CN110264435B - Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium - Google Patents
Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110264435B CN110264435B CN201910504782.5A CN201910504782A CN110264435B CN 110264435 B CN110264435 B CN 110264435B CN 201910504782 A CN201910504782 A CN 201910504782A CN 110264435 B CN110264435 B CN 110264435B
- Authority
- CN
- China
- Prior art keywords
- image
- mip
- dose
- mip image
- pet
- 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
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/08—Volume rendering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- 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
- G06T2207/10104—Positron emission tomography [PET]
-
- 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]
Abstract
The application relates to a method, a device, a computer device and a storage medium for enhancing a low-dose MIP image. The method comprises the following steps: acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer corresponding to a body position; learning the second MIP image through the feedback of the first MIP image by utilizing the generated antagonistic neural network, and establishing an image enhancement model; and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into an image enhancement model for image enhancement, and outputting an MIP enhanced image. By adopting the method, the MIP enhanced image with high quality can be rapidly output by utilizing the image enhanced model; thereby meeting the quality requirement of the MIP image with low dose as a reference picture.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method and an apparatus for enhancing a low-dose MIP image, a computer device, and a storage medium.
Background
Direct Volume Rendering (DVR) is one of the most effective volume data visualization methods, and is widely applied in many fields such as medicine, geography, physics, and the like. The Transfer Function (TF) is responsible for mapping volume data attributes (e.g. density values, gradient modes, etc.) to optical attributes such as color, transparency, etc., the quality of which has a decisive influence on the effectiveness of the DVR. As a special DVR method, Maximum Intensity Projection (MIP) projects the maximum intensity value on the projection light onto the screen without a transfer function, has the advantages of simplicity, practicality and the like, and is widely applied in the fields of medicine and the like.
In a PET-MR device, it is necessary to generate a PET image of a set of MIP images. The MIP image has the effect of: as a reference picture, MR picture localization is assisted. In order to obtain high quality PET images, a full dose of radiotracer is very necessary, but it causes problems with potential patient health damage. And simply by reducing the dose of MIP images, the quality of MIP images is reduced due to more noise, and the quality requirement as a reference image cannot be met.
Disclosure of Invention
In view of the above, it is necessary to provide a method, an apparatus, a computer device and a storage medium for enhancing a low-dose MIP image, which can meet the quality requirement of the low-dose MIP image as a reference map.
A method of enhancement of a low dose MIP image, the method comprising:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer corresponding to a body position;
learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting an MIP enhanced image.
In one embodiment, the reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low-dose radiotracer in the corresponding body position includes:
reconstructing the first PET image under different doses according to preset reconstruction parameters to obtain a corresponding second PET image; and extracting the pixel value of the second PET image to obtain a second MIP image.
In one embodiment, the learning of the second MIP image through the feedback of the first MIP image by using the generative neural network to build the image enhancement model includes:
building a generated antagonistic neural network, wherein the generated antagonistic neural network comprises a generating module and a judging module;
taking the first MIP image as label data, and performing supervised learning on a second MIP image by using the generation module to obtain a secondary image;
distinguishing the secondary image and the first MIP image by using the distinguishing module to obtain a distinguishing result;
and finishing training after the judgment result meets the termination condition to obtain the image enhancement model.
In one embodiment, the method further comprises the following steps:
and after the antagonistic neural network is built and generated, performing one-to-one shuffle processing on the first MIP image and the second MIP image.
In one embodiment, the determination result is fed back to the generation module, and the generation module is used for performing continuous supervised learning on the second MIP image to obtain a new secondary image.
In one embodiment, the method further comprises the following steps:
after the MIP enhanced image is output, the MIP enhanced image is stored as a positioning-assisted reference map.
Correspondingly, the invention also provides a device for enhancing the low-dose MIP image, which comprises an acquisition module, an establishment module and an enhancement module:
the acquisition module is used for acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, reconstructing the first MIP image and the corresponding first PET image, and obtaining a second MIP image under a low dose of radioactive tracer corresponding to a body position;
the establishing module is used for learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
the enhancement module is used for acquiring a current MIP image under the current low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting the MIP enhanced image.
In one embodiment, the storage module further:
and the storage module is used for storing the MIP enhanced image as a reference map for assisting positioning after the MIP enhanced image is output.
Correspondingly, the invention also provides a computer device, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to realize the following steps:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer of a corresponding body position;
learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting an MIP enhanced image.
Accordingly, the present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer of a corresponding body position;
learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting an MIP enhanced image.
According to the method and the device for enhancing the low-dose MIP image, the computer equipment and the storage medium, the obtained first MIP image and the corresponding first PET image are reconstructed firstly, and a second MIP image under the low-dose radioactive tracer of the corresponding body position is obtained; then, learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model; the image enhancement model is directly utilized to carry out image enhancement on the current MIP image, so that the MIP enhanced image with high quality is rapidly output; thereby meeting the quality requirement of the MIP image with low dose as a reference picture.
Drawings
FIG. 1 is a diagram of an embodiment of an application environment of a method for enhancing a low-dose MIP image;
FIG. 2 is a schematic flow chart illustrating a method for enhancing a low-dose MIP image in one embodiment;
FIG. 3 is a schematic flowchart of step S200 in the embodiment of FIG. 2;
FIG. 4 is a schematic flow chart of a method for enhancing a low-dose MIP image in another embodiment;
FIG. 5 is a block diagram of an embodiment of a low dose MIP image enhancement device;
FIG. 6 is a block diagram of a low-dose MIP image enhancement apparatus according to another embodiment;
FIG. 7 is a diagram illustrating an internal structure of a computer device according to an embodiment.
In the figure: 1. a terminal; 2. a server; 100. an acquisition module; 200. establishing a module; 300. a boost module; 400. and a storage module.
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 enhancement method of the low-dose MIP image can be applied to the application environment as shown in figure 1. Wherein, the terminal 1 communicates with the server 2 through the network. Reconstructing a first MIP image and a corresponding first PET image in the terminal 1 to obtain a second MIP image under a low-dose radioactive tracer of a corresponding body position; establishing an image enhancement model by utilizing a generated antagonistic neural network according to the first MIP image and the second MIP image; and acquiring a current MIP image under the current low-dose radioactive tracer, inputting the current MIP image into an image enhancement model for image enhancement, and outputting an MIP enhanced image. After the image enhancement model is established, the terminal 1 may upload the image enhancement model to the server 2 through the network, so as to facilitate loading and use of other terminals 1. The terminal 1 may be various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices connected with the PET-MR device, the terminal 1 may also be the PET-CT device, and the server 2 may be implemented by an independent server or a server cluster composed of a plurality of servers.
In one embodiment, as shown in fig. 2, a method for enhancing a low-dose MIP image is provided, which is described by taking the method as an example for being applied to the terminal in fig. 1, and includes the following steps:
s100, acquiring a first MIP image and a corresponding first PET image under a full-dose radioactive tracer, reconstructing the first MIP image and the corresponding first PET image, and acquiring a second MIP image under a low-dose radioactive tracer corresponding to a body position;
clinical data measured by equipment under a full dose of radiotracer were obtained as training samples. The clinical data for each patient includes a first MIP image and a corresponding first PET image; the second MIP image and the corresponding second PET image under low dose radiotracer can be measured by the device, but since the low dose MIP image again degrades due to more noise, additional algorithms are needed to reduce noise, increasing computational burden. A low dose is referred to a full dose, and a low dose is understood to mean a low dose as long as it is not a full dose. In one embodiment, for the same patient, a first MIP image and a corresponding first PET image may be reconstructed resulting in a second MIP image under a low dose radiotracer for the corresponding body position. The first MIP image is a Maximum Intensity Projection (MIP) image acquired by the device under a full dose of the radiotracer. The first PET image is generated with the first MIP image as a reference map, and the image format thereof may be a DICOM (medical image format) image.
S200, learning the second MIP image through the feedback of the first MIP image by utilizing the generated countermeasure neural network, and establishing an image enhancement model;
in one embodiment, the second MIP image continuously feeds back the learning, image enhancement model built by using the information fed back by the first MIP image by using the generation of a countermeasure neural network (GAN). The image enhancement model may be stored in a database and loaded. In other embodiments, other neural networks may be used, and are not limited in this regard.
S300, obtaining a current MIP image under the current low-dose radioactive tracer, inputting the current MIP image into an image enhancement model for image enhancement, and outputting an MIP enhanced image.
The use cases may be different in different application scenarios. In one embodiment, the image enhancement model can be maintained in the PETMR product release, after the current MIP image of the MIP workflow is generated, the trained image enhancement model is directly loaded, and the image enhancement is performed on the current MIP image, so that the MIP enhancement image with high quality is rapidly output.
According to the method for enhancing the low-dose MIP image, the obtained first MIP image and the corresponding first PET image are reconstructed, and a second MIP image under the low-dose radioactive tracer of the corresponding body position is obtained; then, learning the second MIP image through the feedback of the first MIP image by utilizing the generated countermeasure neural network, and establishing an image enhancement model; the image enhancement model is directly utilized to carry out image enhancement on the current MIP image, so that the MIP enhanced image with high quality is rapidly output; thereby meeting the quality requirement of the MIP image with low dose as a reference picture.
