CN109272472B - Noise and artifact eliminating method for medical energy spectrum CT image - Google Patents
Noise and artifact eliminating method for medical energy spectrum CT image Download PDFInfo
- Publication number
- CN109272472B CN109272472B CN201811198845.0A CN201811198845A CN109272472B CN 109272472 B CN109272472 B CN 109272472B CN 201811198845 A CN201811198845 A CN 201811198845A CN 109272472 B CN109272472 B CN 109272472B
- Authority
- CN
- China
- Prior art keywords
- images
- phantom
- neural network
- training
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000001228 spectrum Methods 0.000 title claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 19
- 238000013170 computed tomography imaging Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 230000000694 effects Effects 0.000 claims abstract description 7
- 238000002591 computed tomography Methods 0.000 claims abstract description 6
- 230000003595 spectral effect Effects 0.000 claims abstract description 6
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 15
- 230000005855 radiation Effects 0.000 claims description 7
- 210000000056 organ Anatomy 0.000 claims description 4
- 210000001519 tissue Anatomy 0.000 claims description 4
- 239000008280 blood Substances 0.000 claims description 3
- 210000004369 blood Anatomy 0.000 claims description 3
- 210000000988 bone and bone Anatomy 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000001678 irradiating effect Effects 0.000 claims description 3
- 210000004185 liver Anatomy 0.000 claims description 3
- 210000004072 lung Anatomy 0.000 claims description 3
- 210000005075 mammary gland Anatomy 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000013441 quality evaluation Methods 0.000 claims description 3
- 210000004872 soft tissue Anatomy 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 abstract 1
- 238000003062 neural network model Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- High Energy & Nuclear Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Optics & Photonics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Heart & Thoracic Surgery (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention relates to the field of computed tomography, and aims to reduce the influence caused by beam hardening and random noise, output more accurate medical images and reduce the harm to human bodies. Therefore, the invention provides a method for eliminating noise and artifacts of a medical energy spectrum CT image, which comprises the following steps: step 1: constructing a virtual phantom; step 2: simulating spectral CT imaging with a plurality of X-ray photons through a phantom; step 3: simulating spectral CT imaging with lower dose X-ray photons through the phantom; step 4: matching the reconstructed images; step5, training the convolutional neural network to finish training; step6 test network training effect. The invention is mainly applied to the design and manufacture occasions of medical CT equipment.
Description
Technical Field
The invention relates to the field of Computed Tomography (Computed Tomography), and aims to solve the problem of low signal-to-noise ratio caused by the fact that the number of photons collected in different energy intervals of an energy spectrum CT is small. In particular to a noise and artifact eliminating method for medical energy spectrum CT images.
Background
The key points of medical CT imaging are mainly divided into two points: the image precision is improved, and the radiation dose is reduced. The multi-energy spectrum CT represented by the single-photon counting type detector and the energy integrating type detector can provide more accurate image information and improve the precision of medical imaging, and the narrower the energy interval is, the closer the energy interval is to the single-energy imaging, the more beam hardening artifacts can be eliminated better, more comprehensive human body information can be provided, and the advantages are increased for improving the accuracy of diagnosis. However, under the condition of reducing the total radiation dose, the number of photons covered by each energy interval is reduced, which results in a low signal-to-noise ratio of the image, low dose noise and artifacts, and difficulty in meeting the accuracy requirement of the medical image.
Generally, the process of energy interaction between X-rays and matter involves a complex energy interaction mechanism, and the generation and collection process of charge collection is a probabilistic event. The single photon counting detector has high requirement on the counting rate, the energy integral detector has serious interlayer crosstalk, and the noise and the artifact brought to the image under the condition of insufficient photon number are difficult to eliminate by the traditional method. In recent years, with the development of machine learning technology, the convolutional neural network has great advantages, and the reconstructed images of each energy interval of the energy spectrum CT are processed by the convolutional neural network, so that the radiation dose of X-rays is further reduced while the actual clinical requirements of the high-precision energy spectrum CT are met.
Disclosure of Invention
The invention aims to provide a medical X-ray energy spectrum CT image processing method based on a convolutional neural network, aiming at overcoming the defects of the prior art and solving the problems of low-dose noise and artifacts in energy spectrum CT imaging. Therefore, the technical scheme adopted by the invention is a noise and artifact eliminating method for medical energy spectrum CT images, which comprises the following steps:
step 1: constructing a virtual phantom, simulating a human body composition construction phantom, and simulating the tissue structures of bones, blood, fat and soft tissues and important organs including livers, lungs and mammary glands;
step 2: simulating energy spectrum CT imaging by a large number of X-ray photons through a phantom, selecting one phantom, irradiating by using clinical standard dose X-rays, simulating the physical process of a detector by using software, reconstructing images to obtain medical images of different energy intervals, replacing the phantom, repeating the process to obtain a data set as a label image of a neural network;
Step 3: simulating spectral CT imaging by using lower-dose X-ray photons through a phantom, and repeating the detection and reconstruction processes in Step2 by using lower-dose X-rays to obtain a data set serving as an input image of a neural network;
step 4: and matching the reconstructed images. Setting images obtained from different energy intervals of the same phantom into a group so that a convolutional neural network can learn the information of a full spectrum, matching label images obtained from different doses of X-rays in Step2 through the same phantom with input images obtained in Step3 to enable the label images to correspond to the input images one by one, and repeating the process until the matching of reconstructed images under all phantom conditions is completed;
step5 training the convolutional neural network. After a proper convolution neural network structure is defined, randomly selecting N groups of input images sorted in Step4 and N groups of corresponding label images to train the neural network, and completing training through a back propagation algorithm until a loss function is converged to the minimum;
and Step6, testing the network training effect, namely, inputting the input image into the network by using the rest part of the matched data set in the Step4 except the image used for neural network training of the Step5 as a test data set, checking the network training effect by using the image quality evaluation standard, finishing the network training if the network training is qualified, and repeating the Step5 and the Step6 if the network training is not qualified.
The large number of X-ray photons is the standard radiation dose used in clinical normal CT scans, and the lower dose is one quarter of the number of photons used in Step 2.
The invention has the characteristics and beneficial effects that:
the invention provides a noise and artifact eliminating method for a medical energy spectrum CT image. The low signal-to-noise ratio images of different energy intervals of the energy spectrum CT are processed through the convolutional neural network model, the X-ray radiation dose is reduced, the image quality is improved, and favorable conditions are provided for accurate medical treatment. Meanwhile, the trained neural network has higher processing speed, saves a lot of time compared with an iterative reconstruction algorithm and the like, and improves the efficiency of image reconstruction.
Description of the drawings:
fig. 1 is a frame diagram of a method for eliminating noise and artifacts from a medical X-ray energy spectrum CT image.
FIG. 2 is a flow chart of a method for obtaining a neural network model.
Detailed Description
Aiming at the problems of low dose noise and artifacts in the energy spectrum CT imaging, the invention provides a medical X-ray energy spectrum CT image processing method based on a convolutional neural network, as shown in figure 1. After one-time low-dose X-ray scanning, images containing noise and artifacts in each energy interval are processed through a pre-trained convolutional neural network model and are restored into reconstructed images with the same quality as the high-dose X-ray irradiation, so that the influence caused by beam hardening and random noise can be reduced, more accurate medical images are output, and the harm to a human body is reduced.
The invention provides a noise and artifact eliminating method for medical energy spectrum CT images, which comprises the steps of firstly obtaining images of different energy intervals reconstructed under the condition of low-dose X-ray through an energy spectrum CT system, and then obtaining the images of different energy intervals suitable for precise medical treatment with the same quality as the images reconstructed under the condition of high-dose X-ray through a pre-trained convolutional neural network model by taking the images as input data. The convolutional neural network model is obtained by training images of different energy intervals reconstructed by the same phantom under the conditions of respectively simulating low-dose X-rays and high-dose X-rays, a frame diagram of the process is shown in figure 1, a flow diagram of a specific obtaining method of the convolutional neural network model is shown in figure 2, and the specific implementation scheme is as follows:
step 1: a virtual phantom is constructed. The human body composition is simulated to build a phantom, and the tissue structures of bones, blood, fat, soft tissues and the like and important organs of livers, lungs, mammary glands and the like are simulated as far as possible.
Step 2: spectral CT imaging is simulated by passing a large number of X-ray photons through a phantom. Selecting a phantom, irradiating by using X-rays with clinical standard dose, simulating the physical process of a detector by using software, and reconstructing images to obtain medical images of different energy intervals. And (4) replacing the phantom, and repeating the process to obtain a data set as a label image of the neural network.
Step 3: spectral CT imaging is simulated with a small number of X-ray photons passing through the phantom. The detection and reconstruction process in Step2 was repeated with a lower dose of X-rays (approximately one-fourth of the number of photons used in Step 2) to obtain a data set as the input image to the neural network.
Step 4: and matching the reconstructed images. Images obtained from different energy intervals of the same phantom are set into a group, so that a convolutional neural network can learn the information of a full spectrum, and label images obtained from different doses of X-rays in Step2 through the same phantom are matched with input images obtained in Step3, and the label images correspond to the input images one by one. The above process is repeated until the matching of the reconstructed images under all phantom conditions is completed.
Step5 training the convolutional neural network. After a proper convolutional neural network structure is defined, training the neural network by randomly selecting N groups of input images sorted in Step4 and N groups of corresponding label images, and completing training by a back propagation algorithm until the loss function is converged to the minimum.
And Step6, testing the network training effect. And (4) using the rest of the matched data set in the Step4 except the image used for neural network training in the Step5 as a test data set, inputting the input image into the network, checking the effect of the network training by using an image quality evaluation standard, finishing the network training if the network training is qualified, and repeating the Step5 and the Step6 if the network training is not qualified.
The high-dose X-ray mentioned in the invention is the standard radiation dose used in clinical normal CT scanning, which varies with different positions of the scanned human tissue, and the low-dose X-ray dose is usually one fourth of the standard dose.
The method can be used for processing the low-dose energy spectrum CT image clinically under the condition that a convolution neural network model capable of restoring the image with the noise and the artifact reconstructed by the low-dose X-ray is generated in the steps. And obtaining images with noise and artifacts in low-dose different energy intervals through an energy spectrum CT imaging system, inputting the images into a convolutional neural network model, and outputting medical images with the same quality as the reconstructed images in high-dose different energy intervals.
The invention is further illustrated by the following examples, but without thereby limiting the invention to the scope of the examples described, and simple variations thereof, based on the inventive concept, should be made by a person skilled in the art within the scope of the invention as claimed. The following detailed description is made with reference to the accompanying drawings:
the phantom adopted when the neural network model is trained accords with the reality of a human body and is highly similar to important organs of the human body. The X-ray source to be employed was generated by an analog GE maximum 125 in which the peak tube voltage and current were set to E, respectively Max120keV, 0.5mAs and EMax120keV, 2mAs to simulate low and high dose X-rays, respectively, other parameters being default settings. And accumulating the charge number of each energy interval in groups through a layered energy integration type detector, and reconstructing an image after energy analysis to form a data set of the neural network.
And randomly selecting N groups of images in the data set as a training set, using the reconstructed images of low-dose different energy intervals as input images, using the reconstructed images of high-dose different energy intervals as label images, inputting the label images into a convolutional neural network for training until a loss function is converged to the minimum, and finishing the training of the network. And inputting the input image into the network as a test set, evaluating the output image by using indexes such as PSNR (peak signal to noise ratio), CNR (contrast to noise ratio) and the like, finishing the training of the convolutional neural network if the required standard is met, reselecting N groups of images for training if the required standard is not met, and adjusting the network structure and the parameters thereof until the output image meets the requirements. After the training of the neural network model is completed, the method can be used for processing the medical image with noise and artifacts reconstructed under the condition of low-dose X-ray.
Claims (2)
1. A noise and artifact eliminating method for medical energy spectrum CT images is characterized by comprising the following steps:
step 1: constructing a virtual phantom, simulating a human body composition construction phantom, and simulating the tissue structures of bones, blood, fat and soft tissues and important organs including livers, lungs and mammary glands;
step 2: simulating energy spectrum CT imaging by a large number of X-ray photons through a phantom, selecting one phantom, irradiating by using clinical standard dose X-rays, simulating the physical process of a detector by using software, reconstructing images to obtain medical images of different energy intervals, replacing the phantom, repeating the process to obtain a data set as a label image of a neural network;
step 3: simulating spectral CT imaging by using lower-dose X-ray photons through a phantom, and repeating the detection and reconstruction processes in Step2 by using lower-dose X-rays to obtain a data set serving as an input image of a neural network;
step 4: matching the reconstructed images, setting the images obtained by the same phantom in different energy intervals into a group so that the convolutional neural network can learn the information of the total energy spectrum, matching the label images obtained by the same phantom in Step2 through different doses of X rays with the input images obtained in Step3 to enable the label images to be in one-to-one correspondence, and repeating the process until the matching of the reconstructed images of all phantoms in different energy intervals is completed;
Step5, training the convolution neural network, after defining a proper convolution neural network structure, randomly selecting N groups of input images sorted in Step4 and N groups of corresponding label images to train the neural network, and completing training by a back propagation algorithm until the loss function is converged to the minimum;
and Step6, testing the network training effect, namely, inputting the input image into the network by using the rest part of the matched data set in the Step4 except the image used for neural network training of the Step5 as a test data set, checking the network training effect by using the image quality evaluation standard, finishing the network training if the network training is qualified, and repeating the Step5 and the Step6 if the network training is not qualified.
2. The method of noise and artifact removal for medical energy spectrum CT images as claimed in claim 1, wherein the large number of X-ray photons is the standard radiation dose used in clinically normal CT scans and the lower dose is one quarter of the number of photons used in Step 2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811198845.0A CN109272472B (en) | 2018-10-15 | 2018-10-15 | Noise and artifact eliminating method for medical energy spectrum CT image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811198845.0A CN109272472B (en) | 2018-10-15 | 2018-10-15 | Noise and artifact eliminating method for medical energy spectrum CT image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109272472A CN109272472A (en) | 2019-01-25 |
CN109272472B true CN109272472B (en) | 2022-07-15 |
Family
ID=65196869
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811198845.0A Active CN109272472B (en) | 2018-10-15 | 2018-10-15 | Noise and artifact eliminating method for medical energy spectrum CT image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109272472B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7237624B2 (en) | 2019-02-07 | 2023-03-13 | 浜松ホトニクス株式会社 | Image processing device and image processing method |
CN109978769B (en) * | 2019-04-04 | 2023-06-20 | 深圳安科高技术股份有限公司 | CT scanning image data interpolation method and system thereof |
CN110246199B (en) * | 2019-05-26 | 2023-05-09 | 天津大学 | Projection domain data noise removing method for energy spectrum CT |
CN110517194A (en) * | 2019-07-19 | 2019-11-29 | 深圳安科高技术股份有限公司 | A kind of training method of substance decomposition model, substance decomposition method and relevant device |
CN110728727B (en) * | 2019-09-03 | 2023-04-18 | 天津大学 | Low-dose energy spectrum CT projection data recovery method |
CN112581554B (en) * | 2019-09-30 | 2024-02-27 | 中国科学院深圳先进技术研究院 | CT imaging method, device, storage equipment and medical imaging system |
CN110742635B (en) * | 2019-10-08 | 2021-10-08 | 南京安科医疗科技有限公司 | Composite energy spectrum CT imaging method |
CN113689359B (en) * | 2021-09-23 | 2024-05-14 | 上海联影医疗科技股份有限公司 | Image artifact removal model and training method and system thereof |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202049120U (en) * | 2011-03-04 | 2011-11-23 | 首都师范大学 | System for eliminating geometric artifacts in CT (computed tomography) image |
CN102652674A (en) * | 2011-03-04 | 2012-09-05 | 首都师范大学 | Method and system for eliminating geometrical artifacts in CT (Computerized Tomography) image |
CN108564553A (en) * | 2018-05-07 | 2018-09-21 | 南方医科大学 | Low-dose CT image noise suppression method based on convolutional neural networks |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9111337B2 (en) * | 2012-10-12 | 2015-08-18 | Mayo Foundation For Medical Education And Research | Low dose cardiac CT imaging with time-adaptive filtration |
-
2018
- 2018-10-15 CN CN201811198845.0A patent/CN109272472B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202049120U (en) * | 2011-03-04 | 2011-11-23 | 首都师范大学 | System for eliminating geometric artifacts in CT (computed tomography) image |
CN102652674A (en) * | 2011-03-04 | 2012-09-05 | 首都师范大学 | Method and system for eliminating geometrical artifacts in CT (Computerized Tomography) image |
CN108564553A (en) * | 2018-05-07 | 2018-09-21 | 南方医科大学 | Low-dose CT image noise suppression method based on convolutional neural networks |
Non-Patent Citations (2)
Title |
---|
基于视觉系统和特征提取的图像质量客观评价方法及应用研究;刘明娜;《中国博士学位论文全文数据库.医药卫生科技》;20160915(第06期);摘要,第17、35-36页 * |
螺旋束CT系统重建及射束硬化伪影抑制方法;周丽平;《中国优秀硕士学位论文全文数据库.信息科技辑》;20180615(第06期);第2-3、61-62页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109272472A (en) | 2019-01-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109272472B (en) | Noise and artifact eliminating method for medical energy spectrum CT image | |
Szczykutowicz et al. | Dual energy CT using slow kVp switching acquisition and prior image constrained compressed sensing | |
Long et al. | Monte Carlo simulations of adult and pediatric computed tomography exams: validation studies of organ doses with physical phantoms | |
CN102846333B (en) | The method and system of the scatter correction in x-ray imaging | |
CN103458967A (en) | Radiation therapy system and therapy planning device | |
JP4344191B2 (en) | Method and system for low-dose image simulation of an imaging system | |
CN103961125A (en) | CT (Computed Tomography) value correcting method for cone-beam CT | |
JP2003232855A (en) | Device for preparing tomogram, method for preparing tomogram, and device for radiographic examination | |
CN110811660B (en) | Method for correcting CT ray beam hardening artifact | |
CN111436958B (en) | CT image generation method for PET image attenuation correction | |
Lin et al. | An efficient polyenergetic SART (pSART) reconstruction algorithm for quantitative myocardial CT perfusion | |
CN110660111A (en) | PET scattering correction and image reconstruction method, device and equipment | |
TW201615152A (en) | Attenuation correction method for positron emission tomography image | |
CN109191462A (en) | A kind of CT anthropomorphic phantom generation method | |
CN105488826A (en) | Energy spectrum CT iterative imaging method and system based on EBP | |
CN106901768B (en) | Adaptive modulation method for reducing cone beam CT irradiation dose | |
CN110246199B (en) | Projection domain data noise removing method for energy spectrum CT | |
CN116630738A (en) | Energy spectrum CT imaging method based on depth convolution sparse representation reconstruction network | |
EP4105888A1 (en) | Systems and methods for computed tomography image reconstruction | |
CN110706299A (en) | Substance decomposition imaging method for dual-energy CT | |
US20230320688A1 (en) | Systems and methods for image artifact mitigation with targeted modular calibration | |
Xiu et al. | An Innovative Beam Hardening Correction Method for Computed Tomography Systems. | |
CN107730569A (en) | A kind of medical image artifact bearing calibration and device | |
CN110827370B (en) | Multi-energy CT (computed tomography) cyclic iterative reconstruction method for non-uniform-thickness component | |
CN112666194B (en) | Virtual digital DR image generation method and DR virtual simulation instrument |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |