CN112529796B - CNN medical CT image denoising method based on multi-feature extraction - Google Patents
CNN medical CT image denoising method based on multi-feature extraction Download PDFInfo
- Publication number
- CN112529796B CN112529796B CN202011406293.5A CN202011406293A CN112529796B CN 112529796 B CN112529796 B CN 112529796B CN 202011406293 A CN202011406293 A CN 202011406293A CN 112529796 B CN112529796 B CN 112529796B
- Authority
- CN
- China
- Prior art keywords
- data
- noise
- model
- image
- layer
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000000605 extraction Methods 0.000 title claims abstract description 22
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 28
- 230000005855 radiation Effects 0.000 claims description 16
- 238000011176 pooling Methods 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 9
- 210000003766 afferent neuron Anatomy 0.000 claims description 3
- 210000000268 efferent neuron Anatomy 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims 1
- 230000006698 induction Effects 0.000 abstract description 2
- 238000002591 computed tomography Methods 0.000 description 37
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000003759 clinical diagnosis Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013170 computed tomography imaging Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
- G06V10/464—Salient features, e.g. scale invariant feature transforms [SIFT] using a plurality of salient features, e.g. bag-of-words [BoW] representations
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a CNN medical CT image denoising method based on multi-feature extraction, which belongs to the technical field of medical image denoising, and particularly relates to a CNN medical CT image denoising method based on multi-feature extraction, which comprises the steps of establishing a model, establishing a plurality of groups of normal models and low-dose models, establishing a convolutional neural network, extracting multi-feature information, and integrating CT image denoising and a noise model, wherein the convolutional neural network has strong learning capacity and can lead the denoised data to be closer to real data through continuous summary and induction, meanwhile, the data of the noise model can be prevented from tending to be unified through removing the filling noise model, the data acquisition speed can be accelerated through adjusting the relation between the convolution step length and the size of a convolution kernel, so that the time consumption of image reconstruction is reduced, more feature information in the image can be captured by using the method of multi-feature extraction convolutional neural network, therefore, the collected characteristic information data are more enriched.
Description
Technical Field
The invention relates to the technical field of medical image denoising, in particular to a CNN medical CT image denoising method based on multi-feature extraction.
Background
In recent years, with the development of Computed Tomography (CT) technology, CT imaging has become more widely used in medical diagnosis. However, high-dose radiation generated in the CT scanning process may cause damage to the human body, and in order to reduce radiation damage, the CT technology field reduces the radiation dose while ensuring that the image quality meets the clinical diagnosis requirement. The existing main method for reducing radiation dose is to reduce tube current, but when the tube current is reduced, the number of photons received by a detector is also reduced, and the number of the received photons can cause projection data to be polluted by noise, so that a CT image reconstructed by the projection data not only has obvious noise, but also can generate streak artifacts, which can adversely affect clinical diagnosis. In response to these problems, many algorithms are proposed to improve the quality of low-dose CT images, and these algorithms can be classified into three major categories, namely, projection domain denoising algorithm, image reconstruction algorithm and image domain denoising algorithm.
The problem of data inconsistency easily occurs in the process of denoising by a projection domain denoising algorithm, so that new noise occurs in a reconstructed image, the algorithm complexity of the image reconstruction algorithm is high, the time consumption is long, the physical condition of a patient cannot be known in time, the image domain denoising algorithm easily destroys the structural information of the image, so that the edge characteristic information of the image is lost, a Convolutional Neural Network (CNN) is a variant of a multi-layer perceptron (MLP), the CNN has strong learning capability and mapping capability, and the method has greater advantages when the complex noise of a low-dose CT image is removed compared with a traditional method.
Disclosure of Invention
The invention aims to provide a CNN medical CT image denoising method based on multi-feature extraction, and aims to solve the problem that the existing CT image denoising method provided in the background technology has more defects.
