WO2013097390A1 - Pet系统中图像的衰减校正方法及装置 - Google Patents

Pet系统中图像的衰减校正方法及装置 Download PDF

Info

Publication number
WO2013097390A1
WO2013097390A1 PCT/CN2012/074736 CN2012074736W WO2013097390A1 WO 2013097390 A1 WO2013097390 A1 WO 2013097390A1 CN 2012074736 W CN2012074736 W CN 2012074736W WO 2013097390 A1 WO2013097390 A1 WO 2013097390A1
Authority
WO
WIPO (PCT)
Prior art keywords
attenuation
image
osem
fbp
attenuation image
Prior art date
Application number
PCT/CN2012/074736
Other languages
English (en)
French (fr)
Inventor
孙智鹏
Original Assignee
沈阳东软派斯通医疗系统有限公司
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 沈阳东软派斯通医疗系统有限公司 filed Critical 沈阳东软派斯通医疗系统有限公司
Priority to US14/359,108 priority Critical patent/US9311723B2/en
Publication of WO2013097390A1 publication Critical patent/WO2013097390A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/002D [Two Dimensional] image generation
    • G06T11/003Reconstruction from projections, e.g. tomography
    • G06T11/005Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/77Retouching; Inpainting; Scratch removal
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10072Tomographic images
    • G06T2207/10104Positron emission tomography [PET]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20004Adaptive image processing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30004Biomedical image processing

