WO2020077592A1 - Ct系统能谱不一致性的校正方法 - Google Patents

Ct系统能谱不一致性的校正方法 Download PDF

Info

Publication number
WO2020077592A1
WO2020077592A1 PCT/CN2018/110848 CN2018110848W WO2020077592A1 WO 2020077592 A1 WO2020077592 A1 WO 2020077592A1 CN 2018110848 W CN2018110848 W CN 2018110848W WO 2020077592 A1 WO2020077592 A1 WO 2020077592A1
Authority
WO
WIPO (PCT)
Prior art keywords
energy spectrum
uniform filter
detector unit
effective energy
filter
Prior art date
Application number
PCT/CN2018/110848
Other languages
English (en)
French (fr)
Inventor
高河伟
张丽
Original Assignee
清华大学
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 清华大学 filed Critical 清华大学
Priority to PCT/CN2018/110848 priority Critical patent/WO2020077592A1/zh
Publication of WO2020077592A1 publication Critical patent/WO2020077592A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]

Definitions

  • the present disclosure belongs to the technical field of image processing, and relates to a correction method of CT system energy spectrum inconsistency.
  • Computerized tomography is the use of X-ray beams and detectors to scan a section (tomography) of an object, using the characteristics of different absorption coefficients of X-rays of human organs or tissues to obtain the physical or chemical characteristics reflecting this section Projection data set, through computer operation to obtain the parameter value of any position on the cross section, and thus tomographic image.
  • the main components of the CT system include: X-ray source, detector, and rotating device.
  • the X-ray source and the detector make relative rotation around the object to obtain CT data at different rotation angles, also known as CT projection values.
  • CT projection values also known as CT projection values.
  • CT imaging the attenuation of X-rays follows an exponential law, CT The projection value is obtained indirectly, and usually requires "negative logarithm" preprocessing.
  • the X-ray photon generated by the X-ray source used in the CT system is a multi-color broad spectrum (Spectrum), so the CT system is generally a polychromatic system (Polychromatic).
  • Polychromatic polychromatic
  • the polychromatic energy spectrum will bring about a ray hardening effect, that is, when X-rays pass through an object, the low-energy X-rays relatively decay more, and the high-energy X-rays relatively decay relatively, so that X-rays Hardened ".
  • Radiation hardening is a fundamental physical problem in CT systems. If not handled properly, CT images will produce various artifacts, such as cupping, shading, streak, and ring artifacts. Artifacts, etc.
  • FIG. 1 is a schematic diagram of components in a CT system that may affect energy spectrum changes / distribution.
  • the energy spectrum in the CT system changes drastically with the spatial position, generally due to the addition of filter materials with drastic spatial changes in the system, as shown by the spatially non-uniform filter in FIG. 1.
  • the filter materials with sharp spatial changes in the components such as bowtie filters, spatial encoders, and primary modulators, etc.
  • the filter will be uneven in scan time, which will also cause the problem of inconsistency of the energy spectrum.
  • the present disclosure provides a correction method for the energy spectrum inconsistency of the CT system to correct the energy spectrum inconsistency.
  • a method for correcting the inconsistency of the energy spectrum of a CT system including: performing non-uniform filtering using the measured transmission value of the non-uniform filter and the effective energy spectrum of the CT system without the non-uniform filter Self-adaptive estimation of the film thickness to obtain the equivalent thickness of the non-uniform filter at each detector unit position; according to the linear attenuation coefficient of the object, thickness information and the equivalent thickness of the non-uniform filter at each detector unit position , Establish the hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system, and obtain the hardening correction look-up table according to the hardening function relationship; and correct the multi-color projection of the detector according to the hardening correction look-up table of each detector unit Value to the corresponding monochrome projection value.
  • the acquisition process of the effective energy spectrum of the CT system without the non-uniform filter includes: obtaining the effective energy spectrum of the CT system without the non-uniform filter by actual measurement; or by establishing a non-uniform filter
  • the model of the effective energy spectrum of the CT system of the filter; and the initial effective energy spectrum obtained in the model of the effective energy spectrum is optimized based on the transmission value data of the known filter of the CT system without the non-uniform filter, and the obtained optimization
  • the effective energy spectrum after is the effective energy spectrum of CT system without non-uniform filter.
  • the model of the effective energy spectrum of the CT system without the non-uniform filter includes the X-ray tube energy spectrum term, the heel-health compensation term, and other costs other than the non-uniform filter to be solved
  • the obtained unoptimized effective energy spectrum is an initial effective energy spectrum, which is obtained by a model or directly obtained by Monte Carlo simulation or measurement as a whole.
  • the model satisfies:
  • ⁇ 0 (E) T (E) ⁇ H (E) ⁇ F (E) ⁇ ⁇ (E)
  • ⁇ 0 (E) represents the CT system's unoptimized effective energy spectrum, which is the initial effective energy spectrum
  • T (E) represents the X-ray tube energy spectrum term, which can be used by Monte Carlo simulation, semi-empirical energy spectrum calculation tools, Or measured
  • H (E) represents the heel heel compensation term, ⁇ w is the linear attenuation coefficient of tungsten, and D w is the thickness of tungsten
  • F (E) represents the other realities except the non-uniform filter to be solved The attenuation term of the filter
  • ⁇ (E) represents the energy spectral response of the detector unit, ⁇ det is the linear attenuation coefficient of the crystal in the detector unit, and D det is the thickness of the crystal in the detector unit.
  • the method for optimizing the initial effective energy spectrum obtained in the model of the effective energy spectrum is: maximum expected estimation.
  • the transmission value of the non-uniform filter is the effective energy spectrum of the CT system without the non-uniform filter, the linear attenuation coefficient of the non-uniform filter, and the non-uniform filter at each detector unit position
  • the spectrum is used as the energy weight of the attenuation term; using the transmission value of the non-uniform filter, the linear attenuation coefficient of the non-uniform filter, and the effective energy spectrum of the CT system without the non-uniform filter, the equation can be solved to obtain the non-uniform filter
  • the equivalent thickness at each detector unit location That is: the equivalent thickness of the non-uniform filter at each detector unit position is the zero root of the following equation:
  • G (T m ) is a function of equivalent thickness
  • m is the transmission value of the non-uniform filter
  • ⁇ (E) is the effective energy spectrum of the CT system
  • ⁇ m is the linear attenuation coefficient of the non-uniform filter, which is known amount
  • T m required to be non-uniform filter equivalent thickness.
  • the method of solving equations using the transmission value of the non-uniform filter, the linear attenuation coefficient of the non-uniform filter, and the effective energy spectrum of the CT system without the non-uniform filter is Newton's iteration, ie
  • the solution of the zero root of the equation is realized by the Newton iteration process, which satisfies:
  • the hardening of the multi-color projection value and the mono-color projection value of the CT system is established
  • the functional relationship includes: calculating the effective energy spectrum after each detector unit includes the non-uniform filter in the entire CT system according to the equivalent thickness of the non-uniform filter at the position of each detector unit; and according to the object ’s
  • the linear attenuation coefficient, the thickness information, and the effective energy spectrum after each detector unit includes a non-uniform filter establish a hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system.
  • the effective energy spectrum after each detector unit includes a non-uniform filter is the effective energy spectrum of each detector unit without the non-uniform filter, the linear attenuation coefficient of the object, and the The function of the equivalent thickness of the non-uniform filter at each detector unit position, where the linear attenuation coefficient of the object and the equivalent thickness of the non-uniform filter at each detector unit position form the attenuation term, each detection
  • the effective energy spectrum of the CT system without the non-uniform filter of the filter unit is used as the energy weight of the attenuation term.
  • the effective energy spectrum of each detector unit satisfies:
  • the effective energy spectrum after including the non-uniform filter For each detector unit, the effective energy spectrum after including the non-uniform filter; ⁇ ij (E) is the effective energy spectrum of the CT system without non-uniform filter for each detector unit; ⁇ m (E) is the object Linear attenuation coefficient; T m (i, j) is the equivalent thickness of the non-uniform filter at each detector unit position; f [T m (i, j)] is a function of the equivalent thickness;
  • g pol is the multicolor projection value of the object;
  • ⁇ o (E) is the linear attenuation coefficient of the object;
  • g mon (L) is the monochrome projection value of the object;
  • L is the thickness of the object.
  • the hardening correction look-up table is obtained according to the relationship of the hardening function, including: selecting the object thickness value (L′ s) within a certain range; selecting the equivalent energy Calculate the monochromatic projection value g mon and the multicolor projection value g pol under each object thickness value according to the hardening function relationship; and establish a mapping relationship from the multicolor projection value to the monochromatic projection value, the mapping relationship is the hardening correction Look-up table; among them, the mapping relationship from the multi-color projection value to the monochromatic projection value can be realized by polynomial fitting.
  • the method further includes: Projection value data, through analysis or iterative image reconstruction process to obtain CT tomographic image of the object.
  • the non-uniform filter is obtained at each detector unit position The equivalent thickness of the system; and based on the linear attenuation coefficient of the object, the thickness information and the equivalent thickness of the non-uniform filter at each detector unit position, the hardening function relationship between the multi-color projection value and the mono-color projection value of the CT system is established, The hardening correction look-up table is obtained according to the relationship of the hardening function; the energy spectrum after each detector unit contains a non-uniform filter is obtained by adaptive estimation of the equivalent thickness one by one, and the multicolor projection value map is converted into a monochromatic projection value , Eliminating the inconsistency of CT system energy spectrum;
  • the estimation of the thickness of the non-uniform filter in this correction method is adaptive, it can be adaptively processed for each CT system, and can also be processed separately for a CT scan in a CT system, or even a CT scan
  • the filter under each angle of the image is processed; that is, the correction method is not only applicable to the case of a spatially unevenly distributed filter, but also to the case of an uneven filter in the scan time, which can be applied to dynamic
  • the thickness value of the filter at each scanning time point on each detector unit it is only necessary to calculate the thickness value of the filter at each scanning time point on each detector unit, and the thickness of a two-dimensional filter can be estimated. , Becomes a three-dimensional estimate on the time scale.
  • FIG. 1 is a schematic diagram of components in a CT system that may affect energy spectrum changes / distribution.
  • FIG. 2 is a flowchart of a method for correcting the inconsistency of the energy spectrum of the CT system according to an embodiment of the present disclosure.
  • FIG. 3 is a schematic diagram of the operation steps of the correction method of the CT system energy spectrum inconsistency shown in FIG. 2.
  • FIG. 4 is a result diagram before and after processing the energy spectrum inconsistency of the primary beam modulator (Primary Modulator) according to the correction method of the energy spectrum inconsistency of the CT system according to an embodiment of the present disclosure.
  • Primary Modulator Primary Modulator
  • the present disclosure provides a correction method for the inconsistency of the energy spectrum of the CT system.
  • the correction method uses the measured transmission value of the non-uniform filter and the effective energy spectrum of the CT system without the non-uniform filter to perform the non-uniform filter thickness self-adjustment.
  • the hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system, and the hardening correction look-up table is obtained according to the hardening function relationship; the energy spectrum after each detector unit contains a non-uniform filter is adaptively adopted by the equivalent thickness
  • the estimation is obtained, and the multicolor projection value map is converted into a monochromatic projection value, which eliminates the inconsistency of the CT system energy spectrum, and the correction method is not only suitable for static spatial non-uniform filters, but also for dynamic non-uniform time Filter processing.
  • a method for correcting the inconsistency of the energy spectrum of a CT system is provided.
  • FIG. 2 is a flowchart of a method for correcting the inconsistency of the energy spectrum of the CT system according to an embodiment of the present disclosure.
  • FIG. 3 is a schematic diagram of the operation steps of the correction method of the CT system energy spectrum inconsistency shown in FIG. 2. The italics in Figure 3 correspond to the contents in the dotted box.
  • the correction method of the CT system energy spectrum inconsistency of the present disclosure includes:
  • Step S21 Establish a model of the effective energy spectrum of the CT system without the non-uniform filter
  • the model of the effective energy spectrum of the CT system without the non-uniform filter includes the X-ray tube energy spectrum term, the heel heel compensation term, the attenuation term of the other true filter in addition to the non-uniform filter to be solved, and the detector
  • the spectrum is the initial effective energy spectrum, which is obtained by the model or directly obtained by Monte Carlo simulation or measurement as a whole.
  • the model of the effective energy spectrum of the CT system without the non-uniform filter satisfies the following expression:
  • ⁇ 0 (E) represents the CT system's unoptimized effective energy spectrum, which is the initial effective energy spectrum
  • T (E) represents the X-ray tube energy spectrum term, which can be used by Monte Carlo simulation, semi-empirical energy spectrum calculation tools, Or measured
  • H (E) represents the heel heel compensation term, ⁇ w is the linear attenuation coefficient of tungsten, and D w is the thickness of tungsten
  • F (E) represents the other realities except the non-uniform filter to be solved The attenuation term of the filter
  • ⁇ (E) represents the energy spectral response of the detector unit, ⁇ det is the linear attenuation coefficient of the crystal in the detector unit, and D det is the thickness of the crystal in the detector unit.
  • the unoptimized effective energy spectrum of the CT system here is the initial effective energy spectrum, which corresponds to the effective energy spectrum of the CT system without the non-uniform filter after optimization in step S22.
  • Step S22 The initial effective energy spectrum obtained in the model of the effective energy spectrum is optimized according to the transmission value data of the known filter of the CT system without the non-uniform filter, and the optimized effective energy spectrum obtained does not contain non-uniform Effective energy spectrum of CT system with uniform filter;
  • the initial effective energy spectrum obtained in step S21 often has some errors, and these errors may bring objective deviations to subsequent calculations. Therefore, the initial effective energy spectrum can be optimized through a series of experimental data of actual known material samples on the CT system.
  • the maximum expected estimate (EM) is used to optimize the initial effective energy spectrum ⁇ 0 (E) according to the transmission value data of the known filter of the CT system, to obtain the optimized effective energy spectrum ⁇ (E), That is, the effective energy spectrum of the CT system without the non-uniform filter, but the optimization method of the present disclosure is not limited to the embodiment shown, and may be other data optimization methods.
  • the effective energy spectrum of the CT system without the non-uniform filter can also be obtained by actual measurement.
  • Step S23 Use the measured transmission value of the non-uniform filter and the effective energy spectrum of the CT system without the non-uniform filter to adaptively estimate the thickness of the non-uniform filter, to obtain the non-uniform filter at each detector unit position Equivalent thickness
  • the transmission value of the non-uniform filter is a function of the effective energy spectrum of the CT system without the non-uniform filter, the linear attenuation coefficient of the non-uniform filter, and the equivalent thickness of the non-uniform filter at each detector unit position, where ,
  • the linear attenuation coefficient of the non-uniform filter and the equivalent thickness of the non-uniform filter at each detector unit position form the attenuation term, and the effective energy spectrum of the CT system without the non-uniform filter is used as the energy weight of the attenuation term .
  • the transmission method of the non-uniform filter, the linear attenuation coefficient of the non-uniform filter, and the effective energy spectrum of the CT system without the non-uniform filter are used to solve the equation.
  • the method used is Newton iteration, and other analysis methods can also be used. Method to solve for equivalent thickness.
  • the equivalent thickness of the non-uniform filter at each detector unit position is the zero root of the following equation:
  • G (T m ) is a function of equivalent thickness
  • m is the transmission value of the non-uniform filter
  • ⁇ (E) is the effective energy spectrum of the CT system without the non-uniform filter
  • ⁇ m is the non-uniform filter
  • the linear attenuation coefficient is a known quantity
  • T m is the equivalent thickness of the non-uniform filter.
  • the initial thickness can be set to 0, which converges after 2, 3 iterations.
  • Step S24 According to the linear attenuation coefficient of the object, the thickness information, and the equivalent thickness of the non-uniform filter at each detector unit position, establish a hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system, and The functional relationship gets the hardening correction look-up table;
  • This step S24 includes: calculating the effective energy spectrum after each detector unit includes the non-uniform filter in the entire CT system according to the equivalent thickness of the non-uniform filter at the position of each detector unit; The linear attenuation coefficient, the thickness information, and the effective energy spectrum after each detector unit includes a non-uniform filter, establish a hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system.
  • the effective energy spectrum after each detector unit includes a non-uniform filter is the effective energy spectrum of the CT system without non-uniform filter for each detector unit, the line attenuation coefficient of the object and the non-uniform filter in each detector A function of the equivalent thickness at the location of the unit, where the linear attenuation coefficient of the object and the equivalent thickness of the non-uniform filter at each detector unit position form the attenuation term, and the non-uniform filter at each detector unit
  • the effective energy spectrum of the CT system is used as the energy weight of the attenuation term.
  • the effective energy spectrum after each detector unit including the non-uniform filter in the entire CT system can be calculated .
  • the effective energy spectrum satisfies the following expression:
  • the effective energy spectrum after including the non-uniform filter is the effective energy spectrum of the CT system without non-uniform filter for each detector unit; ⁇ m (E) is the object Linear attenuation coefficient; T m (i, j) is the equivalent thickness of the non-uniform filter at each detector unit position; f [T m (i, j)] is a function of the equivalent thickness.
  • f [T m (i, j)] can be the equivalent thickness or the equivalent thickness multiplied by a coefficient or other function about the equivalent thickness, which can be adaptively set according to the actual situation.
  • the effective energy spectrum after each detector unit includes a non-uniform filter is the effective energy spectrum of each detector unit without the non-uniform filter, the linear attenuation coefficient of the object and the non-uniform filter
  • the function of the equivalent thickness at the position of each detector unit is a function that is specifically set according to the actual situation, and may also include other related parameters. This embodiment is merely an example, and any reasonable function form is disclosed in this disclosure Within the scope of protection.
  • the effective energy spectrum after each detector unit obtained based on formula (7) includes a non-uniform filter to establish a hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system.
  • the coefficient is ⁇ o (E) and the thickness is L.
  • the spectral hardening function of the object is described by the mathematical expression of the multicolor CT projection value.
  • the target monochromatic projection value is defined as an equivalent energy The product of the linear attenuation coefficient and the thickness of the scanned object under
  • g pol is the multicolor projection value of the object;
  • ⁇ o (E) is the linear attenuation coefficient of the object;
  • g mon (L) is the monochrome projection value of the object;
  • L is the thickness of the object.
  • the process of obtaining the hardening correction lookup table according to the hardening function relationship includes:
  • mapping relationship is the hardening correction look-up table.
  • Step S25 Correct the multi-color projection value of the detector to the corresponding monochromatic projection value according to the hardening correction look-up table of each detector unit;
  • Step S26 Based on the hardened corrected monochromatic projection value data, a CT tomographic image of the object is obtained through an analytical or iterative image reconstruction process;
  • the CT tomographic image of the object is obtained through an analytical or iterative image reconstruction process.
  • the correction method of the energy spectrum inconsistency of the CT system shown in this embodiment is not only applicable to static spatial non-uniform filters, but also applicable to the processing of dynamic temporal non-uniform filters.
  • FIG. 4 is a result diagram before and after processing the energy spectrum inconsistency of the primary beam modulator (Primary Modulator) according to the correction method of the energy spectrum inconsistency of the CT system according to an embodiment of the present disclosure.
  • Primary Modulator Primary Modulator
  • the energy spectrum inconsistency of the primary beam modulator (Primary Modulator) is processed using the correction method of the energy spectrum inconsistency of the CT system shown in this embodiment. 4 is shown.
  • (A), (b) and (c) in Figure 4 are the image reconstruction results of the three copper modulators with different intensities before processing, and (d), (e) and (f) in Figure 4 are the same as (a ), (B), (c) one-to-one corresponding image reconstruction results after energy spectrum inconsistency correction, it can be seen that the corrected CT image has almost no ring artifacts, so the energy spectrum of the CT system of the present disclosure is inconsistent
  • the correction method can correct the ring artifacts caused by the filter material with severe spatial changes in the CT system, and improve the clarity of the CT image.
  • the method for correcting the spectrum inconsistency of the CT system of the present disclosure is adaptive to the estimation of the thickness of the non-uniform filter, it can be adaptively processed for each CT system, and can also be scanned for one CT in one CT system. Separate processing, even for the filter under each angle in a CT scan; that is, the correction method is also applicable to the case where the filter is not uniform in the scan time, and can be applied to the dynamic bowtie filter In the processing of this case, you only need to calculate the thickness value of the filter on each detector unit at each scanning time point, and you can change the thickness estimate of a two-dimensional filter into a time scale. On the three-dimensional estimate.
  • the present disclosure provides a method for correcting the inconsistency of the energy spectrum of a CT system by using the measured transmission value of the non-uniform filter and the effective energy spectrum of the CT system without the non-uniform filter to perform the non-uniform filter Adaptive thickness estimation to obtain the equivalent thickness of the non-uniform filter at each detector unit position; and the equivalent thickness of the non-uniform filter at each detector unit position according to the linear attenuation coefficient of the object, thickness information and the non-uniform filter , Establish the hardening function relationship between the multi-color projection value and the monochromatic projection value of the CT system, and obtain a hardening correction look-up table according to the hardening function relationship; the energy spectrum after each detector unit contains a non-uniform filter is passed through the equivalent thickness one by one It is obtained by adaptive estimation, and the multi-color projection value map is converted into a monochromatic projection value, which eliminates the inconsistency of the CT system energy spectrum; and the correction method is not only applicable to
  • the correction method of the CT system energy spectrum inconsistency of the present disclosure also includes other common methods and steps. Since it is not related to the innovation of the present disclosure, it will not be repeated here.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Engineering & Computer Science (AREA)
  • Radiology & Medical Imaging (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Optics & Photonics (AREA)
  • Pathology (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

一种CT系统能谱不一致性的校正方法，包括：利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度(S23)；根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表(S24)；以及根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值(S25)。该校正方法消除了CT系统能谱不一致性，且不仅适用于静态的空间非均匀滤波片，还适用于对动态的时间非均匀滤波片的处理。

Description

CT系统能谱不一致性的校正方法 技术领域
本公开属于图像处理技术领域，涉及一种CT系统能谱不一致性的校正方法。
背景技术
计算机断层成像(CT，Computerized Tomography)是使用X射线束和探测器扫描物体的某一截面(断层)，利用人体器官或组织对X射线的吸收系数不同的特性获得反映此截面物理或化学特性的投影数据集合，通过计算机运算获得截面上任意位置的参数值，并由此得到断层图像。
CT系统的主要部件包括：X射线源、探测器、以及旋转装置。在CT成像过程中，X射线源与探测器围绕物体做相对旋转运动，以此获得不同旋转角度下的CT数据，也称CT投影值，在CT成像中，X射线的衰减服从指数规律，CT投影值是间接获取的，通常需要“负对数”预处理。
CT系统中所采用的X射线源所产生的X射线光子是一个多色的宽能谱(Spectrum)，因此CT系统一般是一个多色系统(Polychromatic)。根据X射线与物质相互作用的机制，多色能谱会带来射线硬化效应，即X射线穿过物体时其低能X射线相对衰减较多、高能X射线相对衰减较少，从而使得X射线“变硬”了。射线硬化是CT系统的一个根本性物理难题，如果处理不当会使得CT图像产生各种伪影，如杯状伪影(cupping)、明暗阴影(Shading)、条纹状伪影(Streak)、以及环状伪影等。X射线能谱在探测器视野上的空间分布剧烈变化(强烈的不一致性)会导致环状伪影。图1是CT系统中可能影响能谱变化/分布的部件的示意图。参照图1所示，CT系统中的能谱随空间位置剧烈变化，一般是由于系统中加入了空间变化剧烈的滤波材料造成的，如图1中的空间非均匀的滤波片所示，在实际的部件中空间变化剧烈的滤波材料诸如为领结(Bowtie)滤波片、空间编码器(coding)、以及主束调制器(Primary modulator)等。
因此，为了得到清晰的、不含各种伪影的CT断层图像，需要对由于空间非均匀滤波片带来的能谱不一致性进行校正。
根据系统设计的需要，有时候还会出现滤波片在扫描时间上不均匀的 情况，也会带来能谱不一致性的问题。
发明内容
(一)要解决的技术问题
针对上述问题，本公开提供了一种CT系统能谱不一致性的校正方法，以校正能谱不一致性。
(二)技术方案
根据本公开的一个方面，提供了一种CT系统能谱不一致性的校正方法，包括：利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；以及根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值。
在本公开的一些实施例中，不含非均匀滤波片的CT系统有效能谱的获取过程包括：通过实际测量得到不含非均匀滤波片的CT系统有效能谱；或者通过建立不含非均匀滤波片的CT系统有效能谱的模型；以及根据该不含非均匀滤波片的CT系统已知滤波片的透射值数据对有效能谱的模型中获得的初始有效能谱进行优化，获得的优化后的有效能谱为不含非均匀滤波片的CT系统有效能谱。
在本公开的一些实施例中，不含非均匀滤波片的CT系统有效能谱的模型包含X光管能谱项、后脚跟效应补偿项、除了待求解的非均匀滤波片之外的其他本真滤波片的衰减项和探测器单元的能谱响应项，其中，X光管能谱项可通过蒙特卡洛模拟、半经验能谱计算工具、或者测量得到；根据CT系统有效能谱的模型得到的未经优化的有效能谱为初始有效能谱，该初始有效能谱通过模型进行求取或者作为一个整体直接通过蒙特卡洛模拟或者测量得到。该模型满足：
χ 0(E)＝T(E)×H(E)×F(E)×Γ(E)
Figure PCTCN2018110848-appb-000001
Figure PCTCN2018110848-appb-000002
其中，χ 0(E)表示CT系统未经优化的有效能谱，为初始有效能谱；T(E)表示X光管能谱项，可通过蒙特卡洛模拟、半经验能谱计算工具、或者测量得到；H(E)表示后脚跟效应补偿项，μ w为钨的线衰减系数，D w为钨的厚度；F(E)表示除了待求解的非均匀滤波片之外的其他本真滤波片的衰减项；Γ(E)表示探测器单元的能谱响应，μ det为探测器单元中晶体的线衰减系数，D det为探测器单元中晶体的厚度。
在本公开的一些实施例中，对有效能谱的模型中获得的初始有效能谱进行优化的方法为：最大期望估计。
在本公开的一些实施例中，非均匀滤波片的透射值为不含非均匀滤波片的CT系统有效能谱、非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数：其中，非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重；利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解可得出非均匀滤波片在每一个探测器单元位置处的等效厚度。即：非均匀滤波片在每一个探测器单元位置处的等效厚度为如下方程的零根：
Figure PCTCN2018110848-appb-000003
其中，G(T m)为等效厚度的函数；m为非均匀滤波片的透射值；χ(E)为CT系统有效能谱；μ m为非均匀滤波片的线衰减系数，为已知量；T m为非均匀滤波片待求的等效厚度。
在本公开的一些实施例中，利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解采用的方法为牛顿迭代，即方程的零根的求解通过牛顿迭代过程来实现，该牛顿迭代过程满足：
Figure PCTCN2018110848-appb-000004
Figure PCTCN2018110848-appb-000005
在本公开的一些实施例中，根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影 值与单色投影值的硬化函数关系，包括：根据非均匀滤波片在每一个探测器单元位置处的等效厚度，计算出整个CT系统中，每一个探测器单元包括非均匀滤波片之后的有效能谱；以及根据物体的线衰减系数、厚度信息和每一个探测器单元包括非均匀滤波片之后的有效能谱，建立CT系统多色投影值与单色投影值的硬化函数关系。
在本公开的一些实施例中，每一个探测器单元包括非均匀滤波片之后的有效能谱为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱、物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数，其中，物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，每一个探测器单元的不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重。该每一个探测器单元的有效能谱满足：
Figure PCTCN2018110848-appb-000006
其中，
Figure PCTCN2018110848-appb-000007
为每一个探测器单元包括非均匀滤波片之后的有效能谱；χ ij(E)为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱；μ m(E)为物体的线衰减系数；T m(i，j)为非均匀滤波片在每一个探测器单元位置处的等效厚度；f[T m(i，j)]为等效厚度的函数；
硬化函数关系满足：
Figure PCTCN2018110848-appb-000008
Figure PCTCN2018110848-appb-000009
其中，g pol为物体多色投影值；
Figure PCTCN2018110848-appb-000010
为每一个探测器单元包括非均匀滤波片之后的有效能谱；μ o(E)为物体的线衰减系数；
Figure PCTCN2018110848-appb-000011
为在一个等效能量
Figure PCTCN2018110848-appb-000012
下物体的线衰减系数；g mon(L)为物体单色投影值；L为物体的厚度。
在本公开的一些实施例中，依据硬化函数关系得到硬化校正查找表， 包括：选取一定范围内的物体厚度值(L′s)；选取等效能量
Figure PCTCN2018110848-appb-000013
依据硬化函数关系计算每一个物体厚度值下的单色投影值g mon和多色投影值g pol；以及建立一个从多色投影值到单色投影值的映射关系，该映射关系即为硬化校正查找表；其中，从多色投影值到单色投影值的映射关系，可使用多项式拟合来实现。
在本公开的一些实施例中，在根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值的步骤之后还包括：基于硬化校正后的单色投影值数据，经过解析或者迭代的图像重建过程得到物体的CT断层图像。
(三)有益效果
从上述技术方案可以看出，本公开提供的CT系统能谱不一致性的校正方法，具有以下有益效果：
(1)通过利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；以及根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；将每个探测器单元包含非均匀滤波片之后的能谱逐个通过等效厚度的自适应估计求得，将多色投影值映射转换为单色投影值，消除了CT系统能谱不一致性；
(2)由于该校正方法中对非均匀滤波片厚度的估计是自适应的，可以对每一个CT系统自适应处理，也可以对一个CT系统中的一个CT扫描单独处理，甚至是一个CT扫描中的每一个角度下的滤波片做处理；即，本校正方法不仅适用于空间上分布不均匀的滤波片的情形，也适用于在扫描时间上滤波片不均匀的情形，可应用于对动态领结(Bowtie)滤波片的处理中，对该情形，只需要计算每一个扫描时间点的滤波片在每个探测器单元上的厚度值即可，便能将一个二维的滤波片厚度的估计，变成了时间尺度上的三维的估计。
附图说明
图1为CT系统中可能影响能谱变化/分布的部件的示意图。
图2为根据本公开一实施例所示的CT系统能谱不一致性的校正方法的流程图。
图3为如图2所示的CT系统能谱不一致性的校正方法的操作步骤示意图。
图4为根据本公开一实施例所示的CT系统能谱不一致性的校正方法对主束调制器(Primary Modulator)的能谱不一致性进行处理前后的结果图。
具体实施方式
本公开提供了一种CT系统能谱不一致性的校正方法，该校正方法通过利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；并根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；将每个探测器单元包含非均匀滤波片之后的能谱逐个通过等效厚度的自适应估计求得，将多色投影值映射转换为单色投影值，消除了CT系统能谱不一致性，且该校正方法不仅适用于静态的空间非均匀滤波片，还适用于对动态的时间非均匀滤波片的处理。
为使本公开的目的、技术方案和优点更加清楚明白，以下结合具体实施例，并参照附图，对本公开进一步详细说明。
在本公开的第一个示例性实施例中，提供了一种CT系统能谱不一致性的校正方法。
图2为根据本公开一实施例所示的CT系统能谱不一致性的校正方法的流程图。图3为如图2所示的CT系统能谱不一致性的校正方法的操作步骤示意图。图3中的斜体字对应虚线框中的内容。
参照图2和图3所示，本公开的CT系统能谱不一致性的校正方法，包括：
步骤S21：建立不含非均匀滤波片的CT系统有效能谱的模型；
不含非均匀滤波片的CT系统有效能谱的模型包含X光管能谱项、后脚跟效应补偿项、除了待求解的非均匀滤波片之外的其他本真滤波片的衰 减项和探测器单元的能谱响应项，其中，X光管能谱项可通过蒙特卡洛模拟、半经验能谱计算工具、或者测量得到；根据该CT系统有效能谱的模型得到的未经优化的有效能谱为初始有效能谱，该初始有效能谱通过模型进行求取或者作为一个整体直接通过蒙特卡洛模拟或者测量得到。
本实施例中，不含非均匀滤波片的CT系统有效能谱的模型满足如下表达式：
χ 0(E)＝T(E)×H(E)×F(E)×Γ(E)    (1)
Figure PCTCN2018110848-appb-000014
Figure PCTCN2018110848-appb-000015
其中，χ 0(E)表示CT系统未经优化的有效能谱，为初始有效能谱；T(E)表示X光管能谱项，可通过蒙特卡洛模拟、半经验能谱计算工具、或者测量得到；H(E)表示后脚跟效应补偿项，μ w为钨的线衰减系数，D w为钨的厚度；F(E)表示除了待求解的非均匀滤波片之外的其他本真滤波片的衰减项；Γ(E)表示探测器单元的能谱响应，μ det为探测器单元中晶体的线衰减系数，D det为探测器单元中晶体的厚度。
这里的CT系统未经优化的有效能谱为初始有效能谱，与步骤S22中进行优化之后的不含非均匀滤波片的CT系统有效能谱对应。
本实施例中，CT系统的初始有效能谱χ 0(E)除了根据上述公式(1)进行求取之外，还可以作为一个整体直接通过蒙特卡洛模拟或者测量得到。
步骤S22：根据该不含非均匀滤波片的CT系统已知滤波片的透射值数据对有效能谱的模型中获得的初始有效能谱进行优化，获得的优化后的有效能谱为不含非均匀滤波片的CT系统有效能谱；
步骤S21获得的初始有效能谱，往往是有一些误差的，这些误差可能给后续的计算带来客观的偏差。因此，可以通过一系列该CT系统上的实际已知材料样品的实验数据来优化初始有效能谱。
本实施例中，根据该CT系统已知滤波片的透射值数据利用最大期望估计(EM)来对初始有效能谱χ 0(E)进行优化，获得优化后的有效能谱χ(E)，即为不含非均匀滤波片的CT系统有效能谱，但本公开的优化方法不局限于实施例所示，还可以是其他数据优化方法。
除了采用步骤S21和S22来得到不含非均匀滤波片的CT系统有效能谱之外，还可以通过实际测量得到不含非均匀滤波片的CT系统有效能谱。
步骤S23：利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；
非均匀滤波片的透射值为不含非均匀滤波片的CT系统有效能谱、非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数，其中，非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重，利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解可得出非均匀滤波片在每一个探测器单元位置处的等效厚度。可选的，利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解采用的方法为牛顿迭代，也可以采用其他的解析方法求解等效厚度。
本实施例中，非均匀滤波片在每一个探测器单元位置处的等效厚度为如下方程的零根：
Figure PCTCN2018110848-appb-000016
其中，G(T m)为等效厚度的函数；m为非均匀滤波片的透射值；χ(E)为不含非均匀滤波片的CT系统有效能谱；μ m为非均匀滤波片的线衰减系数，为已知量；T m为非均匀滤波片待求的等效厚度。
对上述方程(4)的求解可以通过下面的牛顿迭代过程来实现：
Figure PCTCN2018110848-appb-000017
Figure PCTCN2018110848-appb-000018
在实际计算迭代过程中，初始厚度可以设为0，经过2、3次迭代后即收敛。
步骤S24：根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；
该步骤S24包括：根据非均匀滤波片在每一个探测器单元位置处的等效厚度，计算出整个CT系统中，每一个探测器单元包括非均匀滤波片之后的有效能谱；以及根据物体的线衰减系数、厚度信息和每一个探测器单元包括非均匀滤波片之后的有效能谱，建立CT系统多色投影值与单色投影值的硬化函数关系。
每一个探测器单元包括非均匀滤波片之后的有效能谱为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱、物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数，其中，物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，每一个探测器单元的不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重。
本实施例中，根据步骤S23获得的非均匀滤波片在每一个探测器单元位置处的等效厚度，可以计算出整个CT系统中，每一个探测器单元包括非均匀滤波片之后的有效能谱，该有效能谱满足如下表达式：
Figure PCTCN2018110848-appb-000019
其中，
Figure PCTCN2018110848-appb-000020
为每一个探测器单元包括非均匀滤波片之后的有效能谱；χ ij(E)为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱；μ m(E)为物体的线衰减系数；T m(i，j)为非均匀滤波片在每一个探测器单元位置处的等效厚度；f[T m(i，j)]为等效厚度的函数。
其中，f[T m(i，j)]可以为等效厚度或者等效厚度乘以一个系数或者关于等效厚度的其它函数，该函数根据实际情况可进行适应性设置。
需要说明的是，每一个探测器单元包括非均匀滤波片之后的有效能谱为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱、物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数，此函数为根据实际情况进行具体设置的函数，还可以包括其他相关参数，本实施例仅作为举例，任何合理的函数形式均在本公开的保护范围之内。
基于公式(7)获得的每一个探测器单元包括非均匀滤波片之后的有效能谱建立CT系统多色投影值与单色投影值的硬化函数关系，对于单一物体来说，该物体的线衰减系数为μ o(E)，厚度为L，该物体的能谱硬化函 数通过对多色CT投影值的数学表达式来描述，目标单色投影值定义为在一个等效能量
Figure PCTCN2018110848-appb-000021
下的被扫描物体的线衰减系数与厚度的乘积，该硬化函数关系满足如下表达式：
Figure PCTCN2018110848-appb-000022
Figure PCTCN2018110848-appb-000023
其中，g pol为物体多色投影值；
Figure PCTCN2018110848-appb-000024
为每一个探测器单元包括非均匀滤波片之后的有效能谱；μ o(E)为物体的线衰减系数；
Figure PCTCN2018110848-appb-000025
为在一个等效能量
Figure PCTCN2018110848-appb-000026
下物体的线衰减系数；g mon(L)为物体单色投影值；L为物体的厚度。
本实施例中，依据硬化函数关系得到硬化校正查找表的过程包括：
(a)选取一定范围内的物体厚度值(L′s)；
(b)选取等效能量
Figure PCTCN2018110848-appb-000027
(c)利用公式(7)(8)(9)，计算每一个物体厚度值下的单色投影值g mon和多色投影值g pol；以及
(d)利用多项式拟合或者其他数值计算方法，建立一个从多色投影值到单色投影值的映射关系，该映射关系即为硬化校正查找表。
步骤S25：根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值；
获得硬化校正查找表之后，对于特定探测器单元下的多色投影值，只需要根据该多色投影值查找对应单色投影值就完成了硬化校正的工作，即实现了“线性化”。由于该校正过程，已经包括了对逐个探测器单元对应能谱分别处理，因此CT系统能谱不一致性将会得以消除。
步骤S26：基于硬化校正后的单色投影值数据，经过解析或者迭代的图像重建过程得到物体的CT断层图像；
本实施例中，基于步骤S25得到的硬化校正后的单色投影值数据，即为硬化校正后的CT投影数据，经过解析或者迭代的图像重建过程得到物体的CT断层图像。
本实施例所示的CT系统能谱不一致性的校正方法不仅适用于静态的空间非均匀滤波片，还适用于对动态的时间非均匀滤波片的处理。
至此，本实施例的CT系统能谱不一致性的校正方法介绍完毕。
下面结合一应用实例来说明本公开的CT系统能谱不一致性的校正方法的有益效果。
图4为根据本公开一实施例所示的CT系统能谱不一致性的校正方法对主束调制器(Primary Modulator)的能谱不一致性进行处理前后的结果图。
在本公开的一个应用的实例中，利用本实施例所示的CT系统能谱不一致性的校正方法对主束调制器(Primary Modulator)的能谱不一致性进行处理，处理前后的对比结果参照图4所示。
图4中(a)、(b)、(c)为处理前的三种不同强度的铜调制器的图像重建结果，图4中(d)、(e)、(f)分别为与(a)、(b)、(c)一一对应的进行能谱不一致校正后的图像重建结果，可见，经过校正后的CT图像几乎不存在环状伪影，因此利用本公开的CT系统能谱不一致性的校正方法对CT系统中由于空间变化剧烈的滤波材料引起的环状伪影实现了校正，提高了CT图像的清晰度。
另外，由于本公开的CT系统能谱不一致性的校正方法中对非均匀滤波片厚度的估计是自适应的，可以对每一个CT系统自适应处理，也可以对一个CT系统中的一个CT扫描单独处理，甚至是一个CT扫描中的每一个角度下的滤波片做处理；即，本校正方法也适用于在扫描时间上滤波片不均匀的情形，可应用于对动态领结(Bowtie)滤波片的处理中，对该情形，只需要计算每一个扫描时间点的滤波片在每个探测器单元上的厚度值即可，便能将一个二维的滤波片厚度的估计，变成了时间尺度上的三维的估计。
综上所述，本公开提供了一种CT系统能谱不一致性的校正方法，通过利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；以及根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影 值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；将每个探测器单元包含非均匀滤波片之后的能谱逐个通过等效厚度的自适应估计求得，将多色投影值映射转换为单色投影值，消除了CT系统能谱不一致性；且该校正方法不仅适用于静态的空间非均匀滤波片，还适用于对动态的时间非均匀滤波片的处理，即也适用于在扫描时间上滤波片不均匀的情形，可应用于对动态Bowtie滤波片的处理中，对该情形，只需要计算每一个扫描时间点的滤波片在每个探测器单元上的厚度值即可，其他实现步骤，基本保持不变，便能将一个二维的滤波片厚度的估计，变成了时间尺度上的三维的估计，极大拓宽了校正的范围和尺度。
当然，根据实际需要，本公开的CT系统能谱不一致性的校正方法还包含其他的常见方法和步骤，由于同本公开的创新之处无关，此处不再赘述。
此外，除非特别描述或必须依序发生的步骤，上述步骤的顺序并无限制于以上所列，且可根据所需设计而变化或重新安排。并且上述实施例可基于设计及可靠度的考虑，彼此混合搭配使用或与其他实施例混合搭配使用，即不同实施例中的技术特征可以自由组合形成更多的实施例。
以上所述的具体实施例，对本公开的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本公开的具体实施例而已，并不用于限制本公开，凡在本公开的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。

Claims (10)

  1. 一种CT系统能谱不一致性的校正方法，包括：
    利用测量的非均匀滤波片的透射值和不含非均匀滤波片的CT系统有效能谱进行非均匀滤波片厚度自适应估计，获得非均匀滤波片在每一个探测器单元位置处的等效厚度；
    根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，并依据硬化函数关系得到硬化校正查找表；以及
    根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值。
  2. 根据权利要求1所述的校正方法，其中，所述不含非均匀滤波片的CT系统有效能谱的获取过程包括：
    通过实际测量得到不含非均匀滤波片的CT系统有效能谱；或者
    通过建立不含非均匀滤波片的CT系统有效能谱的模型；以及
    根据该不含非均匀滤波片的CT系统已知滤波片的透射值数据对有效能谱的模型中获得的初始有效能谱进行优化，获得的优化后的有效能谱为不含非均匀滤波片的CT系统有效能谱。
  3. 根据权利要求2所述的校正方法，其中，所述不含非均匀滤波片的CT系统有效能谱的模型包含X光管能谱项、后脚跟效应补偿项、除了待求解的非均匀滤波片之外的其他本真滤波片的衰减项和探测器单元的能谱响应项，其中，X光管能谱项可通过蒙特卡洛模拟、半经验能谱计算工具、或者测量得到；根据所述CT系统有效能谱的模型得到的未经优化的有效能谱为初始有效能谱，该初始有效能谱通过模型进行求取或者作为一个整体直接通过蒙特卡洛模拟或者测量得到。
  4. 根据权利要求2所述的校正方法，其中，所述对有效能谱的模型中获得的初始有效能谱进行优化的方法为：最大期望估计。
  5. 根据权利要求1所述的校正方法，其中，所述非均匀滤波片的透射值为不含非均匀滤波片的CT系统有效能谱、非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数：其中， 非均匀滤波片的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重；利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解可得出非均匀滤波片在每一个探测器单元位置处的等效厚度。
  6. 根据权利要求5所述的校正方法，其中，所述利用非均匀滤波片的透射值、非均匀滤波片的线衰减系数和不含非均匀滤波片的CT系统有效能谱进行方程求解采用的方法为牛顿迭代。
  7. 根据权利要求1所述的校正方法，其中，所述根据物体的线衰减系数、厚度信息和非均匀滤波片在每一个探测器单元位置处的等效厚度，建立CT系统多色投影值与单色投影值的硬化函数关系，包括：
    根据非均匀滤波片在每一个探测器单元位置处的等效厚度，计算出整个CT系统中，每一个探测器单元包括非均匀滤波片之后的有效能谱；以及
    根据物体的线衰减系数、厚度信息和每一个探测器单元包括非均匀滤波片之后的有效能谱，建立CT系统多色投影值与单色投影值的硬化函数关系。
  8. 根据权利要求7所述的校正方法，其中，所述每一个探测器单元包括非均匀滤波片之后的有效能谱为每一个探测器单元的不含非均匀滤波片的CT系统有效能谱、物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度的函数，其中，物体的线衰减系数和非均匀滤波片在每一个探测器单元位置处的等效厚度形成衰减项，每一个探测器单元的不含非均匀滤波片的CT系统有效能谱作为衰减项在能量上的权重。
  9. 根据权利要求8所述的校正方法，其中，所述依据硬化函数关系得到硬化校正查找表，包括：
    选取一定范围内的物体厚度值；
    选取等效能量；
    依据硬化函数关系计算每一个物体厚度值下的单色投影值和多色投影值；以及
    建立一个从多色投影值到单色投影值的映射关系，该映射关系即为硬 化校正查找表；
    其中，从多色投影值到单色投影值的映射关系，可使用多项式拟合来实现。
  10. 根据权利要求1至9中任一项所述的校正方法，在所述根据每一个探测器单元的硬化校正查找表，校正该探测器的多色投影值至对应单色投影值的步骤之后还包括：
    基于硬化校正后的单色投影值数据，经过解析或者迭代的图像重建过程得到物体的CT断层图像。
PCT/CN2018/110848 2018-10-18 2018-10-18 Ct系统能谱不一致性的校正方法 WO2020077592A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/110848 WO2020077592A1 (zh) 2018-10-18 2018-10-18 Ct系统能谱不一致性的校正方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/110848 WO2020077592A1 (zh) 2018-10-18 2018-10-18 Ct系统能谱不一致性的校正方法

Publications (1)

Publication Number Publication Date
WO2020077592A1 true WO2020077592A1 (zh) 2020-04-23

Family

ID=70284413

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/110848 WO2020077592A1 (zh) 2018-10-18 2018-10-18 Ct系统能谱不一致性的校正方法

Country Status (1)

Country Link
WO (1) WO2020077592A1 (zh)

Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5528644A (en) * 1994-01-24 1996-06-18 Kabushiki Kaish Toshiba X-ray CT scanner and method of collecting image data in the same
US6324240B1 (en) * 1998-11-12 2001-11-27 The Board Of Trustees Of The Leland Stanford Junior University Method for beam hardening correction in quantitative computed X-ray tomography
CN1359089A (zh) * 2000-10-17 2002-07-17 西门子公司 对用ct仪记录的原始图象进行射束硬化校正的方法
CN1483383A (zh) * 2003-07-16 2004-03-24 沈阳东软数字医疗系统股份有限公司 一种ct机射束硬化的校正方法
CN1535659A (zh) * 2003-04-04 2004-10-13 GEҽҩϵͳ����Ƽ���˾ Ct系统校正系数计算方法、束硬化后处理方法及ct系统
CN1775176A (zh) * 2004-11-16 2006-05-24 北京航空航天大学 基于原始投影正弦图的ct射束硬化校正方法
CN101023867A (zh) * 2006-02-17 2007-08-29 株式会社东芝 数据校正装置、数据校正方法、磁共振成像装置和x射线ct装置
DE102006046047A1 (de) * 2006-09-28 2008-04-03 Siemens Ag Verfahren zur kombinierten Knochenaufhärtungs- und Streustrahlungskorrektur in der Röntgen-Computertomographie
US20080095302A1 (en) * 2006-10-13 2008-04-24 Siemens Aktiengesellschaft Method for hardening correction in medical imaging
CN101226642A (zh) * 2008-01-25 2008-07-23 西安交通大学 基于ct数据一致性的投影射束硬化校正方法
CN101266216A (zh) * 2007-03-14 2008-09-17 清华大学 标定双能ct系统的方法和图像重建方法
JP2010000240A (ja) * 2008-06-20 2010-01-07 Toshiba Corp Ｘ線ｃｔ装置
CN103134823A (zh) * 2013-03-21 2013-06-05 重庆大学 一种基于卷积的x射线ct系统射束硬化校正方法
CN103559729A (zh) * 2013-11-18 2014-02-05 首都师范大学 一种双能谱ct图像迭代重建方法
CN105608721A (zh) * 2016-01-30 2016-05-25 上海联影医疗科技有限公司 计算机断层成像伪影校正方法及装置
CN109363703A (zh) * 2018-10-18 2019-02-22 清华大学 Ct系统能谱不一致性的校正方法

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5528644A (en) * 1994-01-24 1996-06-18 Kabushiki Kaish Toshiba X-ray CT scanner and method of collecting image data in the same
US6324240B1 (en) * 1998-11-12 2001-11-27 The Board Of Trustees Of The Leland Stanford Junior University Method for beam hardening correction in quantitative computed X-ray tomography
CN1359089A (zh) * 2000-10-17 2002-07-17 西门子公司 对用ct仪记录的原始图象进行射束硬化校正的方法
CN1535659A (zh) * 2003-04-04 2004-10-13 GEҽҩϵͳ����Ƽ���˾ Ct系统校正系数计算方法、束硬化后处理方法及ct系统
CN1483383A (zh) * 2003-07-16 2004-03-24 沈阳东软数字医疗系统股份有限公司 一种ct机射束硬化的校正方法
CN1775176A (zh) * 2004-11-16 2006-05-24 北京航空航天大学 基于原始投影正弦图的ct射束硬化校正方法
CN101023867A (zh) * 2006-02-17 2007-08-29 株式会社东芝 数据校正装置、数据校正方法、磁共振成像装置和x射线ct装置
DE102006046047A1 (de) * 2006-09-28 2008-04-03 Siemens Ag Verfahren zur kombinierten Knochenaufhärtungs- und Streustrahlungskorrektur in der Röntgen-Computertomographie
US20080095302A1 (en) * 2006-10-13 2008-04-24 Siemens Aktiengesellschaft Method for hardening correction in medical imaging
US7822172B2 (en) * 2006-10-13 2010-10-26 Siemens Aktiengesellschaft Method for hardening correction in medical imaging
CN101266216A (zh) * 2007-03-14 2008-09-17 清华大学 标定双能ct系统的方法和图像重建方法
CN101226642A (zh) * 2008-01-25 2008-07-23 西安交通大学 基于ct数据一致性的投影射束硬化校正方法
JP2010000240A (ja) * 2008-06-20 2010-01-07 Toshiba Corp Ｘ線ｃｔ装置
CN103134823A (zh) * 2013-03-21 2013-06-05 重庆大学 一种基于卷积的x射线ct系统射束硬化校正方法
CN103559729A (zh) * 2013-11-18 2014-02-05 首都师范大学 一种双能谱ct图像迭代重建方法
CN105608721A (zh) * 2016-01-30 2016-05-25 上海联影医疗科技有限公司 计算机断层成像伪影校正方法及装置
CN109363703A (zh) * 2018-10-18 2019-02-22 清华大学 Ct系统能谱不一致性的校正方法

Similar Documents

Publication Publication Date Title
CN109363703B (zh) Ct系统能谱不一致性的校正方法
Schmidt et al. A spectral CT method to directly estimate basis material maps from experimental photon-counting data
US10716527B2 (en) Combined sinogram- and image-domain material decomposition for spectral computed tomography (CT)
US10111638B2 (en) Apparatus and method for registration and reprojection-based material decomposition for spectrally resolved computed tomography
Zhao et al. Patient-specific scatter correction for flat-panel detector-based cone-beam CT imaging
US10902650B2 (en) X-ray beam-hardening correction in tomographic reconstruction using the Alvarez-Macovski attenuation model
CN110522466B (zh) 确定患者重量和/或体重指数的方法和装置
JP2008272476A (ja) 単位面積質量画像の作成方法
US10605933B2 (en) X-ray spectral calibration technique for cone-beam CT
CN108338802B (zh) 用于减少图像伪影的方法
CN104050631A (zh) 一种低剂量ct图像重建方法
Li et al. Reduction of beam hardening artifacts in cone-beam CT imaging via SMART-RECON algorithm
US20220042932A1 (en) Estimating Background Radiation from Unknown Sources
Mason et al. Quantitative cone-beam CT reconstruction with polyenergetic scatter model fusion
Fan et al. Image‐domain shading correction for cone‐beam CT without prior patient information
Fragnaud et al. CAD-based X-ray CT calibration and error compensation
Luo et al. A fast beam hardening correction method incorporated in a filtered back-projection based MAP algorithm
Pivot et al. Scatter correction for spectral CT using a primary modulator mask
Zhang et al. An analysis of scatter characteristics in x-ray CT spectral correction
Grimmer et al. Empirical binary tomography calibration (EBTC) for the precorrection of beam hardening and scatter for flat panel CT
WO2020077592A1 (zh) Ct系统能谱不一致性的校正方法
JP6786006B2 (ja) スペクトルｃｔデータセットからの電子密度データセットの生成
Nishihata et al. A non-iterative data-driven beam hardening correction for single-material objects
Chukalina et al. A hardware and software system for tomographic research: reconstruction via regularization
Yu et al. Heel effect adaptive flat field correction of digital x‐ray detectors

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: 18936973

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: 18936973

Country of ref document: EP

Kind code of ref document: A1