In order to achieve the purpose, the invention provides the following technical scheme: a CNN medical CT image denoising method based on multi-feature extraction comprises the following steps:
the method comprises the following steps: establishing a model, namely establishing a normal model, shooting a CT image by normal dose radiation and establishing normal data coordinates, then establishing a low dose model, shooting the CT image by low dose radiation and establishing low dose data coordinates;
step two: establishing a plurality of groups of normal models and low dose models, then subtracting the coordinates of the low dose data of the same group from the coordinates of the normal data to obtain the coordinates of noise data, establishing the noise model according to the coordinates of the noise data, and determining the noise distribution according to the plurality of groups of noise models;
step three: constructing a convolutional neural network, wherein the convolutional neural network comprises an input layer, a hidden layer and an output layer, the hidden layer comprises a convolutional layer, a pooling layer and a full-connection layer, the convolutional layer is used as an afferent neuron, the pooling layer is used as a middle neuron, the full-connection layer is used as an efferent neuron, a CT image is input from the input layer, the convolutional layer is used for carrying out feature extraction on data of the CT image, the pooling layer is used for selecting and filtering information on the extracted feature data, the full-connection layer is used for carrying out nonlinear combination on the filtered feature data and transmitting the combined data to the output layer, and the output layer is used for classifying the data and outputting a classification label by using a normalized exponential function;
step four: extracting multi-feature information, wherein the convolutional layers perform feature extraction on data of the CT image by using convolutional cores, the size of the convolutional cores is n multiplied by n (n is an odd number), the convolution step length is (n-1)/2, the number of the convolutional layers is three, and the sizes of the convolutional cores of each convolutional layer are different;
step five: denoising the CT image, selecting characteristic data with high repeatability by a pooling layer, then filtering information, adding data of noise data coordinates to the shot actual data during filtering to obtain denoised CT image data, and carrying out nonlinear combination on the denoised CT image data by a full-connection layer to integrate the data with large difference with a mathematical expected value;
step six: and integrating the noise model, namely taking CT image data received by an input layer as a low-dose model, taking full-connection layer combined data as a normal model, subtracting the CT image data received by the input layer from the full-connection layer combined data to obtain a filling noise model, substituting the new filling noise model into the noise model after each new filling noise model is obtained, and deleting a group of old filling noise models when the number of groups of filling noise models reaches a set value and adding a group of new filling noise models each time.
Preferably, the normal model and the low-dose model in the same group both collect the same target of the same collection object, and when the noise model is called, the noise model corresponding to the shooting part is directly called.
Preferably, when the low-dose model is established, the data of the collected object is collected by respectively adopting radiation with different doses, so that noise models under different doses are determined, and when the noise models are called, the noise models with the same radiation dose as that during CT shooting are directly adopted.
Preferably, when the model is established, data acquisition is carried out on the body type of the acquisition object, then a three-dimensional coordinate function is established by using the body type data and the noise data of the acquisition object, and when the noise model is called, the noise model with the same body type as that of a person receiving CT photography is directly called.
Preferably, the convolution layer has convolution kernel sizes of (1-9) × (1-9).
Preferably, the number of groups of the filled noise models is 0.5-1 times of the number of groups of the noise models.
Preferably, when the filled noise model is deleted, an average value of the existing filled noise model data is calculated first, and then a group of filled noise models closest to the average value is deleted.
Compared with the prior art, the invention has the beneficial effects that:
1) the convolutional neural network has strong learning capacity, and can enable the denoised data to be closer to real data through continuous summarization and induction, and meanwhile, the data of a noise model can be prevented from tending to unification through removing a filling noise model;
2) the data acquisition speed can be increased by adjusting the relation between the convolution step length and the size of the convolution kernel, so that the time consumption of image reconstruction is reduced;
3) by utilizing the method of extracting the convolutional neural network by multiple features, more feature information in the image can be captured, so that the acquired feature information data is more enriched.