Definitions

  • the present invention relates to the field of medical image processing, and in particular to a method and apparatus for attenuation correction applied to medical images in a PET system.
  • each patient typically needs to perform two scans, one for the Emissio scan and the other for the Transmission scan.
  • the data collected by the Emission scan generally reflects the distribution of the drug in the patient, but there is no attenuation correction, so there is a problem of quantitative inaccuracy; the data collected by the Transmission scan is specifically used to generate the attenuation chord, to The data obtained by the Emission scan is attenuated.
  • the Emission scan data can reflect the distribution of the drug in the patient's body. The scan takes about 4 minutes; the Transmission scan takes about 30 minutes to get a relatively accurate scan image.
  • a segmented attenuation correction algorithm (SAC) is adopted, which firstly filters the Transmissio scan data.
  • FBP Projection
  • FBP Projection
  • a corresponding value is assigned to the pixels of each area, and then the assigned image is forward projected, and the obtained attenuation data is used to attenuate the Emission scan data.
  • the above-mentioned attenuation correction method still causes a bed to take more than 10 minutes, and if the whole body scan is performed for the patient, it takes more than one hour, or there is a problem that the attenuation correction of the medical image scanned for the PET device is inefficient. Therefore, how to propose an innovative attenuation correction method for medical images, in the case of reducing the transmission scan time, can also ensure the accuracy of the image after attenuation correction, thereby increasing the input-output ratio of the device and reducing the patient's scanning band. Coming into the heart and physical discomfort. Summary of the invention
  • the technical problem to be solved by the present invention is to provide a method for attenuating medical images in a PET system to ensure that the image accuracy after attenuation correction can be ensured while reducing the Transmi ss ion scanning time.
  • Another object of the present invention is to apply the above concept to a specific application environment, and to provide an attenuation correction device for medical images in a PET system, thereby ensuring the implementation and application of the method.
  • an embodiment of the present invention provides a method for correcting an image in a PET system, including:
  • the transmission scan chord data is reconstructed by using the Bayesian model-based ordered subset maximum expectation OSEM-B method and the filtered back projection FBP method to obtain an OSEM-B attenuation image and a first FBP attenuation image, respectively;
  • the emission Emission scan chord data of the current PET device is subjected to attenuation correction using an attenuation chord generated in accordance with the effective attenuation image.
  • the weighting the OSEM-B attenuation image and the first FBP attenuation image to obtain an effective attenuation image comprises:
  • ⁇ £ ⁇ is the attenuation coefficient of the OSEM-B attenuation image, and is the attenuation coefficient of the FBP attenuation image, "is a weighting parameter.
  • ⁇ £ ⁇ is the attenuation coefficient of the OSEM-B attenuation image, and is the attenuation coefficient of the FBP attenuation image, "is a weighting parameter.
  • the The acquisition method is specifically as follows:
  • chord data of the inherent noise scan of the transmission device before the current PET device is shipped from the factory and performing FBP reconstruction on the chord data of the noise scan to obtain a second FBP attenuation image
  • the a is calculated using a histogram of the first tissue region and the second tissue region.
  • the calculating, by using a histogram of the first tissue region and the second tissue region is specifically:
  • the transmitting scan chord data is reconstructed by using an ordered subset maximum expectation OSEM-B method based on a Bayesian model, and specifically includes:
  • the w . ⁇ is the attenuation coefficient of water.
  • the invention also provides an image attenuation correction device for a PET system, comprising: an acquisition source data module, configured to acquire transmission transmission scan string data of a current PET device;
  • An OSEM-B reconstruction module is configured to reconstruct the transmission scan chord data by using a Bayesian model-based ordered subset maximum expected value OSEM-B method to obtain an OSEM-B attenuation image;
  • An FBP reconstruction module is configured to reconstruct the transmission scan chord data by using a filtered inverse projection FBP method to obtain a first FBP attenuation image
  • a weighting calculation module configured to weight the OSEM-B attenuation image and the first FBP attenuation image to obtain an effective attenuation image
  • an attenuation correction module configured to perform attenuation correction on the transmitted Emission scan chord data of the current PET device by using an attenuation chord generated according to the effective attenuation image.
  • the weighting calculation module is specifically configured to:
  • the effective attenuation coefficient of the effective attenuation image is calculated by using the formula ⁇ / -- + ⁇ ⁇ ⁇ ) ⁇ ; wherein, - is the attenuation coefficient of the OSEM-B attenuation image, and is the attenuation coefficient of the FBP attenuation image, "is a weighting parameter.
  • the following module is used to obtain the ":”:
  • An acquisition module configured to obtain chord data of the noise scan of the transmission device before the current PET device is shipped from the factory, and perform FBP reconstruction on the chord data of the noise scan to obtain a second FBP attenuation image;
  • An extraction module configured to extract a first tissue region of the effective attenuation image and a second tissue region of the second FBP attenuation image
  • a calculation module configured to calculate the histogram using the first tissue region and the second tissue region.
  • the calculation module is specifically used to calculate a formula using a minimum mean square error mm Hlst Final ⁇ Hist Lang
  • the OSEM-B reconstruction module specifically includes:
  • a first obtaining submodule configured to acquire the stated maximum expected value OSEM method using an ordered subset
  • ⁇ " ⁇ is the attenuation coefficient of water.
  • the embodiment of the present invention not only adopts the FBP method.
  • Transmission scans the chord data for a short period of time, and also uses OSEM-B to reconstruct the scanned chord data of the short-term transmission, and weights the reconstruction results of the two, and finally obtains the effective attenuation coefficient and then Emission.
  • the chord data is scanned for the attenuation correction
  • the quality of the transmission is improved for a long time, and the quality of the transmission is reduced.
  • the transmission scan time of the transmission is only required to be performed in the first embodiment of the present invention.
  • the accuracy of the final attenuation corrected image can be adjusted to the accuracy of the 8-minute Transmission scan using the segmentation attenuation correction in the prior art.
  • FIG. 1 is a flow chart of an embodiment of a method of the present invention
  • step 103 is a flow chart of step 103 in the method embodiment of the present invention.
  • Figure 4 is a schematic diagram of the image quality curve of the NEMA test
  • Figure 5 is a schematic diagram of the lung error rate curve of the NEMA test
  • Figure 6 is a schematic structural view of an embodiment of the device of the present invention.
  • FIG. 7 is a schematic structural diagram of an OSEM-B reconstruction module 602 according to an embodiment of the present invention
  • FIG. 8 is a schematic structural diagram of calculating a weighting parameter in an embodiment of the apparatus of the present invention.
  • a flowchart of a positive method embodiment may include the following steps:
  • Step 101 Acquire transmission transmission chord data of the current PET device.
  • the embodiment of the present invention needs to perform attenuation correction for its transmission scan string data.
  • the transmission scan string data is obtained.
  • the scan chord data in this step can be obtained by performing a Transmission scan in a short time. For example, you can only perform a 4 minute Transmission sweep 4 .
  • Step 102 respectively reconstruct the transmission scan chord data by using the Bayesian model-based ordered subset maximum expected value OSEM-B method and the filtered back projection FBP method to obtain the OSEM-B attenuation image and the first FBP, respectively. Attenuate the image.
  • the transmission scan chord data is not only subjected to the filtered back projection (FBP) method, but also the maximum expected value of the ordered subset based on the Bayesian model is needed (The Ordered Subset Expectation Maximization- Bayesian, OSEM-B) method is used to reconstruct it.
  • FBP filtered back projection
  • OSEM-B Ordered Subset Expectation Maximization- Bayesian
  • OSEM-B is an iterative method often used in Emission reconstruction or Transmission reconstruction. This method can reconstruct accurate and uniform attenuation images by forward and backward projection combined with Bayesian model constraints. Referring to FIG. 2, in this step, the transmission scan chord data is reconstructed by using the OSEM-B method, which may include the following steps:
  • Step 201 Obtain an initial attenuation coefficient of an initial attenuation image of the Transmission scan chord data by using an OSEM method.
  • the OSEM-B algorithm introduced in this embodiment is divided into two parts. First, the calculation of the OSEM method is performed, and then the equation (1) is shown:
  • the initial attenuation coefficient of the obtained initial attenuation image is represented, and y represents the attenuation amount on the response line of a certain negation, indicating the length of a certain response line passing through a certain pixel, that is, the response line i and the pixel
  • the intersection length of 1; s represents the angle subset, and the subscript J represents which response line and which pixel respectively.
  • the response line here refers to the connection between a pair of pixels.
  • the image pixel value of a response line is reduced from 100 to 90, and the attenuation on the response line is 1-00; Is a subset of the image in various directions, for example, the image has a projection in 256 directions, divided into 8 subsets, then a subset has 32 projections.
  • the iterative result of the reconstruction is performed last time, and y can be the measured value, which is the system matrix.
  • Step 202 The initial subtraction coefficient and the Bayesian model are according to a formula
  • Mj ⁇ [ ⁇ j) j + ⁇ 3 ⁇ 4 ⁇ «ter is combined to obtain an attenuation coefficient of the OSEM-B attenuation image; wherein, ⁇ is the attenuation coefficient of water.
  • the Bayesian model is as shown in equation (2):
  • Equation (2) is a model similar to the form of Gaussian function, where "[01] is a parameter, it can be seen that when ⁇ 3 ⁇ 4 When approaching ⁇ " ⁇ , ° is getting closer to ".
  • the PET device has a photon energy of 511KeV.
  • the attenuation values of other parts are very close to water.
  • the OSEM can be combined with the Bayesian model according to formula (3):
  • Equation (3) obtains the attenuation coefficient of the OSEM-B attenuation image reconstructed by the OSEM-B algorithm, but it cannot be directly used for attenuation correction. Because the attenuation value of the tissue part of the OSEM-B attenuation image is too smooth, and the normal human body will fluctuate, but the attenuation value calculated by the formula (3) does not fluctuate, which loses many images due to system noise and correction algorithms. detail. In this way, the image noise distribution of the Transmission scan and the Emission scan does not match. If the attenuation correction of the Emission scan string data is directly performed by the formula (3), the system noise is amplified.
  • the chord data collected by the PET equipment there is more noise in the chord data collected by the PET equipment, and the chord data is not particularly uniform, and can be considered as a system characteristic. If the smoothing of the transmission and Emission string data is inconsistent after the smoothing of the formula (3), the direct use of the attenuation correction may result in inaccurate image or artifacts. Therefore, the weighting step of step 103 is performed. In order to reduce the introduction of system noise, the attenuation coefficient is more realistic.
  • Step 103 Weight the O SEM-B attenuation image and the first FBP attenuation image to obtain an effective attenuation image.
  • the effective attenuation coefficient of the effective attenuation image is calculated by using the weighted form of the OSEM-B and the FBP reconstruction result in the formula (4);
  • ⁇ ⁇ - ⁇ + (1 - ⁇ ) ⁇ ⁇ ( 4 )
  • is the attenuation coefficient of the OSEM-B attenuation image, which is the attenuation coefficient of the FBP attenuation image
  • the value range of "is greater than 0 and less than 1 because FBP reconstructed FBP attenuation image has more noise, preferably, in order not to introduce too much noise, it can be closer
  • the method for obtaining the method may include the following steps: Step 301: Acquire chord data of the inherent noise scan of the transmission device before leaving the factory, and scan the chord of the inherent noise of the device.
  • the map data is subjected to FBP reconstruction to obtain a second FBP attenuation image.
  • the embodiment of the present invention provides an adaptive selection method as shown below. Because PET equipment is more complicated, the distribution of noise in each PET system is different. Therefore, before each PET is shipped, a long-term transmission equipment inherent noise scan is performed, and the string data of the inherent noise of the transmission equipment is used. , and By performing FBP reconstruction on the chord data of the inherent noise scan of the device, a second FBP attenuation image can be obtained.
  • Step 302 Extract a first tissue region of the effective attenuation image and a second tissue region of the second FBP attenuation image. Then, each time the short-term transmission scan of step 101 is performed, the tissue region can be first segmented from the effective attenuation image and the second FBP attenuation image.
  • Step 303 Calculate the a with the histogram of the first tissue region and the second tissue region and then separately calculate the histograms ⁇ , mecanic, , and ⁇ ⁇ of the tissue regions in the two images.
  • the Hist Fmai is a histogram of the first tissue region
  • the histogram is the histogram of the second tissue region.
  • Equation (5) uses the least mean square error fitting method because Hist Final in equation (5) and ⁇ ' in equation (4) are parameters about effective attenuation of the image, so formula (4) and formula ( 5) Combine to eliminate the feature of the effective attenuation image, and then find the value of ".
  • Step 104 Perform attenuation correction on the transmitted Emission scan chord data of the current PET device by using an attenuation chord generated according to the effective attenuation image.
  • the effective attenuation coefficient of the effective attenuation image in step 103 it may be forward projected to generate an effective attenuation chord pattern, and then use the effective attenuation chord diagram to perform attenuation correction on the emission (Emission) scanning chord data of the PET device. .
  • the embodiment of the present invention weights the OSEM-B attenuation image and the FBP attenuation image, and Attenuation correction is performed by taking the effective attenuation coefficient ⁇ ⁇ 1 obtained after the weighting, so that the time for performing the transmission scan even in step 101 is short (the prior art is 8 minutes, and the embodiment of the present invention can be only 4 minutes). It can also make the system noise of the attenuation-corrected Emission scan chord data smaller, and the image precision is also close to the image precision of the 8-minute Transmission scan in the prior art. It can be seen that the embodiment of the present invention can achieve the purpose of obtaining a result that is very similar to the long-term transmission scan by using the transmission scanning chord data of a short time by weighting.
  • FIG. 4 is a schematic diagram of a NEMA image quality curve.
  • the abscissa represents the transmission scan time
  • the ordinate represents the mass value
  • the solid line represents the curve of the embodiment of the present invention
  • the broken line represents the SAC method in the prior art.
  • the curve, d in the figure represents the diameter of the heat source of the meninges, and gives six curves with d of 10, 13, 17, 22, 28, and 37, respectively.
  • the diameter of the heat source of the meninges is larger than the resolution of PET, the image quality reconstructed from the data of the shorter scan time using the embodiment of the present invention is superior to the result of SAC.
  • Fig. 5 is a schematic diagram of the lung error rate curve of NEMA. There is no drug distribution in the lungs, which is a cold zone, and the calculation method according to the embodiment of the present invention shows that the lower the curve of the non-error rate, the better the effect.
  • the abscissa represents the transmission scan time, and the ordinate represents the lung error rate.
  • the lungs when the attenuation corrected image slice numbers are 27, 30, 33, 36, and 39 are respectively shown. Error rate curve. It can be seen from FIG.
  • the present invention can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases, the former is better.
  • Implementation Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium shield including a plurality of instructions for making A computer device (which may be a personal computer, a server, or a network device, etc.) performs various embodiments of the present invention All or part of the steps of the method.
  • the foregoing storage medium includes: a medium that can store program codes, such as a read only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
  • a medium that can store program codes such as a read only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
  • the embodiment of the present invention further provides an attenuation correction device for a medical image in a PET system.
  • FIG. 6 a schematic structural diagram of an embodiment of an attenuation correction apparatus for a medical image in a PET system is shown, which may include:
  • the OSEM-B reconstruction module 602 is configured to reconstruct the transmission scan chord data by using an OSEM-B method based on the ordered subset maximum expected value of the Bayesian model to obtain an OSEM-B attenuation image;
  • the OSEM-B reconstruction module 602 may specifically include: a first obtaining sub-module 701, configured to acquire the transmission short-time scanning chord data by using an ordered subset maximum expected value OSEM method.
  • a second obtaining sub-module 702 configured to combine the initial attenuation coefficient and the Bayesian model a according to the formula ⁇ 2 - ) ⁇ ⁇ ⁇ - ⁇ + ⁇ water to obtain
  • 0SEM-B attenuates the attenuation coefficient of the image; wherein, ⁇ is the attenuation coefficient of water.
  • the FBP reconstruction module 603 is configured to perform reconstruction on the transmission scan chord data by using a filtered back projection FBP method to obtain a first FBP attenuation image; and a weighting calculation module 604, configured to use the OSEM-B attenuation image and the FBP attenuation image Performing a weighting to obtain a valid attenuation image; the weighting calculation module 404 can be specifically configured to:
  • the structure diagram is calculated in an actual application, and includes: an obtaining module 801, configured to acquire chord data of the inherent noise scan of the transmission device before leaving the factory, and the noise The scanned chord data is subjected to FBP reconstruction to obtain a second FBP attenuation image;
  • An extraction module 802 configured to extract a first tissue region of the effective attenuation image and a second tissue region of the second FBP attenuation image;
  • the calculating module 803 is configured to calculate the using the histogram of the first tissue region and the second tissue region.
  • the calculation module 803 can be specifically used to calculate a minimum mean square error
  • the attenuation correction module 605 is configured to perform attenuation correction on the transmitted Emission scan chord data of the current PET device by using the attenuation chord generated according to the effective attenuation image.
  • the attenuation correction device of the embodiment of the present invention may weight the OSEM-B attenuation image and the FBP attenuation image by the weighting calculation module 604, and perform attenuation correction by using the effective attenuation coefficient ⁇ ⁇ 1 obtained after the weighting, so that even if it is performed
  • the transmission scan time is very short (the prior art is 8 minutes, and the embodiment of the present invention can only be 4 minutes), and the system noise of the attenuation-corrected Emission scan string data can be made smaller, and the image precision is also existing.
  • the accuracy of the image for the 8-minute Transmission scan in the technology is very close. It can be seen that the embodiment of the present invention can achieve the purpose of obtaining a result closely related to the long-term transmission scan by using the transmission scanning chord data of 4 ⁇ short time by weighting.
  • the present invention is applicable to a wide variety of general purpose or special purpose computing system environments or configurations.
  • personal computer server computer, handheld or portable device, tablet device, multiprocessor system, microprocessor based system, set-top box, programmable consumer electronics device, network PC, small computer, mainframe computer, including Distributed computing environment for any of the above systems or devices and many more.
  • the invention may be described in the general context of computer-executable instructions executed by a computer, such as a program module.
  • program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types.
  • the invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network.
  • program modules can be located in both local and remote computer storage media including storage devices.
  • the device embodiment since it basically corresponds to the method embodiment, it can be referred to the partial description of the method embodiment.
  • the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, ie may be located One place, or it can be distributed to multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the objectives of the solution of the embodiment. Those skilled in the art can understand and implement without any creative work.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Nuclear Medicine (AREA)

Abstract

一种PET系统中图像的衰减校正方法及装置,所述方法包括:获取当前PET设备的透射扫描弦图数据;对透射扫描弦图数据釆用基于贝叶斯模型的有序子集最大期望值OSEM-B方法以及滤波反投影FBP方法进行重建,分别得到OSEM-B衰减图像和第一FBP衰减图像;将OSEM-B衰减图像和第一FBP衰减图像进行加权以获取有效衰减图像;釆用依据所述有效衰减图像生成的衰减弦图对所述当前PET设备的发射扫描弦图数据进行衰减校正。本发明实施例公开的方法和装置,可以使得衰减图像质量达到较长时间透射扫描的质量精度,同时减少透射扫描的时间。

Description

PET系统中图像的衰减校正方法及装置 技术领域
本发明涉及医学图像处理领域, 特别是涉及一种应用于 PET系统中医 学图像的衰减校正方法及装置。
背景技术
在 PET ( Positron Emission Computed Tomography, 正电子发射型计算 机断层显像)设备扫描过程中, 每个患者一般需要进行两次扫描, 一次是 发射 ( Emissio ) 扫描, 另一次是透射 ( Transmission )扫描。 Emission扫 描釆集到的数据大体上反应了药物在患者体内的分布情况, 但是没有经过 衰减校正, 所以存在定量不准确的问题; Transmission扫描采集到的数据是 专门用于生成衰减弦图, 以对 Emission扫描得到的数据进行衰减校正的。 使用 PET设备进行心脏检查时, Emission的扫描数据可以反映药在患者体 内的分布, 扫描大约耗时 4分钟; 而 Transmission扫描则会大约耗时 30分 钟, 这样才能得到相对准确的扫描图像。
在现有技术的衰减校正方法中, 为了减少 Transmission扫描的时间却 又得到较为准确的图像, 采用了分割衰减校正算法 ( SAC , Segmented Attenuation Correction ),该方法先† Transmissio 的扫 4 数据进行滤波反投 影 (FBP ) 重建, 再将 FBP的重建图像按照图像像素值的大小划分为組织 区域、 肺区域、 床板区域和空气区域, 这四个区域的四种密度不同, 也表 示四种不同的衰减系数, 对各区域的像素赋对应的值, 之后再对赋值后的 图像进行正向投影, 利用得到的衰减弦图对 Emission的扫描数据进行衰减 校正。
但是,采用上述的衰减校正方法还是会导致一个床位需要 10分钟以上 的时间, 而如果对患者进行全身扫描则需要一个小时以上, 还是存在针对 PET设备扫描的医学图像的衰减校正的效率低下的问题, 因此, 如何提出 一种创新的医学图像的衰减校正方法, 在减少 Transmission扫描时间的情 况下, 也能够保证衰减校正后的图像精度, 进而提升设备的投入产出比, 并减轻患者因扫描带来的心里和生理方面的不适感。 发明内容
本发明所要解决的技术问题是, 提供一种 PET 系统中医学图像的衰减 校正方法, 以保证在减少 Transmi ss ion扫描时间的情况下, 也能够保证衰 减校正后的图像精度。
本发明的另一个目的是将上述构思应用于具体的应用环境中, 提供一 种 PET系统中医学图像的衰减校正装置, 从而保证该方法的实现和应用。
为解决上述技术问题, 本发明实施例提供了一种 PET系统中图像的衰 减校正方法, 包括:
获取当前 PET设备的透射 Transmission扫描弦图数据;
对所述 Transmission扫描弦图数据采用基于贝叶斯模型的有序子集最 大期望值 OSEM-B方法以及滤波反投影 FBP方法进行重建, 分别得到 OSEM-B衰减图像和第一 FBP衰减图像;
将所述 OSEM-B衰减图像和第一 FBP衰减图像进行加权以获取有效衰 减图像;
采用依据所述有效衰减图像生成的衰减弦图对所述当前 PET设备的发 射 Emission扫描弦图数据进行衰减校正。
优选的,所述将所述 OSEM-B衰减图像和第一 FBP衰减图像进行加权以 获取有效衰减图像, 具体包括:
, .Final _ OSEM-B ( FBP
采用公式^ / 计算所述有效衰减图像的 有效衰减系数; 其中, ^£ΛΜ为 OSEM-B衰减图像的衰减系数, 为 FBP衰减图像的 衰减系数, "为加权参数。 优选的, 所述"的获取方式具体为:
获取所述当前 PET设备出厂前进行 Transmission设备固有噪声扫描的弦 图数据, 并对所述噪声扫描的弦图数据进行 FBP重建, 以得到第二 FBP衰减 图像;
提取所述有效衰减图像的第一组织区域以及所述笫二 FBP衰减图像的 第二组织区域;
利用所述第一组织区域和第二组织区域的直方图计算所述 a
优选的, 所述利用所述第一組织区域和第二组织区域的直方图计算所 述"具体为: 中,
Figure imgf000005_0001
区域的直方图。
优选的, 所述对所述 Transmission扫描弦图数据采用基于贝叶斯模型的 有序子集最大期望值 OSEM-B方法进行重建, 具体包括:
采用有序子集最大期望值 OSEM方法获取所述 Transmission扫描弦图数 据的初始衰减图像的初始衰减系数 μη ραΛ
将所述初始衰减 系数 μ】 - 与 贝 叶斯模型 按照公式
. . new \ . . new part , , ,
j = Uj ) j ~ 进行结合, 以获取 OSEM-B衰减图像 的衰减系数;
其中, 所述 w。^为水的衰减系数。
本发明还提供了一种 PET系统中图像的衰减校正装置, 包括: 获取源数据模块, 用于获取当前 PET设备的透射 Transmission扫描弦图 数据;
OSEM-B重建模块, 用于对所述 Transmission扫描弦图数据采用基于贝 叶斯模型的有序子集最大期望值 OSEM-B方法进行重建,得到 OSEM-B衰减 图像;
FBP重建模块, 用于对所述 Transmission扫描弦图数据釆用滤波反投 影 FBP方法进行重建, 得到第一 FBP衰减图像;
加权计算模块, 用于将所述 OSEM-B衰减图像和第一 FBP衰减图像进 行加权以获取有效衰减图像;
衰减校正模块, 用于采用依据所述有效衰减图像生成的衰减弦图对所 述当前 PET设备的发射 Emission扫描弦图数据进行衰减校正。 优选的, 所述加权计算模块, 具体用于:
..Final _ OSEM-B ( 、 FBP
采用公式^ / — + ^ ~ θί)μ 计算所述有效衰减图像的 有效衰减系数; 其中, — 为 OSEM-B衰减图像的衰减系数, 为 FBP衰减图像 的衰减系数, "为加权参数。
优选的, 采用如下模块获取所述 ":
获取模块, 用于获取所述当前 PET设备出厂前进行 Transmission设备固 有噪声扫描的弦图数据, 并对所述噪声扫描的弦图数据进行 FBP重建, 以 得到第二 FBP衰减图像;
提取模块, 用于提取所述有效衰减图像的第一组织区域以及所述第二 FBP衰减图像的第二组织区域;
计算模块, 用于利用所述第一组织区域和第二组织区域的直方图计算 所述 。
优选的, 所述计算模块, 具体用于采用最小均方误差的计算公式 mm Hlst Final ~ Hist Lang
计算", 其中, 所述 Hist^为所述第一组织区 域的直方图, 所述 Ηί 为所述第二組织区域的直方图。
优选的, 所述 OSEM-B重建模块, 具体包括:
第一获取子模块, 用于采用有序子集最大期望值 OSEM方法获取所述
, ^ 、 - .new part
Transmission短时扫描弦图数据的初始衰减图像的初始哀减系数 ^ ― ; 第二获取子模块, 用于将所述初始衰减系数 与贝叶斯模型 a)
. .new 1 -, \ . .new part
按照公式 ^ = {[ ~ aj ) j ― + · ^。 进行结合, 以获取 0SEM-B 衰减图像的衰减系数;
其中, 所述^" ^为水的衰减系数。
从上述的技术方案可以看出, 本发明实施例不仅采用 FBP方法对 Transmission短时间内的扫描弦图数据进行重建, 还采用 OSEM-B对 Transmission短时间内的扫描弦图数据进行重建, 并将两者的重建结果做加 权和, 最终得到的有效衰减系数再对 Emission扫描弦图数据进行衰减校正 时, 在同时使得衰减图像质量达到较长时间进行 Transmission扫描的质量精 度, 可以减少 Transmission的扫描时间, 例如本发明实施例一开始只需要执 行 4分钟的 Transmission扫描, 即可使最终衰减校正后的图像精度达到现有 技术中采用分割衰减校正进行 8分钟 Transmission扫描的精度。 附图说明 为了更清楚地说明本申请实施例或现有技术中的技术方案, 下面将对 实施例或现有技术描述中所需要使用的附图作简单地介绍, 显而易见地, 下面描述中的附图仅仅是本申请中记载的一些实施例, 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 他的附图。
图 1为本发明方法实施例的流程图;
图 2为本发明方法实施例中 0SEM-B重建的流程图;
图 3为本发明方法实施例中步骤 103的流程图;
图 4为 NEMA测试的图像质量曲线示意图;
图 5为 NEMA测试的肺错误率曲线示意图;
图 6为本发明装置实施例的结构示意图;
图 7为本发明装置实施例中 OSEM-B重建模块 602的结构示意图; 图 8为本本发明装置实施例中计算加权参数"的结构示意图。 具体实施方式
为了使本技术领域的人员更好地理解本发明方案, 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做出创造性劳动前提 下所获得的所有其他实施例, 都属于本发明保护的范围。
参见图 1, 示出了本发明的一种应用于 PET系统中医学图像的衰减校 正方法实施例的流程图, 可以包括以下步骤:
步骤 101 : 获取当前 PET设备的透射 Transmission扫描弦图数据。 针对一台 PET设备, 本发明实施例需要针对其透射( Transmission )扫 描弦图数据进行衰减校正,首先在该 PET设备进行 Transmission扫描之后, 获取其 Transmission扫描弦图数据。 在本实施例中, 本步骤中的扫描弦图 数据可以是很短时间内进行 Transmission扫描得到的。 例如, 可以只进行 4 分钟的 Transmission扫 4 即可。
步骤 102:分别对所述 Transmission扫描弦图数据釆用基于贝叶斯模型 的有序子集最大期望值 OSEM-B方法以及滤波反投影 FBP方法进行重建, 分别得到 OSEM-B衰减图像和第一 FBP衰减图像。
本实施例与现有技术不同的是, 本实施例中对 Transmission扫描弦图数 据不仅要进行滤波反投影 (FBP ) 方法进行重建, 还需要采用基于贝叶斯 模型的有序子集最大期望值 ( Ordered Subset Expectation Maximization- Bayesian, OSEM-B ) 方法对其进行重建。
其中, OSEM-B是一种在 Emission重建或者 Transmission重建时经常使 用的迭代方法, 该方法分别通过前向和后向投影并结合贝叶斯模型约束, 可以重建出准确而均匀的衰减图像。参考图 2所示,本步驟中对 Transmission 扫描弦图数据采用 OSEM-B方法进行重建, 具体可以包括以下步驟:
步骤 201 :采用 OSEM方法获取所述 Transmission扫描弦图数据的初始衰 减图像的初始衰减系数 。
本实施例中所介绍的 OSEM-B算法分为两个部分,首先进行 OSEM方法 的计算, 再与贝 式 ( 1 ) 所示:
Figure imgf000008_0001
其中, 表示得到的初始衰减图像的初始衰减系数, y表示某奈响应线 上的衰减量, 表示某条响应线穿过某个像素的长度, 即是响应线 i与像素 1的相交长度; s表示角度子集, 下标 J分别表示哪条响应线和哪个像素。 这里的响应线指的是一对像素之间的连线, 例如一条响应线穿过的图像像 素值从 100减少为 90, 那该条响应线上的衰减量就是 1-00; 角度子集指的 是图像在各个方向上的子集, 例如图像有 256个方向上的投影, 共分 8个 子集, 那么一个子集就有 32个投影。 此外, 在公式 ( 1 ) 中, 上一次 进行重建的迭代结果, 而 y可以是测量值, 是系统矩阵。
, .new part ,
步骤 202: 将所述初始衮减系数 与贝叶斯模型 按照公式
, , new 1 \ fi new part . „, . ,
Mj = ^[~ j) j + <¾ ^«ter进行结合, 以获取 OSEM-B衰减图 像的衰减系数; 其中, 所述 ^为水的衰减系数。
在本实施例中, 贝叶斯模型如公式 (2) 所示:
a j = 0 exp[-(/;w - f I β2] (2) 公式 (2)是一个类似高斯函数形式的模型, 其中, " [01]是一个参 数, 可以看出, 当^¾越接近^" ^时, °·越趋近于"。。
而 PET设备的光子能量为 511KeV,其衰减图中除了肺部和骨骼之外, 其他部分的衰减值与水非常接近, OSEM可以与贝叶斯模型按照公式 (3) 结合起来:
公式 (3)得到的就是 OSEM-B算法重建出的 OSEM-B衰减图像的衰 减系数, 但是不能直接用于衰减校正。 因为 OSEM-B衰减图像组织部分的 衰减值过于平滑, 而正常的人体都会有波动, 但是公式(3)算出来的衰减 值没有波动,这就失去了很多由于系统噪声和校正算法所产生的图像细节。 这样一来, Transmission扫描和 Emission扫描的图像噪声分布就不匹配, 如果用公式 (3) 直接进行 Emission扫描弦图数据的衰减校正, 就会放大 系统噪声。 实际上, PET设备采集的弦图数据中噪声比较多, 弦图数据本来就不 会特别均匀, 可以认为是系统特性。 如果经过公式(3 )过度的平滑处理之 后, 会造成 Transmission和 Emission弦图数据的分布不一致, 直接用于衰 减校正可能会使图像定量不准或者出现伪影, 所以要执行步骤 103 的加权 步骤, 以减少系统噪声的引入, 使得衰减系数更为真实。
步骤 103: 将所述 O SEM-B衰减图像和第一 FBP衰减图像进行加权以 获取有效衰减图像。
本步骤釆用公式(4 ) 中 OSEM-B与 FBP重建结果加权的形式计算所 述有效衰减图像的有效衰减系数;
μΜηαί = αμ醒- Β + (1 - α)μΡΒΡ ( 4 ) 其中, μ 为 OSEM-B衰减图像的衰减系数, 为 FBP衰减图像 的衰减系数, "为加权参数, 即为有效衰减系数。 其中, "的取值范围为大于 0且小于 1, 因为 FBP重建的 FBP衰减图 像的噪声较多, 优选情况下, 为了不引入太多噪声, 可以使 更贴近
. .OSEM-B ^
〃 , 所以 a可以取靠近 1的小数。
其中, 参考图 3 , 所述"的获取方式具体可以包括以下步骤: 步骤 301 : 获取所述当前 PET设备出厂前进行 Transmission设备固有 噪声扫描的弦图数据, 并对所述设备固有噪声扫描的弦图数据进行 FBP重 建, 以得到第二 FBP衰减图像。 对于加权过程中的参数"的选择, 本发明实施例提供了一种如下所示 的自适应选择方法。 因为 PET设备比较复杂, 每一台的 PET系统噪声的分 布都有区别,所以,在每台 PET出厂前,都会进行一次长时间的 Transmission 设备固有噪声扫描, 利用 Transmission设备固有噪声扫描的弦图数据, 并 对设备固有噪声扫描的弦图数据进行 FBP重建,可以得到第二 FBP衰减图 像。
步骤 302: 提取所述有效衰减图像的第一组织区域以及所述第二 FBP 衰减图像的第二组织区域。 那么每次在进行步骤 101 的短时间 Transmission扫描时, 就可以先将 组织区域从有效衰减图像和第二 FBP衰减图像中分割出来。 步驟 303 : 利用所述第一组织区域和第二组织区域的直方图计算所述 a 然后分别统计两个图像中组织区域的直方图 ^,„。,和^ ^。 因为图像 中一个区域的噪声比较多时, 直方图就会比较宽, 所以为了使最后的有效 衰减图像和第二 FBP衰减图像更接近, 所以建立公式( 5 ) 所示的目标函 数: a - mm Hist Final ~ Hist Long ( 5 ) 其中, 所 i^ HistFmai为所述第一组织区域的直方图, 所述 为所述第 二组织区域的直方图。
公式 ( 5 ) 利用的是最小均方误差拟合法, 因为公式( 5 ) 中的 HistFinal 和公式(4 ) 中的 ηα'都是关于有效衰减图像的参数, 所以将公式 (4 ) 和 公式 (5 ) 进行结合可以消去有效衰减图像的特征, 即可求出"的值。
步驟 104:采用依据所述有效衰减图像生成的衰减弦图对所述当前 PET 设备的发射 Emission扫描弦图数据进行衰减校正。
在步骤 103得到有效衰减图像的有效衰减系数之后, 可以对其进行正 向投影以生成有效衰减弦图, 再使用该有效衰减弦图对上述 PET设备的发 射 (Emission ) 扫描弦图数据进行衰减校正。
本发明实施例通过将 OSEM-B衰减图像和 FBP衰减图像进行加权, 并釆 取加权之后获得的有效衰减系数 μ ηα1进行衰减校正, 这样可以使得即便是在 步骤 101中进行 Transmission扫描的时间很短 (现有技术是 8分钟, 而本发明 实施例可以仅为 4分钟) , 也能使衰减校正后的 Emission扫描弦图数据的系 统噪声更小, 图像精度也与现有技术中进行 8分钟 Transmission扫描的图像精 度很接近。 可见, 本发明实施例可以通过加权的方式, 达到使用很短时间的 Transmission扫描弦图数据就能得到和长时间 Transmission扫描极为近似的结 果的目的。
为了验证本发明实施例相对于现有技术中 SAC方法的优势, 下面给出分 别采用相同时间的 Transmission 扫描弦图数据对本方法与 SAC 方法进行 NEMA (美国电气制造商协会) 测试的效果示意图。 其中, 图 4为 NEMA图 像质量曲线示意图, 在图 4中, 横坐标代表 Transmission扫描时间, 纵坐标代 表质量值, 实线代表本发明实施例的曲线, 而虚线则是现有技术中的 SAC方 法的曲线, 图中的 d表示脑膜体的热源的直径, 分别给出了 d为 10、 13、 17、 22、 28和 37的六个曲线。从图 4中可以看出,当脑膜体的热源的直径大于 PET 的分辨率时,使用本发明实施例对较短扫描时间的数据重建出的图像质量优于 SAC的结果。
图 5为 NEMA的肺错误率曲线示意图, 在肺部没有药品分布, 为冷区, 并且根据本发明实施例的计算方式可知非错误率的曲线越低则说明效果越好。 而在图 5中, 横坐标代表 Transmission扫描时间, 纵坐标为肺错误率, 图 5中 分別给出了衰减校正后的图像切片序号(slice )为 27、 30、 33、 36和 39时的 肺错误率曲线。 可以从图 5中看出, 使用本发明实施例进行衰减校正, 可以看 出肺错误率大幅度低于现有技术中的 SAC方法, 200s的数据已经明显优于现 有技术的 SAC方法进行 Transmission长时间扫描的数据。
通过以上的方法实施例的描述, 所属领域的技术人员可以清楚地了解到 本发明可借助软件加必需的通用硬件平台的方式来实现, 当然也可以通过硬 件, 但很多情况下前者是更佳的实施方式。基于这样的理解, 本发明的技术方 案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来, 该计算机软件产品存储在一个存储介盾中,包括若干指令用以使得一台计算机 设备(可以是个人计算机, 服务器, 或者网络设备等)执行本发明各个实施例 所述方法的全部或部分步骤。 而前述的存储介质包括: 只读存储器(ROM )、 随机存取存储器( RAM )、 磁碟或者光盘等各种可以存储程序代码的介质。 相应于上面的方法实施例, 本发明实施例还提供一种 PET系统中医学 图像的衰减校正装置。 参见图 6, 示出了一种 PET系统中医学图像的衰减 校正装置实施例的结构示意图, 可以包括:
获取源数据模块 601, 用于获取当前 PET设备的透射 Transmission扫 描弦图数据;
OSEM-B重建模块 602,用于对所述 Transmission扫描弦图数据釆用基 于贝叶斯模型的有序子集最大期望值 OSEM-B 方法进行重建, 得到 OSEM-B衰减图像;
其中, 所述 OSEM-B重建模块 602, 参考图 7所示, 具体可以包括: 第一获取子模块 701 , 用于采用有序子集最大期望值 OSEM方法获取所 述 Transmission短时扫描弦图数据的初始衰减图像的初始衰减系数
- .new _ part
〃7 一 ;
part
第二获取子模块 702,用于将所述初始衰减系数 与贝叶斯模型 a〗按照公式 μ 二 0— )μ^η -ραη + ^water进行结合, 以获取
0SEM-B衰减图像的衰减系数; 其中, 所述 μ 为水的衰减系数。
FBP重建模块 603 ,用于对所述 Transmission扫描弦图数据采用滤波反 投影 FBP方法进行重建, 得到第一 FBP衰减图像; 加权计算模块 604 , 用于将所述 OSEM-B衰减图像和 FBP衰减图像进 行加权以获取有效衰减图像; 所述加权计算模块 404 , 具体可以用于:
..Final _ OSEM-B ι FBP
釆用公式 — + ί-^)μ 计算所述有效衰减图像的 有效衰减系数; 其中, ^ ^为 OSEM-B衰减图像的衰减系数, ™3为 FBP 衰减图像的衰减系数, "为加权参数。
参考图 8所示, 是在实际应用中计算所述 的结构示意图, 包括: 获取模块 801 , 用于获取所述当前 PET设备出厂前进行 Transmission 设备固有噪声扫描的弦图数据, 并对所述噪声扫描的弦图数据进行 FBP重 建, 以得到第二 FBP衰减图像;
提取模块 802 ,用于提取所述有效衰减图像的第一组织区域以及所述第 二 FBP衰减图像的笫二组织区域;
计算模块 803 ,用于利用所述第一组织区域和第二组织区域的直方图计 算所述 。
所述计算模块 803, 具体可以用于采用最小均方误差的计算公式
"=ι ίη||^¾„。/ - ' ¾J|计算", 其中, 所述 H^F 为所述第一组织区 域的直方图, 所述 His 为所述第二組织区域的直方图。
衰减校正模块 605 , 用于采用依据所述有效衰减图像生成的衰减弦图对所 述当前 PET设备的发射 Emission扫描弦图数据进行衰减校正。
本发明实施例的衰减校正装置, 可以通过加权计算模块 604将 OSEM-B 衰减图像和 FBP 衰减图像进行加权, 并釆取加权之后获得的有效衰减系数 μΡίηα1进行衰减校正, 这样可以使得即便是进行 Transmission扫描的时间很短 (现有技术是 8分钟, 而本发明实施例可以仅为 4分钟 ), 也能使衰减校正后 的 Emission扫描弦图数据的系统噪声更小, 图像精度也与现有技术中进行 8 分钟 Transmission扫描的图像精度很接近。 可见, 本发明实施例可以通过加权 的方式, 达到使用 4艮短时间的 Transmission扫描弦图数据就能得到和长时间 Transmission扫描极为近似的结果的目的。
可以理解的是, 本发明可用于众多通用或专用的计算系统环境或配置中。 例如: 个人计算机、 服务器计算机、 手持设备或便携式设备、 平板型设备、 多 处理器系统、 基于微处理器的系统、 置顶盒、 可编程的消费电子设备、 网络 PC、 小型计算机、 大型计算机、 包括以上任何系统或设备的分布式计算环境 等等。
本发明可以在由计算机执行的计算机可执行指令的一般上下文中描述,例 如程序模块。一般地,程序模块包括执行特定任务或实现特定抽象数据类型的 例程、 程序、 对象、 组件、 数据结构等等。 也可以在分布式计算环境中实践本 发明,在这些分布式计算环境中, 由通过通信网络而被连接的远程处理设备来 执行任务。在分布式计算环境中,程序模块可以位于包括存储设备在内的本地 和远程计算机存储介质中。
需要说明的是, 在本文中,诸如第一和第二等之类的关系术语仅仅用来将 一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些 实体或操作之间存在任何这种实际的关系或者顺序。 而且, 术语 "包括',、 "包 含"或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素 的过程、 方法、 物品或者设备不仅包括那些要素, 而且还包括没有明确列出的 其他要素, 或者是还包括为这种过程、 方法、 物品或者设备所固有的要素。 在 没有更多限制的情况下, 由语句 "包括一个…… " 限定的要素, 并不排除在包 括所述要素的过程、 方法、 物品或者设备中还存在另外的相同要素。
对于装置实施例而言, 由于其基本对应于方法实施例, 所以相关之处 参见方法实施例的部分说明即可。 以上所描述的装置实施例仅仅是示意性 的, 其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开 的, 作为单元显示的部件可以是或者也可以不是物理单元, 即可以位于一 个地方, 或者也可以分布到多个网络单元上。 可以根据实际的需要选择其 中的部分或者全部模块来实现本实施例方案的目的。 本领域普通技术人员 在不付出创造性劳动的情况下, 即可以理解并实施。
以上所述仅是本发明的具体实施方式, 应当指出, 对于本技术领域的 普通技术人员来说, 在不脱离本发明原理的前提下, 还可以做出若干改进 和润饰, 这些改进和润饰也应视为本发明的保护范围。

Claims

权 利 要 求
1、 一种 PET系统中图像的衰减校正方法, 其特征在于, 包括: 获取当前 PET设备的透射 Transmission扫描弦图数据;
对所述 Transmission扫描弦图数据采用基于贝叶斯模型的有序子集最 大期望值 OSEM-B 方法以及滤波反投影 FBP 方法进行重建, 分别得到 OSEM-B衰减图像和第一 FBP衰减图像;
将所述 OSEM-B衰减图像和第一 FBP衰减图像进行加权以获取有效衰 减图像;
采用依据所述有效衰减图像生成的衰减弦图对所述当前 PET设备的发 射 Emission扫描弦图数据进行衰减校正。
2、 根据权利要求 1所述的方法, 其特征在于, 所述将所述 OSEM-B衰 减图像和第一 FBP衰减图像进行加权以获取有效衰减图像, 具体包括:
..Final _ OSEM-B i x FBP
釆用公式^ / + ^ ~ θί)μ 计算所述有效衰减图像的 有效衰减系数; 其中, 为 OSEM-B衰减图像的衰减系数, 为 FBP衰减图像 的衰减系数, "为加权参数。
3、 根据权利要求 2所述的方法, 其特征在于, 所述"的获取方式具体 为:
获取所述当前 PET设备出厂前进行 Transmission设备固有噪声扫描的 弦图数据, 并对所述噪声扫描的弦图数据进行 FBP重建, 以得到第二 FBP 衰减图像; 提取所述有效衰减图像的第一组织区域以及所述第二 FBP衰减图像的 第二組织区域; 利用所述第一组织区域和第二组织区域的直方图计算所述 。
4、 根据权利要求 3所述的方法, 其特征在于, 所述利用所述第一组织 区域和第二組织区域的直方图计算所述"具体为:
采用最小均方误差的计算公式 = ηήη|^¾, - ¾ΰ)¾ |计算", 其 中, 所述 H^Fi∞为所述第一组织区域的直方图, 所述 为所述第二组 织区域的直方图。
5、 根据权利要求 1所述的方法, 其特征在于, 所述对所述 Transmission 扫描弦图数据采用基于贝叶斯模型的有序子集最大期望值 OSEM-B方法进 行重建, 具体包括:
采用有序子集最大期望值 OSEM方法获取所述 Transmission扫描弦图数
, ^ , .new part
据的初始衰减图像的初始衰减系数 ― ;
将所述初始哀
Figure imgf000017_0001
型 按照公式 t . new ίΛ -, . . new part , - .
j = ~ aj ) j ~ te 进行结合, 以获取 OSEM-B衰减图像 的衰减系数; 其中 所述^ ^为水的衰减系数。
6、 一种 PET系统中图像的衰减校正装置, 其特征在于, 包括: 获取源数据模块, 用于获取当前 PET设备的透射 Transmission扫描弦 图数据;
OSEM-B重建模块, 用于对所述 Transmission扫描弦图数据釆用基于 贝叶斯模型的有序子集最大期望值 OSEM-B方法进行重建, 得到 OSEM-B 衰减图像; FBP重建模块, 用于对所述 Transmission扫描弦图数据釆用滤波反投 影 FBP方法进行重建, 得到第一 FBP衰减图像;
加权计算模块, 用于将所述 OSEM-B衰减图像和第一 FBP衰减图像进 行加权以获取有效衰减图像;
衰减校正模块, 用于采用依据所述有效衰减图像生成的衰减弦图对所 述当前 PET设备的发射 Emission扫描弦图数据进行衰减校正。
7、 根据权要求 6所述的装置, 其特征在于, 所述加权计算模块, 具体 用于:
..Final _ OSEM-B ( FBP
采用公式^ / - + {ί - θί)β 计算所述有效衰减图像的 有效衰减系数;
其中, μ 为 OSEM-B衰减图像的衰减系数, 为 FBP衰减图像 的衰减系数, "为加权参数。
8、 根据权利要求 7所述的装置, 其特征在于, 采用如下模块获取所述 a .
获取模块, 用于获取所述当前 PET设备出厂前进行 Transmission设备 固有噪声扫描的弦图数据, 并对所述噪声扫描的弦图数据进行 FBP重建, 以得到第二 FBP衰减图像;
提取模块, 用于提取所述有效衰减图像的第一组织区域以及所述第二
FBP衰减图像的第二组织区域;
计算模块, 用于利用所述第一组织区域和第二组织区域的直方图计算 所述 。
9、 根据权利要求 8所述的装置, 其特征在于, 所述计算模块, 具体用 于采用最小均方误差的计算公式" = min Hlst Final ~ Hist Long 计算 ^, 其 中, 所述/ 为所述第一组织区域的直方图, 所述7 为所述第二组织 区域的直方图。
10、 根据权利要求 6所述的装置, 其特征在于, 所述 OSEM-B重建模 块, 具体包括: 笫一获取子模块, 用于采用有序子集最大期望值 OSEM方法获取所述
Transmission短时扫描弦图数据的初始衰减图像的初始衰减系数 μ
. .new part
第二获取子模块, 用于将所述初始衰减系数 ― 与贝叶斯模型 nnew = (\ - ) unew- part +
按照公式 K J ) j^water 进行结合, 以获取
OSEM-B衰减图像的衰减系数;
其中, 所述^ 为水的衰减系数。
PCT/CN2012/074736 2011-12-30 2012-04-26 Pet系统中图像的衰减校正方法及装置 WO2013097390A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US14/359,108 US9311723B2 (en) 2011-12-30 2012-04-26 Attenuation correction method and device for image in PET system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110455239.4A CN103186882B (zh) 2011-12-30 2011-12-30 Pet系统中图像的衰减校正方法及装置
CN201110455239.4 2011-12-30

Publications (1)

Publication Number Publication Date
WO2013097390A1 true WO2013097390A1 (zh) 2013-07-04

Family

ID=48678040

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/074736 WO2013097390A1 (zh) 2011-12-30 2012-04-26 Pet系统中图像的衰减校正方法及装置

Country Status (3)

Country Link
US (1) US9311723B2 (zh)
CN (1) CN103186882B (zh)
WO (1) WO2013097390A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110809782A (zh) * 2018-10-22 2020-02-18 上海联影医疗科技有限公司 衰减校正系统和方法
CN113989400A (zh) * 2021-09-26 2022-01-28 清华大学 一种ct图像生成方法、装置、电子设备及计算机存储介质

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104586416B (zh) * 2013-10-30 2017-07-04 上海联影医疗科技有限公司 衰减图及正电子发射断层图像生成方法和装置
CN104352244B (zh) * 2014-10-21 2016-06-29 沈阳东软医疗系统有限公司 一种数据处理方法和装置
US9858691B2 (en) 2015-07-31 2018-01-02 The Board Of Trustees Of The Leland Stanford Junior University Statistical data acquisition model for GPU based MLEM joint estimation of tissue activity distribution and photon attenuation map from PET data
CN105678711B (zh) * 2016-01-29 2018-08-21 中国科学院高能物理研究所 一种基于图像分割的衰减校正方法
US11302003B2 (en) * 2017-10-26 2022-04-12 Wisconsin Alumni Research Foundation Deep learning based data-driven approach for attenuation correction of pet data
WO2019112658A1 (en) * 2017-12-05 2019-06-13 Siemens Medical Solutions Usa, Inc. Improved multi-focal non-parallel collimator-based imaging
CN108389242A (zh) * 2018-04-10 2018-08-10 上海联影医疗科技有限公司 投影图像的校正方法及装置
US10679385B1 (en) * 2018-12-17 2020-06-09 General Electric Company System and method for statistical iterative reconstruction and material decomposition
CN111179372B (zh) * 2019-12-31 2024-03-26 上海联影智能医疗科技有限公司 图像衰减校正方法、装置、计算机设备和存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1903128A (zh) * 2005-07-25 2007-01-31 株式会社岛津制作所 正电子ct设备的3d图像重构方法和正电子ct设备
CN101690418A (zh) * 2007-05-31 2010-03-31 通用电气公司 促进修正图像重建中的增益波动的方法和系统
CN101953693A (zh) * 2009-07-16 2011-01-26 西门子公司 完整或部分正电子发射断层造影衰减图的基于模型的估计

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6856666B2 (en) * 2002-10-04 2005-02-15 Ge Medical Systems Global Technology Company, Llc Multi modality imaging methods and apparatus
US9470801B2 (en) * 2004-01-13 2016-10-18 Spectrum Dynamics Llc Gating with anatomically varying durations
US7378660B2 (en) * 2005-09-30 2008-05-27 Cardiovascular Imaging Technologies L.L.C. Computer program, method, and system for hybrid CT attenuation correction
US8098916B2 (en) * 2007-10-30 2012-01-17 General Electric Company System and method for image-based attenuation correction of PET/SPECT images
CN101744618B (zh) * 2009-12-17 2012-01-25 北京亿仁赛博医疗设备有限公司 同机房传动pet/ct/mr图像采集、配准和成像方法
CN101856236B (zh) * 2010-06-13 2012-07-04 苏州瑞派宁科技有限公司 一种应用适应性的正电子发射断层成像方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1903128A (zh) * 2005-07-25 2007-01-31 株式会社岛津制作所 正电子ct设备的3d图像重构方法和正电子ct设备
CN101690418A (zh) * 2007-05-31 2010-03-31 通用电气公司 促进修正图像重建中的增益波动的方法和系统
CN101953693A (zh) * 2009-07-16 2011-01-26 西门子公司 完整或部分正电子发射断层造影衰减图的基于模型的估计

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110809782A (zh) * 2018-10-22 2020-02-18 上海联影医疗科技有限公司 衰减校正系统和方法
CN110809782B (zh) * 2018-10-22 2023-09-12 上海联影医疗科技股份有限公司 衰减校正系统和方法
CN113989400A (zh) * 2021-09-26 2022-01-28 清华大学 一种ct图像生成方法、装置、电子设备及计算机存储介质

Also Published As

Publication number Publication date
US9311723B2 (en) 2016-04-12
CN103186882A (zh) 2013-07-03
US20140328532A1 (en) 2014-11-06
CN103186882B (zh) 2015-09-30

Similar Documents

Publication Publication Date Title
WO2013097390A1 (zh) Pet系统中图像的衰减校正方法及装置
Liu et al. Deep learning with noise‐to‐noise training for denoising in SPECT myocardial perfusion imaging
Jia et al. A GPU tool for efficient, accurate, and realistic simulation of cone beam CT projections
Wang et al. Scatter correction for cone‐beam computed tomography using moving blocker strips: a preliminary study
Sun et al. An iterative projection‐based motion estimation and compensation scheme for head x‐ray CT
CN107871332A (zh) 一种基于残差学习的ct稀疏重建伪影校正方法及系统
CN109949411B (zh) 一种基于三维加权滤波反投影和统计迭代的图像重建方法
US7747057B2 (en) Methods and apparatus for BIS correction
Heußer et al. Prior‐based artifact correction (PBAC) in computed tomography
Joemai et al. Metal artifact reduction for CT: Development, implementation, and clinical comparison of a generic and a scanner‐specific technique
Zamyatin et al. Extension of the reconstruction field of view and truncation correction using sinogram decomposition
CN102846333A (zh) 用于x射线成像中的散射校正的方法和系统
Zhang et al. Few‐view cone‐beam CT reconstruction with deformed prior image
US20170340287A1 (en) Method And Apparatus For Motion Correction In CT Imaging
CN109523458B (zh) 一种结合稀疏诱导动态引导滤波的高精度稀疏角度ct重建方法
Brock et al. Reconstruction of a cone‐beam CT image via forward iterative projection matching
Tang et al. Statistical CT noise reduction with multiscale decomposition and penalized weighted least squares in the projection domain
Sharma et al. Interior micro‐CT with an offset detector
Jang et al. Head motion correction based on filtered backprojection for x‐ray CT imaging
WO2021062885A1 (zh) Ct成像方法、装置、存储介质及医学成像系统
Xie et al. Contextual loss based artifact removal method on CBCT image
Hu et al. Sinogram Restoration for Low‐Dosed X‐Ray Computed Tomography Using Fractional‐Order Perona‐Malik Diffusion
Fan et al. Image‐domain shading correction for cone‐beam CT without prior patient information
CN113313649B (zh) 图像重建方法及装置
Sun et al. Variational Bayesian blind restoration reconstruction based on shear wave transform for low-dose medical CT image

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: 12863329

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14359108

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 04.11.2014)

122 Ep: pct application non-entry in european phase

Ref document number: 12863329

Country of ref document: EP

Kind code of ref document: A1