CN112120724A - System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior - Google Patents

System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior Download PDF

Info

Publication number
CN112120724A
CN112120724A CN202010982001.6A CN202010982001A CN112120724A CN 112120724 A CN112120724 A CN 112120724A CN 202010982001 A CN202010982001 A CN 202010982001A CN 112120724 A CN112120724 A CN 112120724A
Authority
CN
China
Prior art keywords
module
pet
data
dictionary
pet image
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202010982001.6A
Other languages
Chinese (zh)
Inventor
邓贞宙
封子纪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanchang Hualiang Optoelectronics Co ltd
Original Assignee
Nanchang Hualiang Optoelectronics Co ltd
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 Nanchang Hualiang Optoelectronics Co ltd filed Critical Nanchang Hualiang Optoelectronics Co ltd
Priority to CN202010982001.6A priority Critical patent/CN112120724A/en
Publication of CN112120724A publication Critical patent/CN112120724A/en
Pending legal-status Critical Current

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]
    • A61B6/037—Emission tomography
    • 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
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • 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
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5258—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/003—Reconstruction from projections, e.g. tomography
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2210/00—Indexing scheme for image generation or computer graphics
    • G06T2210/41—Medical

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Optics & Photonics (AREA)
  • General Health & Medical Sciences (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Veterinary Medicine (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Nuclear Medicine (AREA)

Abstract

The invention provides a system and a method for restoring lost PET data of fragments based on sparsity prior, wherein the system comprises: the device comprises a detector module, a preprocessing module, a detection module and a data restoration module. The method comprises the following steps: detecting external event information, converting the external event information into an electric signal, preprocessing the external signal, detecting PET data lost by fragments, restoring the PET data lost by the fragments, and obtaining a high-resolution PET image, wherein the method depends on two coupled dictionaries, a D1 dictionary is used for training low-resolution PET image data, and a D2 dictionary is used for training high-resolution PET image data; finally, based on the sparse representation of the low resolution PET image of D1, a high resolution PET image was obtained from the D2 dictionary from a linear relationship between the input and output data. The PET data lost by the fragments are restored by utilizing sparsity prior for image reconstruction; the restored image data has good robustness and high resolution, and has good effects on the aspects of detail restoration and the like.

Description

System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior
Technical Field
The invention relates to a data recovery system and a method, in particular to a recovery system and a method of fragment-lost PET data based on sparsity prior.
Background
Positron Emission Tomography (PET) is a non-invasive in vivo imaging method, can non-invasively, quantitatively and dynamically evaluate the metabolic level, biochemical reaction, functional activity and perfusion of various organs in a human body, and is a clinical functional imaging device with the highest sensitivity. Radionuclides with short half-lives are synthesized with compounds required for human metabolism, such as glucose, choline, and acetic acid, and then injected into a human body, and these proton-rich radionuclides spontaneously convert protons into neutrons and emit positrons and neutrals. Positron annihilations produce a pair of gamma photons. These high-penetrating gamma photons are coincidence detected and the distribution of the locations where annihilation events occur is analytically or statistically reconstructed to reconstruct an image of the interior of the patient. The positron emission tomography imaging instrument plays an important role in assisting in diagnosing tumor and cancer, cardiovascular and cerebrovascular diseases, nervous system diseases and the like.
Generally, if the PET equipment malfunctions during the patient scanning process (for example, some detection units or electronic modules or image acquisition and collection parts are aged or temporarily off-line due to overdose), some information at some geometrical angles of the PET is lost, and the acquired coincidence data of the PET electronic system is incomplete. The conventional PET apparatus basically selects to terminate the scanning process, transfer to other normal machines for experiments or require the patient to perform a secondary scheduled scan, etc. This wastes the medication that is injected this time, and costs to hospitals and patients. The patient needs to reserve the injection of the radioactive drug again and carry out PET scanning again, so that the injury of the patient caused by the radiation of the radioactive drug and other instruments is increased, the diagnosis and treatment time of the patient can be delayed under more serious conditions, and serious consequences can be caused.
Disclosure of Invention
In order to solve the defects of the prior art, the invention provides a system and a method for restoring lost PET data of fragments based on sparsity prior.
In order to achieve the purpose, the invention adopts the following technical scheme:
the system for restoring the fragment lost PET data based on the sparsity prior comprises a detector module, a preprocessing module, a detection module and a data restoring module;
and the detector module comprises a crystal array module, a photoelectric converter module and a detector electronics module.
And the preprocessing module comprises a noise processing module.
And the data recovery module comprises a D1 dictionary module and a D2 dictionary module.
Furthermore, the detector module is used for acquiring external event information and converting the external event information into an electric signal, and the detector module comprises a crystal array module, a photoelectric converter module and a detector electronics module. The crystal array module is used for receiving external gamma rays and converting the gamma rays into a certain number of visible light photons, the gamma rays are incident to the crystal array module and are ionized and excited, atoms are demagnetized and excited to generate fluorescence photons, and the number of generated visible light photons is related to the energy of the ray photons. The photoelectric converter module is used for converting optical signals into analog electric signals, the optical signals generated by the crystal array module are transmitted to the photoelectric converter module, the optical signals are converted into voltage or current pulse signals, the voltage or current pulse signals are multiplied and amplified through electronics, and the signal size which can be processed by the back-end circuit is output. The detector electronics module is used for extracting event information from the simulated electric pulse signal, and comprises a pulse processing module and a coincidence processing module, wherein the pulse processing module is used for extracting information of a single pulse event, including time, energy and position information, and the coincidence processing module classifies the single pulse event into paired coincidence events according to the single pulse information.
Further, the preprocessing module is used for preprocessing the signal acquired by the detector module, and the preprocessing module includes a noise processing module. The noise processing module is used for carrying out noise reduction filtering processing on the data.
Further, the detection module is used for detecting the PET data with lost fragments.
Further, the data recovery module is used for recovering the fragment missing PET data, and the data recovery module comprises a D1 dictionary module and a D2 dictionary module. The D1 and D2 dictionary modules are trained from a large number of real PET image blocks. The D1 dictionary module is used for obtaining sparse representation of each plaque of the input PET image, training low-resolution PET image data, the D2 dictionary module is used for generating a high-resolution PET image through the sparse representation, training the high-resolution image data, and obtaining the high-resolution PET image from the D2 dictionary module according to linear relation between the input data and the output data based on the sparse representation of the low-resolution PET image of the D1 dictionary module.
The method for restoring the fragment missing PET data based on sparsity prior comprises the following steps:
s1: detecting external event information and converting the external event information into an electric signal;
s2: preprocessing an external signal;
s3: detecting missing PET data;
s4: and restoring the lost PET data to obtain a high-resolution PET image.
In step S1, the detector module obtains external event information, wherein the photoelectric converter module converts the optical signal generated by the crystal array module into an analog electrical signal.
In step S2, the preprocessing module is configured to perform preprocessing on the signal, including noise reduction filtering processing.
In the step S4, the lost PET data is restored to obtain a high-resolution PET image, and the method relies on two coupled dictionaries, the D1 dictionary is used for training the low-resolution PET image data, and the D2 dictionary is used for training the high-resolution PET image data. Finally, based on the sparse representation of the low resolution PET image of D1, a high resolution PET image was obtained from the D2 dictionary from a linear relationship between the input and output data.
In step S4, given the low-resolution PET image Y, two constraints are satisfied for generating the high-resolution PET image X that exhibits the same condition:
(1) for sparse reconstruction, the high resolution PET image X and the low resolution PET image Y should satisfy the condition: y — SHX, where H represents the sparse matrix and S represents the sampling matrix. The global model in the constraint condition is used for ensuring the robustness of the recovered PET image and inhibiting artifacts possibly generated by the local model.
(2) The high resolution PET image X is segmented into a series of small segments X that can be sparsely represented in the D2 dictionary: x ≈ D2α, for α ∈ RkHas | | | alpha | | non-conducting phosphor0K, where α is a sparse representation of the segment x of the input low resolution PET image Y, reflecting a sparse prior。
In said step S4, using the local model, we train the dictionaries of both D1 and D2 representing image textures, so that the low-resolution and high-resolution PET images have the same sparse representation. The process of finding the sparsest representation from the D1 dictionary is: sparse condition: min | | alpha | luminance0s.t.
Figure BDA0002687872900000033
And the method is used for acquiring the representation of each local segment, wherein F is a feature extraction operator of the low-resolution PET image, so that alpha in the calculation is closely related to the image to be restored, and the prediction is more reasonable. Then, the constraint condition (1) is used as a premise, and a representative result is used as a basis for restoring the whole image.
In step S4, in order to find a sparse representation of an image, the most important step is to obtain an appropriate sparse condition:
Figure BDA0002687872900000031
using a given pair of PET image slices: { Y | Y ═ Y1,y2,...,ynAnd X | X ═ X1,x2,...,xnTo train two dictionaries of relevance, D1 and D2, of the formula:
Figure BDA0002687872900000032
wherein Dc＝{D1,D2}，Xc＝{Y,X}，l1Norm | Z | counting1For enhancing sparsity, the parameter λ is used to balance the sparsity of the solution.
The method is used for restoring the PET data with lost fragments based on the sparse prior of the image, and the PET image restoration model obtained by the method is good in robustness, high in resolution and good in effect on the aspects of detail restoration and the like.
Drawings
FIG. 1 is a flowchart of a system and method for recovering missing PET data based on sparsity prior according to the present invention.
Fig. 2 is a block signal transmission diagram of a system and method for recovering missing PET data based on sparsity prior according to an embodiment of the present invention.
Detailed Description
In order that the invention may be better understood, reference will now be made to the following examples.
The system for restoring the fragment missing PET data based on the sparsity prior comprises a detector module 100, a preprocessing module 200, a detection module 300 and a data restoring module 400, as shown in FIG. 2.
The detector module 100 is used for acquiring gamma rays generated by positron annihilation in a living body and converting the gamma rays into electrical signals, and the detector module 100 comprises a crystal array module 110, a photoelectric converter module 120 and a detector electronics module 130. The crystal array module 110 is configured to receive external gamma rays and convert the gamma rays into a certain number of visible light photons, the gamma rays are incident on the crystal array module and are ionized and excited, atoms are exited to generate fluorescence photons, and the number of generated visible light photons is related to the energy of the ray photons. The optical-to-electrical converter module 120 is configured to convert the optical signal into an analog electrical signal, the optical signal generated by the crystal array module 110 is transmitted to the optical-to-electrical converter module 120, the optical signal is converted into a voltage or current pulse signal, and the voltage or current pulse signal is multiplied and amplified by electronics to output a signal that can be processed by the back-end circuit. The detector electronics module 130 is configured to extract information about events from the simulated electrical pulse signal, and includes a pulse processing module 131 and a coincidence processing module 132, the pulse processing module 131 is configured to extract information about single-pulse events, including time, energy, and location information, and the coincidence processing module 132 classifies the single-pulse events as paired coincidence events according to the information about the single pulse.
The preprocessing module 200 is used for preprocessing the signals acquired by the detector module 100, and the preprocessing module 200 includes a noise processing module 210. The noise processing module 210 is configured to perform noise reduction filtering on the data, perform a frequency domain gaussian low-pass filtering, and remove noise.
The detection module 300 is configured to detect missing PET data.
The data recovery module 300 is used for recovering the fragment missing PET data, and the data recovery module 300 comprises a D1 dictionary module 310 and a D2 dictionary module 320. The D1 dictionary module 310 and the D2 dictionary module 320 are trained from a large number of real PET image blocks. Wherein the D1 dictionary module 310 is configured to obtain a sparse representation of each blob of the input PET image, train the low resolution PET image data, the D2 dictionary module 320 is configured to generate the high resolution PET image from the sparse representation, train the high resolution image data, and obtain the high resolution PET image from the D2 dictionary module based on the sparse representation of the low resolution PET image of the D1 dictionary module according to a linear relationship between the input and output data.
The method for restoring the fragment missing PET data based on sparsity prior comprises the following steps:
step S1: detecting external event information and converting the external event information into an electric signal;
step S2: preprocessing an external signal;
step S3: detecting missing PET data;
step S4: and restoring the lost PET data to obtain a high-resolution PET image.
In step S1, the detector module obtains external event information, wherein the photoelectric converter module converts the optical signal generated by the crystal array module into an analog electrical signal.
In step S2, the preprocessing module is configured to perform preprocessing on the signal, including noise reduction filtering processing.
In the step S4, the lost PET data is restored to obtain a high-resolution PET image, and the method relies on two coupled dictionaries, the D1 dictionary is used for training the low-resolution PET image data, and the D2 dictionary is used for training the high-resolution PET image data. Finally, based on the sparse representation of the low resolution PET image of D1, a high resolution PET image was obtained from the D2 dictionary from a linear relationship between the input and output data.
In step S4, given the low-resolution PET image Y, two constraints are satisfied for generating the high-resolution PET image X that exhibits the same condition:
(1) for sparse reconstruction, the high resolution PET image X and the low resolution PET image Y should satisfy the condition: y — SHX, where H represents the sparse matrix and S represents the sampling matrix. The global model in the constraint condition is used for ensuring the robustness of the recovered PET image and inhibiting artifacts possibly generated by the local model.
(2) The high resolution PET image X is segmented into a series of small segments X that can be sparsely represented in the D2 dictionary: x ≈ D2α, for α ∈ RkHas | | | alpha | | non-conducting phosphor0K, where α is a sparse representation of the segment x of the input low resolution PET image Y, which reflects the sparse prior.
In said step S4, using the local model, we train the dictionaries of both D1 and D2 representing image textures, so that the low-resolution and high-resolution PET images have the same sparse representation. The process of finding the sparsest representation from the D1 dictionary is: sparse condition: min | | alpha | luminance0s.t.
Figure BDA0002687872900000063
And the method is used for acquiring the representation of each local segment, wherein F is a feature extraction operator of the low-resolution PET image, so that alpha in the calculation is closely related to the image to be restored, and the prediction is more reasonable. Then, the constraint condition (1) is used as a premise, and a representative result is used as a basis for restoring the whole image.
In step S4, in order to find a sparse representation of an image, the most important step is to obtain an appropriate sparse condition:
Figure BDA0002687872900000061
using a given pair of PET image slices: { Y | Y ═ Y1,y2,...,ynAnd X | X ═ X1,x2,...,xnTo train two dictionaries of relevance, D1 and D2, of the formula:
Figure BDA0002687872900000062
wherein Dc＝{D1,D2}，Xc＝{Y,X}，l1Norm | Z | counting1For enhancing sparsity, the parameter λ is used to balance the sparsity of the solution.
The method is used for restoring the PET data with lost fragments based on the sparse prior of the image, and the PET image restoration model obtained by the method is good in robustness, high in resolution and good in effect on the aspects of detail restoration and the like.
Although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the spirit and scope of the invention.

Claims (10)

1. A system for restoring lost PET data of fragments based on sparsity prior is characterized in that: comprises a detector module, a preprocessing module, a detection module and a data restoration module, wherein,
the detector module comprises a crystal array module, a photoelectric converter module and a detector electronics module and is used for capturing positron annihilation event information in the PET system, acquiring external event information and converting the external event information into an electric signal;
the preprocessing module comprises a noise processing module and is used for preprocessing the information transmitted by the detector module;
the detection module is used for detecting the PET data with lost fragments;
and the data recovery module comprises a D1 dictionary module and a D2 dictionary module and is used for recovering the PET image data with lost fragments.
2. The sparsity-priors-based system for restoration of segment-missing PET data according to claim 1, wherein: the detector electronics module is used to extract information of the event, including time, energy, and location information, from the simulated electrical pulse signal.
3. The sparsity-priors-based system for restoration of segment-missing PET data according to claim 1, wherein: the D1 dictionary module and the D2 dictionary module are trained from a large number of real PET image blocks using regularization methods.
4. The sparsity-priors-based system for restoration of segment-missing PET data according to claim 1, wherein: the D1 dictionary module is used to obtain a sparse representation of each blob of the input PET image, train the low resolution PET image data, the D2 dictionary module is used to generate the high resolution PET image from the sparse representation, train the high resolution image data, and obtain the high resolution PET image from the D2 dictionary based on a linear relationship between the input and output data. .
5. The method for restoring the fragment lost PET data based on the sparsity prior is characterized by comprising the following steps: the method comprises the following steps:
step S1: detecting external event information and converting the external event information into an electric signal;
step S2: preprocessing an external signal;
step S3: detecting missing PET data;
step S4: and restoring the lost PET data of the fragments by using a sparsity prior technology to obtain a high-resolution PET image.
6. The sparsity-prior based restoration method of segment-missing PET data according to claim 5, wherein: in step S1, the PET detector is a detector for capturing positron annihilation event information in a PET system, and structurally, the PET detector module includes a crystal array module, a photoelectric converter module, and a detector electronics module, and the specific steps of acquiring data include:
t11: the crystal array module converts gamma photons into visible light photons and soft ultraviolet light photons;
t12: the photoelectric converter module converts the optical signal into an analog electrical signal;
t13: the detector electronics module extracts information of the event from the simulated electrical pulse signal.
7. The sparsity-prior based restoration method of segment-missing PET data according to claim 5, wherein: in the step S4, the segment missing PET data is restored to obtain a high resolution PET image, the method relies on two coupled dictionaries, the D1 dictionary is used for training the low resolution PET image data, the D2 dictionary is used for training the high resolution PET image data, and finally, based on the sparse representation of the D1 low resolution PET image, the high resolution PET image is obtained from the D2 dictionary according to the linear relationship between the input and output data.
8. The sparsity-prior based restoration method of segment-missing PET data according to claim 7, wherein: in step S4, the low-resolution PET image Y trained from the D1 dictionary generates the high-resolution PET image X that exhibits the same conditions, and satisfies two constraint conditions:
(1) for sparse reconstruction, the high resolution PET image X and the low resolution PET image Y should satisfy the condition: y — SHX, where H represents the sparse matrix and S represents the sampling matrix. The global model in the constraint condition is used for ensuring the robustness of the recovered PET image and inhibiting artifacts possibly generated by the local model.
(2) The high resolution PET image X is segmented into a series of small segments X that can be sparsely represented in the D2 dictionary: x ≈ D2α, for α ∈ RkHas | | | alpha | | non-conducting phosphor0K, where α is a sparse representation of the segment x of the input low resolution PET image Y, which reflects the sparse prior.
9. The sparsity-prior based restoration method of segment-missing PET data according to claim 8, wherein: in step S4, the process of finding the most sparse representation according to the D1 dictionary is: sparse condition:
Figure FDA0002687872890000021
and the method is used for acquiring the representation of each local segment, wherein F is a feature extraction operator of the low-resolution PET image so as to ensure that alpha in the calculation is closely related to the image to be restored and the prediction is more reasonable(ii) a Then, the constraint condition (1) is used as a premise, and a representative result is used as a basis for restoring the whole image.
10. The sparsity-prior based restoration method of segment-missing PET data according to claim 9, wherein: in step S4, to obtain a suitable thinning condition:
Figure FDA0002687872890000022
Figure FDA0002687872890000023
using a given pair of PET image slices: { Y | Y ═ Y1,y2,...,ynAnd X | X ═ X1,x2,...,xnTo train two dictionaries of relevance, D1 and D2, of the formula:
Figure FDA0002687872890000031
wherein Dc＝{D1,D2}，Xc＝{Y,X}，l1Norm | Z | counting1For enhancing sparsity, the parameter λ is used to balance the sparsity of the solution.
CN202010982001.6A 2020-09-17 2020-09-17 System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior Pending CN112120724A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010982001.6A CN112120724A (en) 2020-09-17 2020-09-17 System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010982001.6A CN112120724A (en) 2020-09-17 2020-09-17 System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior

Publications (1)

Publication Number Publication Date
CN112120724A true CN112120724A (en) 2020-12-25

Family

ID=73841748

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010982001.6A Pending CN112120724A (en) 2020-09-17 2020-09-17 System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior

Country Status (1)

Country Link
CN (1) CN112120724A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009129137A1 (en) * 2008-04-17 2009-10-22 Wisconsin Alumni Research Foundation Method for image reconstruction employing sparsity-constrained iterative correction
CN104778671A (en) * 2015-04-21 2015-07-15 重庆大学 Image super-resolution method based on SAE and sparse representation
CN107346556A (en) * 2017-07-06 2017-11-14 浙江大学 A kind of PET image reconstruction method based on block dictionary learning and sparse expression

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009129137A1 (en) * 2008-04-17 2009-10-22 Wisconsin Alumni Research Foundation Method for image reconstruction employing sparsity-constrained iterative correction
CN104778671A (en) * 2015-04-21 2015-07-15 重庆大学 Image super-resolution method based on SAE and sparse representation
CN107346556A (en) * 2017-07-06 2017-11-14 浙江大学 A kind of PET image reconstruction method based on block dictionary learning and sparse expression

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
ZHANLI HU等: "Super-resolution PET image reconstruction with sparse representation", 《2017 IEEE NUCLEAR SCIENCE SYMPOSIUM AND MEDICAL IMAGING CONFERENCE》 *
张必翔: "专业学位研究生实验课程 医科篇", 华中科技大学出版社 *
王静等: "基于正则变分模型的医学影像非线性复原方法", 《中国医学物理学杂志》 *
童基均等: "基于飞行时间信息与稀疏正则化的PET图像重建", 《计算机辅助设计与图形学学报》 *
陈健等: "图像超分辨率复原方法及应用", 《激光与光电子学进展》 *

Similar Documents

Publication Publication Date Title
US9554759B2 (en) Digital radiography detector image readout process
US8003948B2 (en) Data acquisition for positron emission tomography
CN110772274A (en) Multi-pixel gamma imaging system and method
CN102361593B (en) Apparatus and method for data acquisition using an imaging apparatus
JP2008170440A (en) System and method for excluding damage pet data damage during simultaneous pet-mr acquisition
EP1552789A3 (en) Efficient single photon emission imaging
EP4037833A2 (en) Systems and methods of using self-attention deep learning for image enhancement
Lee et al. Advances in imaging instrumentation for nuclear cardiology
CN109961419B (en) Correction information acquisition method for attenuation correction of PET activity distribution image
US20180372890A1 (en) Calibrating time in pet device
EP2389661A1 (en) Nuclear image reconstruction
Hendee Physics and applications of medical imaging
CN111436958A (en) CT image generation method for PET image attenuation correction
WO2010132523A1 (en) Image analysis method and system
CN110275196B (en) Method, apparatus and product for determining interaction characteristics using a plurality of pixels
Taguchi et al. Photon counting detector computed tomography
US11431928B2 (en) System and method for mitigating electromagnetic interference when acquiring image data
CN112120724A (en) System and method for restoring fragment lost PET (positron emission tomography) data based on sparsity prior
Malczewski PET image reconstruction using compressed sensing
CN102028484A (en) Detector module for an emission tomography scanner
Miller et al. Progress of BazookaSPECT
CN112205996B (en) Image encryption system and method based on photon random offset
EP2711738A1 (en) A method and a device to generate virtual X-ray computed tomographic image data
US8892184B2 (en) Systems and methods for reducing interference in a dual modality imaging system
CN110584693B (en) Dual-mode imaging method, device and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20201225