Drawings
FIG. 1 is a schematic diagram of a denoising process according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "disposed," "sleeved/connected," "connected," and the like are to be construed broadly, e.g., "connected," which may be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example (b):
referring to fig. 1, the present invention provides a technical solution: a CNN medical CT image denoising method based on multi-feature extraction comprises the following steps:
the method comprises the following steps: establishing a model, namely establishing a normal model, shooting a CT image by normal dose radiation and establishing normal data coordinates, then establishing a low dose model, shooting the CT image by low dose radiation and establishing low dose data coordinates;
step two: establishing a plurality of groups of normal models and low dose models, then subtracting the low dose data coordinates of the same group from the normal data coordinates to obtain noise data coordinates, establishing a noise model according to the noise data coordinates, and determining noise distribution according to the plurality of groups of noise models;
step three: constructing a convolutional neural network, wherein the convolutional neural network comprises an input layer, a hidden layer and an output layer, the hidden layer comprises a convolutional layer, a pooling layer and a full-connection layer, the convolutional layer is used as an afferent neuron, the pooling layer is used as a middle neuron, the full-connection layer is used as an efferent neuron, a CT image is input from the input layer, the convolutional layer is used for carrying out feature extraction on data of the CT image, the pooling layer is used for selecting and filtering information on the extracted feature data, the full-connection layer is used for carrying out nonlinear combination on the filtered feature data and transmitting the combined data to the output layer, and the output layer is used for classifying the data and outputting a classification label by using a normalized exponential function;
step four: extracting multi-feature information, wherein the convolutional layers perform feature extraction on data of the CT image by using convolutional cores, the size of the convolutional cores is n multiplied by n (n is an odd number), the convolution step length is (n-1)/2, the number of the convolutional layers is three, and the sizes of the convolutional cores of each convolutional layer are different;
step five: denoising the CT image, selecting characteristic data with high repeatability by a pooling layer, then filtering information, adding data of noise data coordinates to the shot actual data during filtering to obtain denoised CT image data, and carrying out nonlinear combination on the denoised CT image data by a full-connection layer to integrate the data with large difference with a mathematical expected value;
step six: and integrating the noise model, namely taking CT image data received by an input layer as a low-dose model, taking full-connection layer combined data as a normal model, subtracting the CT image data received by the input layer from the full-connection layer combined data to obtain a filling noise model, substituting the new filling noise model into the noise model after each new filling noise model is obtained, and deleting a group of old filling noise models when the number of groups of filling noise models reaches a set value and adding a group of new filling noise models each time.
The normal model and the low-dose model in the same group collect the same target of the same collection object, and when the noise model is called, the noise model corresponding to the shooting part is directly called.
When the low-dose model is established, the data of the collected object are collected by adopting radiation with different doses respectively, so that noise models under different doses are determined, and when the noise models are called, the noise models with the same radiation dose as that of CT shooting are directly adopted.
When the model is built, data acquisition is carried out on the body type of an acquisition object, then a three-dimensional coordinate function is built by using the body type data and the noise data of the acquisition object, and when the noise model is called, the noise model with the same body type as that of a person receiving CT shooting is directly called.
The convolution layer has convolution kernels of (1-9) × (1-9).
The number of groups of the filled noise models is 0.5-1 times of the number of groups of the noise models.
Filling noise model when deleting, firstly calculate the average value of the existing filling noise model data, and then delete the group of filling noise models closest to the average value.
While there have been shown and described the fundamental principles and essential features of the invention and advantages thereof, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing exemplary embodiments, but is capable of other specific forms without departing from the spirit or essential characteristics thereof; the present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein, and any reference signs in the claims are not to be construed as limiting the claims.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. A CNN medical CT image denoising method based on multi-feature extraction is characterized by comprising the following steps: the CNN medical CT image denoising method based on multi-feature extraction comprises the following steps:
the method comprises the following steps: establishing a model, namely establishing a normal model, shooting a CT image by normal dose radiation and establishing normal data coordinates, then establishing a low dose model, shooting the CT image by low dose radiation and establishing low dose data coordinates;
step two: establishing a plurality of groups of normal models and low dose models, then subtracting the low dose data coordinates of the same group from the normal data coordinates to obtain noise data coordinates, establishing a noise model according to the noise data coordinates, and determining noise distribution according to the plurality of groups of noise models;
step three: constructing a convolutional neural network, wherein the convolutional neural network comprises an input layer, a hidden layer and an output layer, the hidden layer comprises a convolutional layer, a pooling layer and a full-connection layer, the convolutional layer is used as an afferent neuron, the pooling layer is used as a middle neuron, the full-connection layer is used as an efferent neuron, a CT image is input from the input layer, the convolutional layer is used for carrying out feature extraction on data of the CT image, the pooling layer is used for selecting and filtering information on the extracted feature data, the full-connection layer is used for carrying out nonlinear combination on the filtered feature data and transmitting the combined data to the output layer, and the output layer is used for classifying the data and outputting a classification label by using a normalized exponential function;
step four: extracting multi-feature information, wherein the convolutional layers perform feature extraction on data of the CT image by using convolutional cores, the size of the convolutional cores is n multiplied by n (n is an odd number), the convolution step length is (n-1)/2, the number of the convolutional layers is three, and the sizes of the convolutional cores of all the convolutional layers are different;
step five: denoising the CT image, selecting characteristic data with high repeatability by a pooling layer, then filtering information, adding data of noise data coordinates to the shot actual data during filtering to obtain denoised CT image data, and carrying out nonlinear combination on the denoised CT image data by a full-connection layer to integrate the data with large difference with a mathematical expected value;
step six: the method comprises the steps of integrating a noise model, wherein CT image data received by an input layer are used as a low-dose model, data combined by full connecting layers are used as a normal model, then the CT image data received by the input layer are subtracted from the data combined by the full connecting layers to obtain a filling noise model, the new filling noise model is brought into the noise model after the new filling noise model is obtained each time, a group of old filling noise models are deleted when a group of new filling noise models are added each time after the group number of the filling noise models reaches a set value, the low-dose model is established, data collection is carried out on a collection object by adopting radiation with different doses respectively, so that the noise models under different doses are determined, and when the noise model is called, the noise model with the same radiation dose as that of CT shooting is directly adopted.
2. The CNN medical CT image denoising method based on multi-feature extraction as claimed in claim 1, wherein: and the normal model and the low-dose model in the same group both collect the same target of the same collection object, and when the noise model is called, the noise model corresponding to the shooting part is directly called.
3. The CNN medical CT image denoising method based on multi-feature extraction as claimed in claim 1, wherein: when the model is built, data acquisition is carried out on the body type of an acquisition object, then a three-dimensional coordinate function is built by using the body type data and the noise data of the acquisition object, and when the noise model is called, the noise model with the same body type as that of a person receiving CT shooting is directly called.
4. The CNN medical CT image denoising method based on multi-feature extraction as claimed in claim 1, wherein: the convolution layer has convolution kernel sizes of (1-9) × (1-9).
5. The CNN medical CT image denoising method based on multi-feature extraction as claimed in claim 1, wherein: the number of the groups of the filling noise models is 0.5-1 times of the number of the groups of the noise models.
6. The CNN medical CT image denoising method based on multi-feature extraction as claimed in claim 1, wherein: when the filling noise model is deleted, the average value of the existing filling noise model data is calculated firstly, and then a group of filling noise models closest to the average value is deleted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011406293.5A CN112529796B (en) | 2020-12-02 | 2020-12-02 | CNN medical CT image denoising method based on multi-feature extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011406293.5A CN112529796B (en) | 2020-12-02 | 2020-12-02 | CNN medical CT image denoising method based on multi-feature extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112529796A CN112529796A (en) | 2021-03-19 |
| CN112529796B true CN112529796B (en) | 2022-07-05 |
Family
ID=74998431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011406293.5A Expired - Fee Related CN112529796B (en) | 2020-12-02 | 2020-12-02 | CNN medical CT image denoising method based on multi-feature extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112529796B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118066A (en) * | 2015-09-16 | 2015-12-02 | 东南大学 | Low-dosage CT image decomposition method based on three-dimensional distinctive feature representation |
| CN109102550A (en) * | 2018-06-08 | 2018-12-28 | 东南大学 | Whole network low-dose CT imaging method and device based on convolution residual error network |
| CN110009613A (en) * | 2019-03-28 | 2019-07-12 | 东南大学 | Low-dose CT imaging method, apparatus and system based on the dense network of depth |
| CN110246199A (en) * | 2019-05-26 | 2019-09-17 | 天津大学 | A kind of projection domain data noise minimizing technology towards power spectrum CT |
| CN110827216A (en) * | 2019-10-23 | 2020-02-21 | 上海理工大学 | Multi-generator generation countermeasure network learning method for image denoising |
| CN111047524A (en) * | 2019-11-13 | 2020-04-21 | 浙江工业大学 | Low-dose CT lung image denoising method based on deep convolutional neural network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2748798B1 (en) * | 2011-12-13 | 2018-04-11 | Koninklijke Philips N.V. | Automatic determination of regularization factor for iterative image reconstruction with regularization and/or image de-noising |
| CN108492269B (en) * | 2018-03-23 | 2021-06-25 | 西安电子科技大学 | Low-dose CT image denoising method based on gradient regular convolution neural network |
-
2020
- 2020-12-02 CN CN202011406293.5A patent/CN112529796B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118066A (en) * | 2015-09-16 | 2015-12-02 | 东南大学 | Low-dosage CT image decomposition method based on three-dimensional distinctive feature representation |
| CN109102550A (en) * | 2018-06-08 | 2018-12-28 | 东南大学 | Whole network low-dose CT imaging method and device based on convolution residual error network |
| CN110009613A (en) * | 2019-03-28 | 2019-07-12 | 东南大学 | Low-dose CT imaging method, apparatus and system based on the dense network of depth |
| CN110246199A (en) * | 2019-05-26 | 2019-09-17 | 天津大学 | A kind of projection domain data noise minimizing technology towards power spectrum CT |
| CN110827216A (en) * | 2019-10-23 | 2020-02-21 | 上海理工大学 | Multi-generator generation countermeasure network learning method for image denoising |
| CN111047524A (en) * | 2019-11-13 | 2020-04-21 | 浙江工业大学 | Low-dose CT lung image denoising method based on deep convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112529796A (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Heinrich et al. | Residual U-net convolutional neural network architecture for low-dose CT denoising | |
| CN109559359B (en) | Artifact removal method for reconstructing image by sparse angle data based on deep learning | |
| CN112258415B (en) | Chest X-ray film super-resolution and denoising method based on generation countermeasure network | |
| CN109859285B (en) | Electrical impedance image reconstruction method based on cavity convolution network | |
| CN109598697B (en) | Determination of a two-dimensional mammography data set | |
| CN109697741A (en) | A kind of PET image reconstruction method, device, equipment and medium | |
| CN111325686A (en) | Low-dose PET three-dimensional reconstruction method based on deep learning | |
| CN112651885A (en) | Method and apparatus for reducing image recording noise | |
| CN112435164B (en) | Simultaneous super-resolution and denoising method for generating low-dose CT lung image based on multiscale countermeasure network | |
| CN103562960B (en) | For generating the assigned unit between the image-region of image and element class | |
| CN112598649B (en) | 2D/3D spine CT non-rigid registration method based on generation of countermeasure network | |
| Gajera et al. | CT-scan denoising using a charbonnier loss generative adversarial network | |
| WO2022000192A1 (en) | Ct image construction method, ct device, and storage medium | |
| CN113424222A (en) | System and method for providing stroke lesion segmentation using a conditional generation countermeasure network | |
| Wang et al. | Domain‐adaptive denoising network for low‐dose CT via noise estimation and transfer learning | |
| CN115546442A (en) | Multi-view stereo matching reconstruction method and system based on perception consistency loss | |
| Nguyen et al. | 3D Unet generative adversarial network for attenuation correction of SPECT images | |
| Green et al. | 3-D Neural denoising for low-dose Coronary CT Angiography (CCTA) | |
| CN116645283A (en) | Low-dose CT image denoising method based on self-supervision perceptual loss multi-scale convolutional neural network | |
| Pradhan et al. | Lung cancer detection using 3D convolutional neural networks | |
| CN110598724B (en) | Cell low-resolution image fusion method based on convolutional neural network | |
| Xia et al. | Deep residual neural network based image enhancement algorithm for low dose CT images | |
| CN112529796B (en) | CNN medical CT image denoising method based on multi-feature extraction | |
| CN116596785A (en) | Texture preserving low dose CT image denoising based on texture feature guidance | |
| CN110176045A (en) | A method of dual-energy CT image is generated by single energy CT image |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220705 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |