CN112329920A - Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model - Google Patents
Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model Download PDFInfo
- Publication number
- CN112329920A CN112329920A CN202011230654.5A CN202011230654A CN112329920A CN 112329920 A CN112329920 A CN 112329920A CN 202011230654 A CN202011230654 A CN 202011230654A CN 112329920 A CN112329920 A CN 112329920A
- Authority
- CN
- China
- Prior art keywords
- data
- parameter
- weighted image
- magnetic resonance
- imaging model
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 70
- 238000012549 training Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 230000009466 transformation Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000006870 function Effects 0.000 description 26
- 238000013135 deep learning Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 210000000845 cartilage Anatomy 0.000 description 2
- 210000003127 knee Anatomy 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004071 biological effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- 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
- G06T11/006—Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
-
- 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
- G06T11/008—Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses an unsupervised training method and an unsupervised training device for a magnetic resonance parameter imaging model. The unsupervised training method comprises the following steps: inputting the acquired undersampling k-space data into a magnetic resonance parameter imaging model to be trained to obtain parameter weighted image data; updating a loss function according to the acquired undersampled k-space data and the parameter weighted image data; and updating the network parameters of the magnetic resonance parameter imaging model to be trained according to the updated loss function. By introducing the loss function based on the parameter weighted image, the calculation of the loss function can be completed by using the undersampled k-space data and the parameter weighted image data obtained according to the undersampled k-space data, and the unsupervised training of the model is realized without adopting the fully-sampled k-space data.
Description
Technical Field
The invention belongs to the technical field of image reconstruction of magnetic resonance parameter imaging signals, and particularly relates to an unsupervised training method and a training device of a magnetic resonance parameter imaging model, a computer readable storage medium and computer equipment.
Background
Quantitative Magnetic Resonance Parametric imaging (Quantitative Magnetic Resonance Parametric Mapping) is an emerging tool for assessing and determining basic biological properties of tissues. It aims to measure the absolute relaxation of magnetic resonance, providing comparable measurements across sites and time points. The most common method of acquiring magnetic resonance parameter maps is to acquire weighted images with varying imaging parameters (e.g., inversion Time (TI) in the T1 map, echo Time (TE) in the T2 map, or spin-lock Time (TSL) in the T1 ρ map.) then estimate the parameter maps by fitting these images pixel-by-pixel to corresponding physical exponential models.
In terms of fast imaging, the currently common techniques are parallel imaging and compressed sensing. Parallel imaging uses the correlation between multi-channel coils to accelerate acquisition, and compressed sensing uses the prior information of sparsity of an imaged object to reduce k-space sampling points. However, due to the limitation of hardware and other conditions, the parallel imaging acceleration times are limited, and the compressed sensing technology has very long reconstruction time and is difficult to select sparse transformation and reconstruction parameters due to the adoption of iterative reconstruction. In recent years, magnetic resonance image reconstruction using a deep learning method has received increasing attention. The deep learning method utilizes a neural network to learn the optimal parameters required by reconstruction from a large amount of training data or directly learn the mapping relation from undersampled data to fully-acquired images, thereby obtaining better imaging quality and higher acceleration times than the traditional parallel imaging or compressed sensing method.
Although the method of deep learning makes up the defects of the traditional rapid imaging method, some problems exist at the same time, for example, the data-driven deep learning lacks theoretical guidance, and a large amount of training data is often needed to obtain a good effect. In the existing deep learning magnetic resonance parameter imaging method, a fully acquired parameter map is required to be used as a reference map, and fully acquired data in actual scanning is difficult to obtain, so that network training is not facilitated.
Disclosure of Invention
(I) technical problems to be solved by the invention
The technical problem solved by the invention is as follows: aiming at a magnetic resonance parameter imaging model based on deep learning, how to carry out unsupervised training on the magnetic resonance parameter imaging model under the condition of not fully acquiring image data.
(II) the technical scheme adopted by the invention
An unsupervised training method of a magnetic resonance parametric imaging model, the unsupervised training method comprising:
inputting the acquired undersampling k-space data into a magnetic resonance parameter imaging model to be trained to obtain parameter weighted image data;
updating a loss function according to the acquired undersampled k-space data and the parameter weighted image data;
and updating the network parameters of the magnetic resonance parameter imaging model to be trained according to the updated loss function.
Preferably, the formula for calculating the loss function is:
wherein L (m; theta) represents a loss function, f represents under-acquired k-space data, m represents parameter weighted image data, A represents an encoding matrix, theta represents a network parameter, and psi represents sparse transformation.
Preferably, the imaging model to be trained includes a reconstruction network and a parameter fitting network, and the method for inputting the acquired undersampled k-space data into the magnetic resonance parametric imaging model to be trained to obtain the parameter weighted image data includes:
inputting the acquired undersampled k-space data, the first parameter weighted image initial data and the second parameter weighted image initial data into the reconstruction network to obtain first parameter weighted image updating data;
and inputting the first parameter weighted image updating data into the parameter fitting network to obtain second parameter weighted image updating data, and taking the second parameter weighted image updating data as the parameter weighted image data.
Preferably, the parameter-fitting network comprises a fitting sub-network and a signal relaxation physics model, wherein the method of inputting the first parameter-weighted image update data into the parameter-fitting network to obtain a second parameter-weighted image update data comprises:
inputting the first parametric weighted image update data into the fitting sub-network to obtain parametric image data and reference image data;
inputting the parametric image data and the reference image data to the signal relaxation physics model to obtain second parametric weighted image update data.
Preferably, after obtaining the second parameter-weighted image updating data, the training method further comprises:
respectively taking the first parameter weighted image updating data and the second parameter weighted image updating data obtained by the iteration as first parameter weighted image initial data and second parameter weighted image initial data of the next iteration, inputting the first parameter weighted image updating data and newly acquired under-acquired k-space data into the reconstruction network to obtain first parameter weighted image data of the next iteration, and inputting the first parameter weighted image updating data of the next iteration into the parameter fitting network to obtain second parameter weighted image updating data of the next iteration;
and repeating the steps according to the preset iteration number to obtain the second parameter weighted image updating data of the preset iteration number.
Preferably, the method for updating the loss function according to the acquired undersampled k-space data and the parametric weighted image data comprises: and calculating a loss function according to the acquired undersampled k-space data and the second parameter weighted image updating data of the preset iteration number.
The application also discloses an unsupervised training device of the magnetic resonance parameter imaging model, the unsupervised training device includes:
the data input module is used for inputting the acquired under-acquired k space data into an imaging model to be trained so as to obtain parameter weighted image data;
a loss function calculation module for updating a loss function according to the acquired under-acquired k-space data and the parameter weighted image data;
and the network parameter updating module is used for updating the network parameters of the imaging model to be trained according to the loss function.
The invention also discloses a computer readable storage medium, which stores a training program of the magnetic resonance parameter imaging model, and the training program of the magnetic resonance parameter imaging model is executed by a processor to realize an unsupervised training method of any one of the magnetic resonance parameter imaging models.
The invention also discloses a computer device, which comprises a computer readable storage medium, a processor and a training program of the magnetic resonance parameter imaging model stored in the computer readable storage medium, wherein the reconstruction program of the simultaneous multi-layer imaging signals realizes an unsupervised training method of any one of the magnetic resonance parameter imaging models when being executed by the processor.
(III) advantageous effects
The invention discloses an unsupervised training method of a magnetic resonance parameter imaging model, which has the following technical effects compared with the traditional training method:
the unsupervised training method can complete the calculation of the loss function by introducing the loss function based on the parameter weighted image, utilizing the undersampled k-space data and the parameter weighted image data obtained according to the undersampled k-space data, does not need to adopt the fully-acquired k-space data, and realizes the unsupervised training of the model.
Drawings
FIG. 1 is a flow chart of a method of unsupervised training of a magnetic resonance parametric imaging model according to an embodiment of the present invention;
FIG. 2 is a flow chart of the calculation of parametric weighted image data according to an embodiment of the present invention;
FIG. 3 is a training architecture diagram of a magnetic resonance parametric imaging model of an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a reconstruction network according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a fitting sub-network according to an embodiment of the present invention;
FIG. 6 is a graph showing the results of comparative experiments in accordance with examples of the present invention;
FIG. 7 is a schematic diagram of a training apparatus for a magnetic resonance parametric imaging model in accordance with an embodiment of the present invention;
FIG. 8 is a functional block diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in 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 invention and are not intended to limit the invention.
Before describing in detail the various embodiments of the present application, the inventive concepts of the present application are first briefly described: the magnetic resonance parametric imaging model based on deep learning often needs to adopt fully-acquired k-space data during training, and the fully-acquired k-space data is often difficult to acquire during the training process. According to the method and the device, the loss function based on the parameter weighted image is adopted, the under-acquired k-space data and the parameter weighted image data obtained according to the under-acquired k-space data are utilized to complete calculation of the loss function, and therefore unsupervised training of the model is achieved.
Specifically, as shown in fig. 1, the unsupervised training method for the magnetic resonance parameter imaging model includes the following steps:
step S10: and inputting the acquired undersampled k-space data into a magnetic resonance parameter imaging model to be trained to obtain parameter weighted image data.
In this embodiment, the undersampled k-space data may be acquired through various embodiments, for example, manually undersampled from the fully-acquired k-space data, or directly acquiring the undersampled k-space data by using a device, which is not particularly limited herein.
Step S20: and updating a loss function according to the acquired undersampled k-space data and the parameter weighted image data.
The loss function in the present embodiment is a loss function based on a parameter-weighted image. Illustratively, the loss function preferably takes the form:
the method comprises the steps that L (m; theta) represents a loss function, f represents under-acquired k-space data, m represents parameter weighted image data, A represents an encoding matrix, theta represents network parameters, and psi represents sparse transformation.
Step S30: and updating the network parameters of the magnetic resonance parameter imaging model to be trained according to the updated loss function.
As an embodiment, as shown in fig. 2 and fig. 3, the magnetic resonance parametric imaging model to be trained includes a reconstruction network (Recon-net) and a parameter fitting network (fitting), where the parameter fitting network includes a fitting sub-network (fitting-net) and a signal relaxation physical model (MR signal model), and the method for inputting the acquired undersampled k-space data into the imaging model to be trained to obtain the parametric weighted image data includes:
step S101: and inputting the acquired undersampled k-space data, the first parameter weighted image initial data and the second parameter weighted image initial data into the reconstruction network to obtain first parameter weighted image updating data.
Step S102: and inputting the first parameter weighted image updating data into the parameter fitting network to obtain second parameter weighted image updating data, and taking the second parameter weighted image updating data as the parameter weighted image data.
Wherein, step S102 includes the following steps: inputting the first parametric weighted image update data into the fitting sub-network to obtain parametric image data and reference image data; inputting the parametric image data and the reference image data to the signal relaxation physics model to obtain second parametric weighted image update data.
Further, after obtaining the second parameter-weighted image updating data, the training method further includes:
respectively taking the first parameter weighted image updating data and the second parameter weighted image updating data obtained by the iteration as first parameter weighted image initial data and second parameter weighted image initial data of the next iteration, inputting the first parameter weighted image updating data and undersampling k-space data into the reconstruction network to obtain first parameter weighted image data of the next iteration, and inputting the first parameter weighted image updating data of the next iteration into the parameter fitting network to obtain second parameter weighted image updating data of the next iteration;
and repeating the steps according to the preset iteration number to obtain the second parameter weighted image updating data of the preset iteration number.