In one embodiment, reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose radiotracer corresponding to the body position comprises the steps of:
reconstructing the first PET image under different doses according to preset reconstruction parameters to obtain a corresponding second PET image; and extracting the pixel value of the second PET image to obtain a second MIP image.
Specifically, the first PET image is processed by offline-recon, and dynamic reconstruction parameters are set for reconstruction, so that a corresponding second PET image at a low dose is obtained. And then storing the pixel value contents of a series of second PET images as gray level images, namely obtaining second MIP images. In one embodiment, dynamic reconstruction parameters are set for reconstruction, and the time for the ablation may be set to 10 minutes for obtaining a second PET image of the patient with a low dose using the data ablation function.
The following describes the establishment of the image enhancement model in detail.
Specifically, as shown in fig. 3, step S200 includes the following steps;
s210, building a generated antagonistic neural network, wherein the generated antagonistic neural network comprises a generating module and a judging module;
s220, taking the first MIP image as label data, and performing supervised learning on the second MIP image by using the generation module to obtain a secondary image;
s230, distinguishing the sub-image and the first MIP image by using a distinguishing module to obtain a distinguishing result;
and S240, finishing training after the judgment result meets the termination condition to obtain an image enhancement model.
In one embodiment, the constructed generation confrontation neural network at least comprises a generation module and a discrimination module; the generation module and the discrimination module are mutually independent and are mutually game-learning. In this embodiment, the generation of the G portion of the anti-neural network is improved such that the input is not a simple noise signal, but a two-dimensional, low-dose second MIP image. Generating a model to generate a sub-image by using the first MIP image as label data, distinguishing the generated sub-image from the first MIP image by learning of the model, improving the generation model according to a distinguishing result of the distinguishing model, and generating a new sub-image; if the judgment result does not meet the termination, generating a model to continuously generate a new sub-image according to the feedback of the judgment result, and continuously learning and distinguishing the newly generated sub-image and the first MIP image by the judgment model; and finishing training until the judgment result meets the termination condition to obtain the image enhancement model.
In the generative model, the loss function of the generative model comprises a weighting of the two parts of the first paradigm of generating the side image produced by the anti-neural network and the first MIP image. In the continuous learning process, the auxiliary image and the first MIP image are closer and closer by utilizing the weighting of the two parts of the first model.
In the discrimination model, the loss function of the discrimination model can accurately identify the discrimination of the first MIP image to the secondary image; the value of the loss function output of the discriminant model is a probability value between 0 and 1. In this embodiment, a preset threshold is set, and when the output probability value is higher than the preset threshold, the determination result is 1; and meeting the termination condition to obtain the image enhancement model. When the output probability value is lower than a preset threshold value, the judgment result is 0; if the termination condition is not met, feeding back to the generated model to generate a new secondary image, and continuing learning and training.
In one embodiment, the epoch number is set to achieve a gradient descent. The 1 epoch is equal to one training using all samples in the training set, that is, the value of the epoch is that the first MIP image and the second MIP image are read several times and input into the generative model and the discriminant model. Monitoring the output of the whole iteration process, and inputting the second MIP image into the generation model to obtain a secondary image; and monitors the output of the two loss functions until the termination condition is met.
In one embodiment, the method further comprises the following steps:
after the antagonistic neural network is built and generated, one-to-one shuffle processing is carried out on the first MIP image and the second MIP image, so that the first MIP image and the second MIP image are distributed and ordered, and the training of an image enhancement model is guaranteed.
As shown in fig. 4, in an embodiment, on the basis of the embodiment in fig. 2, the method further includes the following steps:
and S400, after the MIP enhanced image is output, storing the MIP enhanced image as a reference map for assisting positioning. The reference image can be a Tag storing the MIP enhanced image into DICOM pixel data and writing into a database and a disk. Therefore, an enhanced MIP enhanced image is provided, the format of the MIP enhanced image is a DICOM image, and the MIP enhanced image can be directly used as a reference image for assisting positioning.
It should be understood that although the various steps in the flow charts of fig. 2-4 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. 2-4 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 5, an apparatus for enhancing a low-dose MIP image includes an acquisition module 100, a creation module 200, and an enhancement module 300: an obtaining module 100, configured to obtain a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, reconstruct the first MIP image and the corresponding first PET image, and obtain a second MIP image under a low dose of radioactive tracer in a corresponding body position; the establishing module 200 is configured to learn the second MIP image through feedback of the first MIP image by using a generated countermeasure neural network, and establish an image enhancement model; the enhancing module 300 is configured to obtain a current MIP image under a current low-dose radioactive tracer, input the current MIP image into an image enhancing model for image enhancement, and output an MIP enhanced image.
The enhancement device of the low-dose MIP image can rapidly output a high-quality MIP enhanced image; thereby meeting the quality requirement of the MIP image with low dose as a reference picture.
In an embodiment, the obtaining module 100 is further configured to perform reconstruction on the first PET image under different doses according to preset reconstruction parameters to obtain a corresponding second PET image; and extracting the pixel value of the second PET image to obtain a second MIP image.
In one embodiment, the establishing module 200 is further configured to build a generating antagonistic neural network, and the generating antagonistic neural network includes a generating module and a judging module; taking the first MIP image as label data, and performing supervised learning on the second MIP image by using a generating module to obtain a secondary image; judging the sub-image and the first MIP image by using a judging module to obtain a judging result; and finishing training after the judgment result meets the termination condition to obtain the image enhancement model.
In an embodiment, the establishing module 200 is further configured to, after obtaining the determination result, if the determination result does not satisfy the termination condition, feed back the determination result to the generating module, and perform continuous supervised learning on the second MIP image by using the generating module to obtain a new secondary image.
As shown in fig. 6, in one embodiment, on the basis of the embodiment of fig. 5, a module 400 is further stored: the storage module 400 is configured to store the MIP enhanced image as a positioning-assisted reference map after outputting the MIP enhanced image.
Specific limitations regarding the enhancing apparatus for the low-dose MIP image can be referred to the above limitations regarding the enhancing method for the low-dose MIP image, and will not be described herein again. The modules in the low-dose MIP image enhancement 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, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing the model and data related to model training. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of enhancing a low dose MIP image.
Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
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:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer corresponding to a body position; learning the second MIP image through the feedback of the first MIP image by utilizing the generated countermeasure neural network, and establishing an image enhancement model; and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into an image enhancement model for image enhancement, and outputting an MIP enhanced image.
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:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer corresponding to a body position; learning the second MIP image through the feedback of the first MIP image by utilizing the generated countermeasure neural network, and establishing an image enhancement model; and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into an image enhancement model for image enhancement, and outputting an MIP enhanced image.
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 may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
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 (9)
1. A method of enhancing a low dose MIP image, the method comprising:
acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, and reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low dose of radioactive tracer of a corresponding body position;
the reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low-dose radiotracer corresponding to the body position includes:
reconstructing the first PET image under different doses according to preset reconstruction parameters, and obtaining a corresponding second PET image by using a data cutting function; extracting pixel values of the second PET image to obtain a second MIP image;
learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
and acquiring a current MIP image under a low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting an MIP enhanced image.
2. The method of claim 1, wherein learning the second MIP image through the feedback of the first MIP image using the generative antagonistic neural network establishes an image enhancement model comprising:
building a generated antagonistic neural network, wherein the generated antagonistic neural network comprises a generating module and a judging module;
taking the first MIP image as label data, and performing supervised learning on a second MIP image by using the generation module to obtain a secondary image;
distinguishing the secondary image and the first MIP image by using the distinguishing module to obtain a distinguishing result;
and finishing training after the judgment result meets the termination condition to obtain an image enhancement model.
3. The method of claim 2, further comprising:
and after the antagonistic neural network is built and generated, performing one-to-one shuffle processing on the first MIP image and the second MIP image.
4. The method of claim 2, further comprising:
and after the judgment result is obtained, if the judgment result does not meet the termination condition, feeding the judgment result back to the generation module, and performing continuous supervised learning on the second MIP image by using the generation module to obtain a new auxiliary image.
5. The method of any one of claims 1 to 4, further comprising:
after the MIP enhanced image is output, the MIP enhanced image is stored as a positioning-assisted reference map.
6. An apparatus for enhancing low-dose MIP images, the apparatus comprising an acquisition module, a building module and an enhancing module:
the acquisition module is used for acquiring a first MIP image and a corresponding first PET image under a full dose of radioactive tracer, reconstructing the first MIP image and the corresponding first PET image, and obtaining a second MIP image under a low dose of radioactive tracer corresponding to a body position;
the reconstructing the first MIP image and the corresponding first PET image to obtain a second MIP image under a low-dose radiotracer corresponding to the body position includes:
reconstructing the first PET image under different doses according to preset reconstruction parameters, and obtaining a corresponding second PET image by using a data cutting function; extracting pixel values of the second PET image to obtain a second MIP image;
the establishing module is used for learning the second MIP image through the feedback of the first MIP image by utilizing a generated countermeasure neural network, and establishing an image enhancement model;
the enhancement module is used for acquiring a current MIP image under the current low-dose radioactive tracer, inputting the current MIP image into the image enhancement model for image enhancement, and outputting the MIP enhanced image.
