CN113643394A - Scattering correction method, device, computer equipment and storage medium - Google Patents
Scattering correction method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113643394A CN113643394A CN202110832302.5A CN202110832302A CN113643394A CN 113643394 A CN113643394 A CN 113643394A CN 202110832302 A CN202110832302 A CN 202110832302A CN 113643394 A CN113643394 A CN 113643394A
- Authority
- CN
- China
- Prior art keywords
- image
- projection data
- data
- cone beam
- scattering
- 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
Links
- 238000012937 correction Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000003062 neural network model Methods 0.000 claims abstract description 61
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000004590 computer program Methods 0.000 claims description 14
- 238000001914 filtration Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 8
- 238000012952 Resampling Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 description 18
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000001959 radiotherapy Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- 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
- G06T11/005—Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
-
- 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
-
- 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
- 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]
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application relates to a scatter correction method, a scatter correction device, a computer device and a storage medium. The method comprises the following steps: inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data; carrying out front projection processing on the processed cone beam CT image to obtain second projection data; obtaining initial scattering data according to the first projection data and the second projection data; obtaining target projection data according to the first projection data and the initial scattering data; and reconstructing the target projection data to obtain a cone beam CT image after scattering correction. The method can reduce the error of the obtained cone beam CT image after the scattering correction.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method and an apparatus for scatter correction, a computer device, and a storage medium.
Background
With the development of Cone Beam CT (CBCT), CBCT has been widely used in the fields of medical diagnosis, image-guided radiotherapy, etc., but due to the characteristics of its hardware structure, it is more seriously affected by scattered photons than conventional CT, so that it is necessary to perform scatter correction on the cone beam CT image.
In the traditional technology, the method for performing scattering correction on the cone beam CT image mainly comprises the steps of firstly calculating the actually scanned ray source energy spectrum and the detector response curve, then establishing a mathematical geometric model among the ray source, the object and the detector, which is the same as the actually scanned image, and finally calculating images received by the detector under different angles, wherein the process is also called front projection, and then obtaining a projection image after scattering correction by using an original cone beam CT projection image and the obtained front projection image.
However, the conventional scatter correction method has a problem that the error of the obtained scatter-corrected projection image is large.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a scatter correction method, apparatus, computer device, and storage medium that can reduce the resulting scatter-corrected projection image error in response to the above-described technical problems.
A method of scatter correction, the method comprising:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
In one embodiment, the obtaining target projection data according to the first projection data and the initial scattering data includes:
filtering the initial scattering data to obtain filtered scattering data corresponding to the initial scattering data;
and obtaining the target projection data according to the first projection data and the filtered scattering data.
In one embodiment, the obtaining the target projection data according to the first projection data and the filtered scattering data includes:
and determining the difference value of the first projection data and the filtered scattering data as the target projection data.
In one embodiment, the obtaining initial scattering data according to the first projection data and the second projection data includes:
and determining the difference value of the first projection data and the second projection data under the corresponding angle as the initial scattering data.
In one embodiment, before determining the difference image of the first projection image and the second projection image as the initial scattering image, the method further comprises:
dividing the first projection image by a preset reference image, and converting the gray value of the first projection image into a corresponding attenuation rate to obtain a processed projection image; the reference image is an image obtained by controlling the cone beam CT equipment to perform null scanning;
the determining a difference image of the first projection image and the second projection image as the initial scattering image includes:
and determining a difference image of the processed projection image and the second projection image as the initial scattering image.
In one embodiment, before the inputting the original cone beam CT image into the preset neural network model and obtaining the processed cone beam CT image, the method further includes:
performing scattering correction on the original cone-beam CT image by adopting a preset scattering correction method to obtain an initially corrected cone-beam CT image;
the method for inputting the original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image comprises the following steps:
and inputting the cone beam CT image subjected to initial correction into the neural network model to obtain the processed cone beam CT image.
In one embodiment, the method further comprises:
acquiring a sample cone beam CT image and a sample CT image;
resampling the sample CT image to obtain a resampled CT image; the number of layers of the resampled CT image corresponds to the number of layers of the sample cone-beam CT image one by one;
inputting the sample cone-beam CT image into a preset initial neural network model to obtain a processed sample cone-beam CT image;
and training the initial neural network model according to the processed sample cone beam CT image and the resampled CT image to obtain the neural network model.
A scatter correction device, the device comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, and the original cone beam CT image is obtained by reconstructing first projection data;
the processing module is used for carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
the second acquisition module is used for acquiring initial scattering data according to the first projection data and the second projection data;
the third acquisition module is used for acquiring target projection data according to the first projection data and the initial scattering data;
and the reconstruction module is used for reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
According to the scattering correction method, the device, the computer equipment and the storage medium, the original cone beam CT image is input into the preset neural network model, the neural network model can well reduce the noise of the input original cone beam CT image, the accuracy of each pixel CT value of the processed cone beam CT image is improved, the processed cone beam CT image and the original cone beam CT image can be kept consistent in geometry, the image effect is similar to the CT image, the noise is lower, the CT value is quite accurate, therefore, the processed cone beam CT image is subjected to front projection processing, the obtained second projection data not only has higher calculation accuracy, but also the problem that the edge of a reconstruction result has more artifacts due to the fact that the front projection geometry is not corresponding after the CT image and the CBCT image are registered in the traditional technology is solved, and a projection image can be accurately obtained for subsequent scattering image estimation, in addition, according to the first projection data and the second projection data, an initial scattering image can be accurately obtained, further, target projection data can be accurately obtained according to the first projection data initial scattering data, further, the target projection data can be reconstructed, and a cone beam CT image after scattering correction is obtained.
Drawings
FIG. 1 is a diagram of an exemplary embodiment of a scatter correction method;
FIG. 2 is a schematic flow chart diagram of a scatter correction method in one embodiment;
FIG. 3 is a schematic flow chart diagram of a scatter correction method in one embodiment;
FIG. 4 is a schematic flow chart diagram of a scatter correction method in one embodiment;
FIG. 5 is a schematic flow chart diagram of a scatter correction method in one embodiment;
fig. 6 is a block diagram showing the structure of a scatter correction apparatus according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The scattering correction method provided by the embodiment of the application can be applied to computer equipment shown in fig. 1. The computer device comprises a processor and a memory connected by a system bus, wherein a computer program is stored in the memory, and the steps of the method embodiments described below can be executed when the processor executes the computer program. Optionally, the computer device may further comprise a network interface, a display screen and an input device. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a nonvolatile storage medium storing an operating system and a computer program, and an internal memory. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. Optionally, the computer device may be a server, a personal computer, a personal digital assistant, other terminal devices such as a tablet computer, a mobile phone, and the like, or a cloud or a remote server, and the specific form of the computer device is not limited in the embodiment of the present application.
In one embodiment, as shown in fig. 2, a scatter correction method is provided, which is exemplified by the method applied to the computer device in fig. 1, and includes the following steps:
s201, inputting the original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data.
The cone beam CT image is a three-dimensional image obtained by an X-ray generator performing annular DR (digital projection) around a projection with a low dose (usually, a tube current is about 10 ma), and then recombining data obtained in an "intersection" after digital projection for multiple times (180-360 times) around the projection in a computer. Specifically, the computer equipment inputs an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image; wherein the original cone beam CT image is obtained by reconstructing the first projection data. It should be noted that the obtained processed cone-beam CT image is an image subjected to noise reduction and artifact removal processing, and has similar effect to the CT image, so that the obtained processed cone-beam CT image is more consistent with the actual scanning result. Optionally, the network structure of the preset neural network model may be any network structure, and this embodiment is not limited herein. Alternatively, the first projection data may be a projection value or a projection image.
S202, carrying out front projection processing on the processed cone beam CT image to obtain second projection data.
Specifically, the computer device performs a front projection process on the processed cone beam CT image obtained above to obtain second projection data. It should be noted that the second projection data obtained here is a front projection image at a plurality of projection angles, where the basic principle of front projection is to calculate the radiation source energy spectrum and the detector response curve of actual scanning, then establish a mathematical geometric model between the radiation source, the object, and the detector, which is the same as that of actual scanning, and finally calculate the images received by the detector at different angles, and this process is also referred to as front projection. Since the front projection process only considers the attenuation effect of the substance on the main ray and does not consider compton scattering, the signals received by the detector are all main ray signals contributing to reconstruction after attenuation (in cone beam CT scanning, it is generally considered that photons received by the detector are mainly main ray photons and scattered photons which are unfavorable for reconstruction). Alternatively, the second projection data may be a projection value or a projection image.
S203, obtaining initial scattering data according to the first projection data and the second projection data.
Specifically, the computer device obtains initial scattering data according to the first projection data and the second projection data. Optionally, the computer device may adjust the gray value of the second projection data to make the gray value of the first projection data and the gray value of the second projection data have the same meaning, and then subtract the two data to obtain the initial scattering data.
And S204, obtaining target projection data according to the first projection data and the initial scattering data.
And the target projection data is the projection data after the scattering correction corresponding to the initial scattering data. Optionally, the computer device may determine a difference between the first projection data and the initial scattering data as the target projection data.
S205, reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
It can be understood that, since the processed cone beam CT image is subjected to the front projection processing, projection data at a plurality of projection angles are obtained, so that the initial scattering data obtained by the computer device is also data at a plurality of angles, and further the target projection data obtained by the computer device is also a plurality of, and the computer device performs image reconstruction on the plurality of target projection data, so as to obtain the cone beam CT image after the scattering correction.
In the scattering correction method, the original cone beam CT image is input into the preset neural network model, the neural network model can well reduce the noise of the input original cone beam CT image, the accuracy of the CT value of each pixel of the processed cone beam CT image is improved, the processed cone beam CT image and the original cone beam CT image can keep consistent in geometry, the image effect is similar to the CT image, the noise is low, the CT value is very accurate, the processed cone beam CT image is subjected to front projection processing, the obtained second projection data not only has higher calculation accuracy, but also overcomes the problem that the edge of a reconstruction result has more artifacts due to the fact that the front projection geometry does not correspond after the CT image and the CBCT image are registered in the prior art, and the method can accurately obtain a projection image for subsequent scattering image estimation, in addition, according to the first projection data and the second projection data, the initial scattering image can be accurately obtained, further, the target projection data can be accurately obtained according to the first projection data and the initial scattering data, further, the target projection data can be reconstructed, and the cone beam CT image after scattering correction is obtained.
In the above scenario in which the target projection data is obtained according to the first projection data and the initial scattering data, in an embodiment, as shown in fig. 3, the step S204 includes:
s301, filtering the initial scattering data to obtain filtered scattering data corresponding to the initial scattering data.
Specifically, the computer device performs filtering processing on the obtained initial scattering data to obtain a filtered scattering image corresponding to the initial scattering data. Optionally, in this embodiment, the filtering process performed on the initial scattering data may be an average filtering process, or a low-pass filtering process.
S302, obtaining target projection data according to the first projection data and the filtered scattering data.
Specifically, the computer device obtains target projection data according to the first projection data and the obtained filtered scattering data. And the target projection data is the projection data after the scattering correction corresponding to the filtered scattering data. Optionally, the computer device may determine the difference data between the first projection image and the filtered scatter data as the target projection data.
In this embodiment, because the computer device performs filtering processing on the initial scattering data, the filtered scattering data corresponding to the initial scattering data can be accurately obtained, so that a projection image after scattering correction, that is, a target projection image, corresponding to the filtered scattering image can be accurately obtained according to the first projection data and the obtained filtered scattering data, and the accuracy of the obtained target projection image is improved.
In the above scenario in which the initial scattering data is obtained according to the first projection data and the second projection data, in an embodiment, the step S203 includes: and determining the difference value of the first projection data and the second projection data under the corresponding angle as initial scattering data.
It can be understood that, since the processed cone beam CT image is subjected to the front projection processing, projection data of a plurality of projection angles are obtained, that is, the obtained second projection data is projection data under a plurality of projection angles, and then the computer device obtains the initial scattering data according to the second projection data and the first projection data under the corresponding angle of the second projection data in a scene where the initial scattering data is obtained according to the first projection data and the second projection data, that is, the computer device may determine a difference value of the first projection data and the second projection data under the corresponding angle as the initial scattering data. Optionally, before determining, by the computer device, a difference value of the first projection data and the second projection data at a corresponding angle as initial scattering data, the computer device may further divide the first projection data by preset reference data, convert a gray value of the first projection data into a corresponding attenuation rate, obtain processed projection data, and determine, by the computer device, the difference value of the processed projection data and the second projection data at a corresponding angle as the initial scattering data; wherein the preset reference data is data obtained by controlling the cone beam CT equipment to perform null scan.
In this embodiment, the process of determining the difference value of the first projection data and the second projection data at the corresponding angle as the initial scattering data by the computer device is relatively simple, so that the computer device can quickly obtain the initial scattering data, thereby improving the efficiency of obtaining the initial scattering data.
In the scenario where the original cone beam CT image is input into the preset neural network model, before inputting the original cone beam CT image into the preset neural network model, the computer device may perform preset scatter correction processing on the original cone beam CT image, and input the corrected image into the preset neural network model, in an embodiment, before S201, the method further includes: performing scattering correction on the original cone-beam CT image by adopting a preset scattering correction method to obtain an initially corrected cone-beam CT image; the S201 includes: and inputting the cone beam CT image subjected to initial correction into a neural network model to obtain a processed cone beam CT image.
Specifically, the computer device performs scatter correction on the original cone-beam CT image by using a preset scatter correction method to obtain an initially corrected cone-beam CT image, and inputs the initially corrected cone-beam CT image into the neural network model to obtain the processed cone-beam CT image. Optionally, the preset scattering correction method may be an SKS scattering correction method, or may be another scattering correction method, for example, MC simulation, or the like.
In this embodiment, the computer device may perform a scatter correction on the original cone-beam CT image by using a preset scatter correction method to obtain an initially corrected cone-beam CT image, and the initially corrected cone-beam CT image is a corrected image, so that the obtained initially corrected cone-beam CT image is input to the neural network model, and the neural network model may accurately process the input initially corrected cone-beam CT image, thereby improving the accuracy of the obtained processed cone-beam CT image.
In the above-mentioned scene where the original cone beam CT image is input into the preset neural network model to obtain the processed cone beam CT image, the preset neural network model is a pre-trained model, and in an embodiment, as shown in fig. 4, the method further includes,
s401, acquiring a sample cone beam CT image and a sample CT image.
Specifically, a computer device acquires a sample cone-beam CT image and a sample CT image. Optionally, in this embodiment, the sample cone-beam CT image and the sample CT image may be images corresponding to the same part of the same examiner, or images corresponding to different parts of different examiners, and it can be understood that training using a sample corresponding to the same part of the same examiner can achieve a better training effect for the network. Alternatively, the computer device may acquire the sample cone-beam CT image and the sample CT image from a PACS (Picture Archiving and Communication Systems) server.
S402, resampling the sample CT image to obtain a resampled CT image; the number of layers of the resampled CT image corresponds to the number of layers of the sample cone beam CT image one to one.
Specifically, the computer device resamples the acquired sample CT image to obtain a resampled CT image, wherein the number of layers of the resampled CT image corresponds to the number of layers of the sample cone beam CT image one to one. It should be noted that the layer thicknesses of the same part scanned by the sample cone beam CT image and the sample CT image are different, and the layer thickness of the sample CT image is generally larger than that of the sample cone beam CT image, so that the sample CT image needs to be resampled to obtain resampled CT images corresponding to the layer number of the sample cone beam CT image one to one.
And S403, inputting the sample cone beam CT image into a preset initial neural network model to obtain a processed sample cone beam CT image.
Specifically, the computer device inputs the acquired sample cone beam CT image into a preset initial neural network model to obtain a processed sample cone beam CT image. The obtained processed sample cone-beam CT image is an image subjected to noise reduction and artifact removal processing, and has an effect similar to that of the sample CT image. Optionally, the network structure of the preset initial neural network model may be any network structure, and this embodiment is not limited herein.
S404, training the initial neural network model according to the processed sample cone beam CT image and the resampled CT image to obtain the neural network model.
Specifically, the computer device trains the initial neural network model according to the obtained processed sample cone-beam CT image and the resampled CT image to obtain the neural network model. Optionally, the computer device may obtain a loss function value of the initial neural network model according to the obtained processed sample cone-beam CT image and the resampled CT image, and train the initial neural network model according to the loss function value to obtain the neural network model. Optionally, the computer device may determine, as the neural network model, an initial neural network model corresponding to a time when the value of the loss function reaches a stable value or a minimum value. For example, the initial neural network model may be a Cycle-GAN network, and the computer device may determine, when the effect of the processed sample cone-beam CT image generated by the initial Cycle-GAN network reaches a target effect, a corresponding Cycle-GAN network at that time as the neural network model.
In this embodiment, the computer device may obtain the resampled CT images corresponding to the number of layers of the obtained sample cone beam CT image by resampling the obtained sample CT image, so that the sample cone beam CT image may be input into the preset initial neural network model to obtain the processed sample cone beam CT image, and the initial neural network model may be accurately trained according to the processed sample cone beam CT image and the obtained resampled CT image, thereby improving the accuracy of the obtained neural network model.
To facilitate understanding by those skilled in the art, the scatter correction method provided in the present application is described in detail below, and as shown in fig. 5, the method may include:
s1, performing scattering correction on the original cone beam CT projection image by using an SKS scattering correction method to obtain a cone beam CT image after primary correction;
s2, inputting the cone beam CT image preliminarily corrected by the SKS method into a trained neural network model to obtain a processed cone beam CT image;
s3, performing front projection by using the processed CT image generated by the neural network model, and acquiring projection images under a plurality of projection angles, wherein the gray value of the image is the corresponding ray attenuation rate;
s4, dividing the original cone beam CT projection image by the blank scanning image, converting the gray value of the original cone beam CT projection image into a corresponding attenuation rate, and then subtracting the projection image generated under the corresponding angle to obtain an initial scattering image;
s5, performing mean filtering and low-pass filtering on the obtained initial scattering image to make the initial scattering image smoother;
s6, converting the gray value of the original cone beam CT image into a corresponding ray attenuation rate, and then subtracting the filtered scattering image to obtain a projection image after scattering correction;
and S7, reconstructing the projection after the scatter correction by using an FDK reconstruction algorithm, and finally obtaining the cone beam CT image after the scatter correction.
It should be noted that, for the descriptions in S1-S7, reference may be made to the descriptions related to the above embodiments, and the effects are similar, and the description of this embodiment is not repeated herein.
It should be understood that although the various steps in the flow charts of fig. 2-5 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-5 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 6, there is provided a scatter correction apparatus comprising: the device comprises a first acquisition module, a processing module, a second acquisition module, a third acquisition module and a reconstruction module, wherein:
the first acquisition module is used for inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
the processing module is used for carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
the second acquisition module is used for acquiring initial scattering data according to the first projection data and the second projection data;
the third acquisition module is used for acquiring target projection data according to the first projection data and the initial scattering data;
and the reconstruction module is used for reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the third obtaining module includes: a filtering unit and a first obtaining unit, wherein:
and the filtering unit is used for filtering the initial scattering data to obtain the filtered scattering data corresponding to the initial scattering data.
And the first acquisition unit is used for acquiring target projection data according to the first projection data and the filtered scattering data.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the first obtaining unit is configured to determine a difference between the first projection data and the filtered scatter data as the target projection data.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the second obtaining module includes: a second acquisition unit, wherein:
and the second acquisition unit is used for determining the difference value of the first projection data and the second projection data under the corresponding angle as initial scattering data.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the apparatus further includes: a fourth acquisition module, wherein:
the fourth acquisition module is used for dividing the first projection data by preset reference data and converting the gray value of the first projection data into a corresponding attenuation rate to obtain processed projection data; the reference data is data obtained by controlling the cone beam CT equipment to perform null scanning;
and the second acquisition unit is used for determining the difference value of the processed projection data and the second projection data under the corresponding angle as an initial scattering image.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the apparatus further includes: a correction module, wherein:
and the correction module is used for performing scattering correction on the original cone-beam CT image by adopting a preset scattering correction method to obtain an initially corrected cone-beam CT image.
And the first acquisition module is used for inputting the initially corrected cone beam CT image into the neural network model to obtain a processed cone beam CT image.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
On the basis of the foregoing embodiment, optionally, the apparatus further includes: a fifth obtaining module, a resampling module, a sixth obtaining module and a training module, wherein:
and the fifth acquisition module is used for acquiring the sample cone beam CT image and the sample CT image.
The resampling module is used for resampling the sample CT image to obtain a resampled CT image; the number of layers of the resampled CT image corresponds to the number of layers of the sample cone beam CT image one to one.
And the sixth acquisition module is used for inputting the sample cone beam CT image into a preset initial neural network model to obtain a processed sample cone beam CT image.
And the training module is used for training the initial neural network model according to the processed sample cone beam CT image and the resampled CT image to obtain the neural network model.
The scatter correction device provided in this embodiment may implement the above method embodiments, and the implementation principle and technical effect are similar, which are not described herein again.
For the specific definition of the scatter correction device, reference may be made to the above definition of the scatter correction method, which is not described herein again. The modules in the scattering correction device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
The implementation principle and technical effect of the computer-readable storage medium provided by the above embodiments are similar to those of the above method embodiments, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method of scatter correction, the method comprising:
inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, wherein the original cone beam CT image is obtained by reconstructing first projection data;
carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
obtaining initial scattering data according to the first projection data and the second projection data;
obtaining target projection data according to the first projection data and the initial scattering data;
and reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
2. The method of claim 1, wherein obtaining target projection data from the first projection data and the initial scatter data comprises:
filtering the initial scattering data to obtain filtered scattering data corresponding to the initial scattering data;
and obtaining the target projection data according to the first projection data and the filtered scattering data.
3. The method of claim 2, wherein deriving the target projection data from the first projection data and the filtered scatter data comprises:
and determining the difference value of the first projection data and the filtered scattering data as the target projection data.
4. The method of claim 1, wherein said deriving initial scatter data from said first projection data and said second projection data comprises:
and determining the difference value of the first projection data and the second projection data under the corresponding angle as the initial scattering data.
5. The method of claim 4, wherein prior to determining the difference between the first projection data and the second projection data at corresponding angles as the initial scatter data, the method further comprises:
dividing the first projection data by preset reference data, and converting the gray value of the first projection data into a corresponding attenuation rate to obtain processed projection data; the reference data is data obtained by controlling the cone beam CT equipment to perform null scanning;
determining the difference between the first projection data and the second projection data at the corresponding angle as the initial scattering data, including:
and determining the difference value of the processed projection data and the second projection data under the corresponding angle as the initial scattering data.
6. The method according to any one of claims 1 to 5, wherein before inputting the original cone beam CT image into a preset neural network model and obtaining the processed cone beam CT image, the method further comprises:
performing scattering correction on the original cone-beam CT image by adopting a preset scattering correction method to obtain an initially corrected cone-beam CT image;
the method for inputting the original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image comprises the following steps:
and inputting the cone beam CT image subjected to initial correction into the neural network model to obtain the processed cone beam CT image.
7. The method of claim 1, further comprising:
acquiring a sample cone beam CT image and a sample CT image;
resampling the sample CT image to obtain a resampled CT image; the number of layers of the resampled CT image corresponds to the number of layers of the sample cone-beam CT image one by one;
inputting the sample cone-beam CT image into a preset initial neural network model to obtain a processed sample cone-beam CT image;
and training the initial neural network model according to the processed sample cone beam CT image and the resampled CT image to obtain the neural network model.
8. A scatter correction device, characterized in that the device comprises:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for inputting an original cone beam CT image into a preset neural network model to obtain a processed cone beam CT image, and the original cone beam CT image is obtained by reconstructing first projection data;
the processing module is used for carrying out front projection processing on the processed cone beam CT image to obtain second projection data;
the second acquisition module is used for acquiring initial scattering data according to the first projection data and the second projection data;
the third acquisition module is used for acquiring target projection data according to the first projection data and the initial scattering data;
and the reconstruction module is used for reconstructing the target projection data to obtain a cone beam CT image after scattering correction.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110832302.5A CN113643394A (en) | 2021-07-22 | 2021-07-22 | Scattering correction method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110832302.5A CN113643394A (en) | 2021-07-22 | 2021-07-22 | Scattering correction method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113643394A true CN113643394A (en) | 2021-11-12 |
Family
ID=78418094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110832302.5A Pending CN113643394A (en) | 2021-07-22 | 2021-07-22 | Scattering correction method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113643394A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117218868A (en) * | 2023-11-07 | 2023-12-12 | 福建理工大学 | Traffic flow prediction method based on geometric scattering map network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574828A (en) * | 2015-12-22 | 2016-05-11 | 沈阳东软医疗系统有限公司 | Image scattering correction method, device and apparatus |
| CN109064521A (en) * | 2018-07-25 | 2018-12-21 | 南京邮电大学 | A kind of CBCT using deep learning removes pseudo- image method |
| CN111127579A (en) * | 2019-12-26 | 2020-05-08 | 南京安科医疗科技有限公司 | CT reconstruction method for removing cone beam artifacts based on neural network |
| CN112435307A (en) * | 2020-11-26 | 2021-03-02 | 浙江大学 | Deep neural network assisted four-dimensional cone beam CT image reconstruction method |
-
2021
- 2021-07-22 CN CN202110832302.5A patent/CN113643394A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574828A (en) * | 2015-12-22 | 2016-05-11 | 沈阳东软医疗系统有限公司 | Image scattering correction method, device and apparatus |
| CN109064521A (en) * | 2018-07-25 | 2018-12-21 | 南京邮电大学 | A kind of CBCT using deep learning removes pseudo- image method |
| CN111127579A (en) * | 2019-12-26 | 2020-05-08 | 南京安科医疗科技有限公司 | CT reconstruction method for removing cone beam artifacts based on neural network |
| CN112435307A (en) * | 2020-11-26 | 2021-03-02 | 浙江大学 | Deep neural network assisted four-dimensional cone beam CT image reconstruction method |
Non-Patent Citations (1)
| Title |
|---|
| HEHE CUI 等: "Local-filtration Based Scatter Correction for Conebeam CT Using Planning CT as a Prior", 《2019 INTERNATIONAL CONFERENCE ON MEDICAL IMAGING PHYSICS AND ENGINEERING (ICMIPE)》, pages 1 - 6 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117218868A (en) * | 2023-11-07 | 2023-12-12 | 福建理工大学 | Traffic flow prediction method based on geometric scattering map network |
| CN117218868B (en) * | 2023-11-07 | 2024-03-22 | 福建理工大学 | Traffic flow prediction method based on geometric scattering map network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110570492B (en) | CT artifact suppression method, device and medium based on neural network | |
| US10964072B2 (en) | Methods, systems, and media for noise reduction in computed tomography images | |
| EP3362987B1 (en) | System and method for image correction | |
| CN111080584B (en) | Quality control method for medical image, computer device and readable storage medium | |
| JP2020506742A (en) | Deep learning-based estimation of data for use in tomographic reconstruction | |
| US10867375B2 (en) | Forecasting images for image processing | |
| JP6214226B2 (en) | Image processing apparatus, tomography apparatus, image processing method and program | |
| US20140328532A1 (en) | Attenuation correction method and device for image in pet system | |
| CN108606805B (en) | Scattering correction method, device and equipment | |
| CN110084868B (en) | Image correction method, apparatus, computer device, and readable storage medium | |
| US20200175730A1 (en) | Imaging system extended field-of-view | |
| CN112581554B (en) | CT imaging method, device, storage equipment and medical imaging system | |
| US20230206404A1 (en) | Image processing apparatus, image processing method, and computer-readable medium | |
| CN106228520B (en) | Image enchancing method and device | |
| CN114943780A (en) | Image reconstruction processing method and system and medical imaging equipment | |
| CN113643394A (en) | Scattering correction method, device, computer equipment and storage medium | |
| US20230419455A1 (en) | System and method for image correction | |
| US11484279B2 (en) | Systems to assess projection data inconsistency | |
| US10984564B2 (en) | Image noise estimation using alternating negation | |
| US20220414832A1 (en) | X-ray imaging restoration using deep learning algorithms | |
| CN114913259A (en) | Truncation artifact correction method, CT image correction method, apparatus, and medium | |
| CN114004738A (en) | Method, apparatus, device and medium for processing digital mammary tomography image | |
| CN113592968A (en) | Method and device for reducing metal artifacts in tomographic images | |
| JP2021036969A (en) | Machine learning device, machine learning method, and program | |
| CN112446931A (en) | Reconstruction data processing method and device, medical imaging system and storage medium |
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 |