Specifically, the above iterative process may be represented in the following form:
wherein n is iteration frequency, M is first parameter weighted image updating data, f is under-acquired k space data, pi represents reconstruction network, and M0,TxRespectively reference image data and parametric image data generated by the fitting sub-network U,
representing reference image data and parameter mapThe image data is updated via a second parametrically weighted image generated by a signal relaxation physics model S. The first parameter weighted image initial data and the second parameter weighted image initial data adopted in the first iteration are set according to industry experience.
Further, a network structure of the reconstruction network is shown in fig. 4, the reconstruction network adopts an improved original dual algorithm networking method, a traditional original dual algorithm is expanded on the network, and a parameter weighted image can be directly reconstructed from undersampled k-space data through a combination relation between an approximation operator and parameters in a neural network learning algorithm. The specific reconstruction process can be expressed in the form:
wherein d isn+1For the dual parameters of the (n + 1) th iteration, Γ and Λ are respectively two sub-networks, the network structure is shown in fig. 4, the number on each network layer represents the number of channels of the layer, and a residual error network is adopted for better training the network.
Further, the fitting sub-network adopts an optimized end-to-end Unet network, and the specific network structure is shown in FIG. 5. To reduce the number of training parameters, the fitting sub-network employs parameter sharing.
As an embodiment, in step S20, a loss function is calculated based on the acquired undersampled k-space data, second parameter weighted image update data for a predetermined number of iterations. Wherein the predetermined number of iterations is preferably 5.
Further, since the parameter fitting does not involve an iterative optimization algorithm, the parameter fitting network is intended to adopt an optimized end-to-end Unet network, and a specific network structure is shown in fig. 5. The down-sampling is implemented by convolution with step size 2, the left features are directly added to the right features instead of stacking, thereby reducing the number of training parameters, and bilinear quadratic interpolation is used as an up-sampling layer to reduce the checkerboard artifacts. Since the structural image is complex and the parametric map is always non-negative, the network layer that finally generates the parametric map needs to be processed using an activation function. In order to reduce the number of training parameters, the parameter graph generated by the network generates a parameter weighted image through a signal relaxation model to carry out the next iteration, and the fitting sub-network Mapping-net adopts parameter sharing.
The unsupervised training method of the embodiment can complete the calculation of the loss function by introducing the loss function based on the parameter weighted image, and by using the undersampled k-space data and the parameter weighted image data obtained according to the undersampled k-space data, and the unsupervised training of the model is realized without adopting the fully-sampled k-space data.
In order to verify that the magnetic resonance parameter imaging model obtained by the unsupervised training method has beneficial effects, the applicant performs a comparison experiment with T1ρFor imaging, a 5.2-fold retrospective undersampling experiment is performed on a 3T magnetic resonance imaging system to prove the effectiveness of the invention. As shown in the figure, ref represents a full acquisition image, DEMO represents an image obtained by using the magnetic resonance parameter imaging model of the embodiment, k-t SLR represents an image obtained by using a low rank method, and rec _ PF represents an image obtained by using a conventional compressed sensing method. From the experimental results, T of the region of interest (knee cartilage)1ρThe value is superimposed on the reconstructed image of 5ms TSL, in the table knee cartilage T1ρThe mean and standard deviation of the values, std, can be seen to be the closest to the fully acquired image (ref) using the magnetic resonance parametric imaging model of this embodiment.
Further, another embodiment discloses an unsupervised training apparatus for a magnetic resonance parametric imaging model, as shown in fig. 7, the unsupervised training apparatus comprising:
the data input module 100 is configured to input the acquired under-acquired k-space data to an imaging model to be trained to obtain parameter weighted image data;
a loss function calculating module 200, configured to update a loss function according to the acquired undersampled k-space data and the parameter weighted image data;
and a network parameter updating module 300, configured to update the network parameters of the imaging model to be trained according to the loss function.
Wherein the data input module 100 is further configured to:
inputting the acquired undersampled k-space data, the first parameter weighted image initial data and the second parameter weighted image initial data into the reconstruction network to obtain first parameter weighted image updating data; and
and inputting the first parameter weighted image updating data into the parameter fitting network to obtain second parameter weighted image updating data, and taking the second parameter weighted image updating data as the parameter weighted image data.
Further, the data input module 100 is further configured to: inputting the first parameter weighted image updating data as first parameter weighted image initial data of a next iteration, the second parameter weighted image updating data as second parameter weighted image initial data of the next iteration and undersampled k-space data into the reconstruction network to obtain first parameter weighted image updating data of the next iteration, and inputting the first parameter weighted image updating data of the next iteration into the parameter fitting network to obtain second parameter weighted image updating data of the next iteration.
Another embodiment discloses a computer readable storage medium storing an unsupervised training program of a magnetic resonance parametric imaging model, which when executed by a processor implements the above-mentioned unsupervised training method of the magnetic resonance parametric imaging model.
Another embodiment also discloses a computer device, which comprises, at a hardware level, as shown in fig. 8, a processor 12, an internal bus 13, a network interface 14, and a computer-readable storage medium 11. The processor 12 reads a corresponding computer program from the computer-readable storage medium and then runs, forming a request processing apparatus on a logical level. Of course, besides software implementation, the one or more embodiments in this specification do not exclude other implementations, such as logic devices or combinations of software and hardware, and so on, that is, the execution subject of the following processing flow is not limited to each logic unit, and may also be hardware or logic devices. The computer readable storage medium 11 stores thereon an unsupervised training program of the magnetic resonance parameter imaging model, which when executed by the processor implements the above-mentioned unsupervised training method of the magnetic resonance parameter imaging model.
Computer-readable storage media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer-readable storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
Although a few embodiments of the present invention have been shown and described, it would be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the claims and their equivalents, and that such changes and modifications are intended to be within the scope of the invention.
Claims (9)
1. An unsupervised training method for a magnetic resonance parametric imaging model, the unsupervised training method comprising:
inputting the acquired undersampling k-space data into a magnetic resonance parameter imaging model to be trained to obtain parameter weighted image data;
updating a loss function according to the acquired undersampled k-space data and the parameter weighted image data;
and updating the network parameters of the magnetic resonance parameter imaging model to be trained according to the updated loss function.
2. The method of unsupervised training of a magnetic resonance parametric imaging model of claim 1, wherein the loss function is calculated by the formula:
wherein L (m; theta) represents a loss function, f represents under-acquired k-space data, m represents parameter weighted image data, A represents an encoding matrix, theta represents a network parameter, and psi represents sparse transformation.
3. The method of unsupervised training of an mri model according to claim 1, wherein the to-be-trained imaging model comprises a reconstruction network and a parameter fitting network, and the method of inputting the acquired undersampled k-space data into the to-be-trained mri model to obtain the parameter weighted image data comprises:
inputting the acquired undersampled k-space data, the first parameter weighted image initial data and the second parameter weighted image initial data into the reconstruction network to obtain first parameter weighted image updating data;
and inputting the first parameter weighted image updating data into the parameter fitting network to obtain second parameter weighted image updating data, and taking the second parameter weighted image updating data as the parameter weighted image data.
4. A method of unsupervised training of a magnetic resonance parametric imaging model according to claim 3, characterized in that the parameter fitting network comprises a fitting sub-network and a signal relaxation physics model, wherein the method of inputting the first parameter weighted image update data into the parameter fitting network to obtain second parameter weighted image update data comprises:
inputting the first parametric weighted image update data into the fitting sub-network to obtain parametric image data and reference image data;
inputting the parametric image data and the reference image data to the signal relaxation physics model to obtain second parametric weighted image update data.
5. A method of unsupervised training of a magnetic resonance parametric imaging model according to claim 1, wherein after obtaining the second parametric weighted image update data, the method of training further comprises:
respectively taking the first parameter weighted image updating data and the second parameter weighted image updating data obtained by the iteration as first parameter weighted image initial data and second parameter weighted image initial data of the next iteration, inputting the first parameter weighted image updating data and undersampling k-space data into the reconstruction network to obtain first parameter weighted image data of the next iteration, and inputting the first parameter weighted image updating data of the next iteration into the parameter fitting network to obtain second parameter weighted image updating data of the next iteration;
and repeating the steps according to the preset iteration number to obtain the second parameter weighted image updating data of the preset iteration number.
6. The method of unsupervised training of a magnetic resonance parametric imaging model of claim 5, wherein the method of updating the loss function based on the acquired undersampled k-space data and the parametric weighted image data is by: and calculating a loss function according to the acquired undersampled k-space data and the second parameter weighted image updating data of the preset iteration number.
7. An unsupervised training device for a magnetic resonance parametric imaging model, the unsupervised training device comprising:
the data input module is used for inputting the acquired under-acquired k space data into an imaging model to be trained so as to obtain parameter weighted image data;
a loss function calculation module for updating a loss function according to the acquired under-acquired k-space data and the parameter weighted image data;
and the network parameter updating module is used for updating the network parameters of the imaging model to be trained according to the loss function.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a training program of a magnetic resonance parameter imaging model, which when executed by a processor implements the unsupervised training method of a magnetic resonance parameter imaging model of any one of claims 1 to 6.
9. A computer device, characterized in that the computer device comprises a computer readable storage medium, a processor and a training program of a magnetic resonance parametric imaging model stored in the computer readable storage medium, the unsupervised training program of the magnetic resonance parametric imaging model, when executed by the processor, implementing the unsupervised training method of the magnetic resonance parametric imaging model of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011230654.5A CN112329920A (en) | 2020-11-06 | 2020-11-06 | Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011230654.5A CN112329920A (en) | 2020-11-06 | 2020-11-06 | Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112329920A true CN112329920A (en) | 2021-02-05 |
Family
ID=74317037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011230654.5A Pending CN112329920A (en) | 2020-11-06 | 2020-11-06 | Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112329920A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538611A (en) * | 2021-06-21 | 2021-10-22 | 复旦大学 | Unsupervised cardiac magnetic resonance parameter quantitative image reconstruction method |
| CN114967121A (en) * | 2022-05-13 | 2022-08-30 | 哈尔滨工业大学 | End-to-end single lens imaging system design method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103443643A (en) * | 2010-09-01 | 2013-12-11 | 原子能与替代能源委员会 | Method for performing parallel magnetic resonance imaging |
| WO2017113205A1 (en) * | 2015-12-30 | 2017-07-06 | 中国科学院深圳先进技术研究院 | Rapid magnetic resonance imaging method and apparatus based on deep convolutional neural network |
| CN106970343A (en) * | 2017-04-11 | 2017-07-21 | 深圳先进技术研究院 | A kind of MR imaging method and device |
| CN110664378A (en) * | 2019-10-28 | 2020-01-10 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
| US20200105031A1 (en) * | 2018-09-30 | 2020-04-02 | The Board Of Trustees Of The Leland Stanford Junior University | Method for Performing Magnetic Resonance Imaging Reconstruction with Unsupervised Deep Learning |
| US20200341094A1 (en) * | 2019-04-25 | 2020-10-29 | Siemens Healthcare Gmbh | Multi-contrast mri image reconstruction using machine learning |
| CN111856362A (en) * | 2019-04-24 | 2020-10-30 | 深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
-
2020
- 2020-11-06 CN CN202011230654.5A patent/CN112329920A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103443643A (en) * | 2010-09-01 | 2013-12-11 | 原子能与替代能源委员会 | Method for performing parallel magnetic resonance imaging |
| WO2017113205A1 (en) * | 2015-12-30 | 2017-07-06 | 中国科学院深圳先进技术研究院 | Rapid magnetic resonance imaging method and apparatus based on deep convolutional neural network |
| CN106970343A (en) * | 2017-04-11 | 2017-07-21 | 深圳先进技术研究院 | A kind of MR imaging method and device |
| US20200105031A1 (en) * | 2018-09-30 | 2020-04-02 | The Board Of Trustees Of The Leland Stanford Junior University | Method for Performing Magnetic Resonance Imaging Reconstruction with Unsupervised Deep Learning |
| CN111856362A (en) * | 2019-04-24 | 2020-10-30 | 深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
| US20200341094A1 (en) * | 2019-04-25 | 2020-10-29 | Siemens Healthcare Gmbh | Multi-contrast mri image reconstruction using machine learning |
| CN110664378A (en) * | 2019-10-28 | 2020-01-10 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method, device, system and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| JING CHENG: "Learning reconstruction without ground-truth data: an unsupervised way for fast MR imaging", 《RESEARCHGATE》, 8 November 2019 (2019-11-08) * |
| 王一达;宋阳;谢海滨;童睿;李建奇;杨光;: "卷积神经网络重建欠采的磁共振图像", 磁共振成像, no. 06, 19 June 2018 (2018-06-19) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538611A (en) * | 2021-06-21 | 2021-10-22 | 复旦大学 | Unsupervised cardiac magnetic resonance parameter quantitative image reconstruction method |
| CN114967121A (en) * | 2022-05-13 | 2022-08-30 | 哈尔滨工业大学 | End-to-end single lens imaging system design method |
| CN114967121B (en) * | 2022-05-13 | 2023-02-03 | 哈尔滨工业大学 | Design method of end-to-end single lens imaging system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108896943B (en) | Magnetic resonance quantitative imaging method and device | |
| US10489943B2 (en) | System and method for sparse image reconstruction | |
| Cole et al. | Unsupervised MRI reconstruction with generative adversarial networks | |
| WO2020114329A1 (en) | Fast magnetic resonance parametric imaging and device | |
| US11346911B2 (en) | Magnetic resonance fingerprinting image reconstruction and tissue parameter estimation | |
| CN112329920A (en) | Unsupervised training method and unsupervised training device for magnetic resonance parameter imaging model | |
| CN110827369B (en) | Undersampling model generation method, image reconstruction method, apparatus and storage medium | |
| CN111028153A (en) | Image processing and neural network training method and device and computer equipment | |
| US11948676B2 (en) | Qualitative and quantitative MRI using deep learning | |
| CN113924503B (en) | Parameter map determination for time domain magnetic resonance | |
| CN113534031B (en) | Image domain data generating method, computer device, and readable storage medium | |
| CN113298902A (en) | Method for reconstructing magnetic resonance image in convoluted field of view, computer device and storage medium | |
| CN114010180B (en) | Magnetic resonance rapid imaging method and device based on convolutional neural network | |
| CN111127575A (en) | Image reconstruction method, computer-readable medium, and computer device | |
| CN112336337B (en) | Training method and device for magnetic resonance parameter imaging model, medium and equipment | |
| CN113030813B (en) | Magnetic resonance T2Quantitative imaging method and system | |
| CN111856364B (en) | Magnetic resonance imaging method, device and system and storage medium | |
| CN115356672B (en) | Multi-dimensional magnetic resonance imaging method, system and storage medium | |
| CN112150570B (en) | Compressed sensing magnetic resonance imaging method based on iterative p-threshold projection algorithm | |
| WO2021129235A1 (en) | Rapid three-dimensional magnetic resonance parameter imaging method and device | |
| JP2023069890A (en) | Magnetic resonance imaging device, image processing device, and image processing method | |
| CN114565690A (en) | Magnetic resonance imaging method based on wave gradient coding field and deep learning model | |
| CN103236049B (en) | Based on the partial K space image reconstruction method of sequence similarity interpolation | |
| CN113470128A (en) | Method for reconstructing simultaneous multi-slice imaging signals, storage medium and computer device | |
| CN111856365A (en) | Magnetic resonance imaging method, magnetic resonance imaging method and magnetic resonance imaging device |
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 |