7. The apparatus of claim 6, further comprising a storage module to:
and the storage module is used for storing the MIP enhanced image as a reference map for assisting positioning after the MIP enhanced image is output.
8. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 5 when executing the computer program.
9. 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 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910504782.5A CN110264435B (en) | 2019-06-12 | 2019-06-12 | Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910504782.5A CN110264435B (en) | 2019-06-12 | 2019-06-12 | Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110264435A CN110264435A (en) | 2019-09-20 |
| CN110264435B true CN110264435B (en) | 2022-05-17 |
Family
ID=67917725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910504782.5A Active CN110264435B (en) | 2019-06-12 | 2019-06-12 | Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110264435B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991636B (en) * | 2019-11-14 | 2023-07-04 | 东软医疗系统股份有限公司 | Training method and device of generated countermeasure network, image enhancement method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108537794A (en) * | 2018-04-19 | 2018-09-14 | 上海联影医疗科技有限公司 | Medical image processing method, device and computer readable storage medium |
| CN109035356A (en) * | 2018-07-05 | 2018-12-18 | 四川大学 | A kind of system and method based on PET pattern imaging |
| CN109493951A (en) * | 2017-11-08 | 2019-03-19 | 上海联影医疗科技有限公司 | For reducing the system and method for dose of radiation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2890300B1 (en) * | 2012-08-31 | 2019-01-02 | Kenji Suzuki | Supervised machine learning technique for reduction of radiation dose in computed tomography imaging |
-
2019
- 2019-06-12 CN CN201910504782.5A patent/CN110264435B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109493951A (en) * | 2017-11-08 | 2019-03-19 | 上海联影医疗科技有限公司 | For reducing the system and method for dose of radiation |
| CN108537794A (en) * | 2018-04-19 | 2018-09-14 | 上海联影医疗科技有限公司 | Medical image processing method, device and computer readable storage medium |
| CN109035356A (en) * | 2018-07-05 | 2018-12-18 | 四川大学 | A kind of system and method based on PET pattern imaging |
Non-Patent Citations (1)
| Title |
|---|
| Low-Dose CT Image Denoising Based on Improved WGAN-gp;Xiaoli Li et al.;《Journal of New Media》;20190131;第2-3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110264435A (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993726B (en) | Medical image detection method, device, equipment and storage medium | |
| CN109754447B (en) | Image generation method, device, equipment and storage medium | |
| CN109166130B (en) | Image processing method and image processing device | |
| CN109598745B (en) | Image registration method and device and computer equipment | |
| CN111161270B (en) | Vascular segmentation method for medical image, computer device and readable storage medium | |
| CN110334722B (en) | Image classification method and device, computer equipment and storage medium | |
| CN109697740B (en) | Image reconstruction method and device and computer equipment | |
| CN111161269B (en) | Image segmentation method, computer device, and readable storage medium | |
| WO2019024568A1 (en) | Ocular fundus image processing method and apparatus, computer device, and storage medium | |
| CN110084868B (en) | Image correction method, apparatus, computer device, and readable storage medium | |
| WO2020252911A1 (en) | Facial recognition method for missing individual, apparatus, computer device and storage medium | |
| CN109616169B (en) | Similar patient mining method, similar patient mining device, computer equipment and storage medium | |
| CN110210519B (en) | Classification method, computer device, and storage medium | |
| CN110796652B (en) | Image processing method, computer device, and storage medium | |
| CN111210465A (en) | Image registration method and device, computer equipment and readable storage medium | |
| CN108830782B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN111583184A (en) | Image analysis method, network, computer device, and storage medium | |
| CN110008835B (en) | Sight line prediction method, device, system and readable storage medium | |
| CN110310257B (en) | Medical image processing method, apparatus, computer device and storage medium | |
| CN111026757B (en) | Method, device, equipment and storage medium for generating standard statistical format data | |
| CN110264435B (en) | Method and device for enhancing low-dose MIP (MIP) image, computer equipment and storage medium | |
| CN111179157A (en) | Method and device for processing face region in medical image and related product | |
| EP3709309A1 (en) | Medical data collection for machine learning | |
| CN110598025A (en) | Film reading method, system, device, equipment and storage medium | |
| KR20230093848A (en) | Low-dose ct reconstructing method based on machine learning and image processing apparatus |
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 | ||
| CB02 | Change of applicant information |
Address after: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |