方法
WO2020010583A1 (zh) 放疗设备准直器安装检测方法、装置及系统
Mohammadi et al. Evaluation of relative transmitted dose for a step and shoot head and neck intensity modulated radiation therapy using a scanning liquid ionization chamber electronic portal imaging device
JP2023023437A (ja) 粒子線治療システム、および治療計画装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20943849

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20943849

Country of ref document: EP

Kind code of ref document: A1