EP4639191A1 - Method and system for magnetic resonance imaging processing - Google Patents

Method and system for magnetic resonance imaging processing

Info

Publication number
EP4639191A1
EP4639191A1 EP23820855.7A EP23820855A EP4639191A1 EP 4639191 A1 EP4639191 A1 EP 4639191A1 EP 23820855 A EP23820855 A EP 23820855A EP 4639191 A1 EP4639191 A1 EP 4639191A1
Authority
EP
European Patent Office
Prior art keywords
magnetic resonance
magnetic field
shimming
main magnetic
operator
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
Application number
EP23820855.7A
Other languages
German (de)
French (fr)
Inventor
Peter Forthmann
Christoph Günther Leussler
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koninklijke Philips NV
Original Assignee
Koninklijke Philips NV
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips NV filed Critical Koninklijke Philips NV
Publication of EP4639191A1 publication Critical patent/EP4639191A1/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/28—Details of apparatus provided for in groups G01R33/44 - G01R33/64
    • G01R33/38—Systems for generation, homogenisation or stabilisation of the main or gradient magnetic field
    • G01R33/387—Compensation of inhomogeneities
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/445—MR involving a non-standard magnetic field B0, e.g. of low magnitude as in the earth's magnetic field or in nanoTesla spectroscopy, comprising a polarizing magnetic field for pre-polarisation, B0 with a temporal variation of its magnitude or direction such as field cycling of B0 or rotation of the direction of B0, or spatially inhomogeneous B0 like in fringe-field MR or in stray-field imaging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/565—Correction of image distortions, e.g. due to magnetic field inhomogeneities
    • G01R33/56563—Correction of image distortions, e.g. due to magnetic field inhomogeneities caused by a distortion of the main magnetic field B0, e.g. temporal variation of the magnitude or spatial inhomogeneity of B0
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/28—Details of apparatus provided for in groups G01R33/44 - G01R33/64
    • G01R33/38—Systems for generation, homogenisation or stabilisation of the main or gradient magnetic field
    • G01R33/381—Systems for generation, homogenisation or stabilisation of the main or gradient magnetic field using electromagnets
    • G01R33/3815—Systems for generation, homogenisation or stabilisation of the main or gradient magnetic field using electromagnets with superconducting coils, e.g. power supply therefor
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/5608—Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels

Definitions

  • the invention relates to the field of magnetic resonance imaging. More particularly, the present invention relates to the field of processing a magnetic resonance image acquired with a magnetic resonance imaging device.
  • High-temperature superconductors enable magnets to be ramped very quickly, which in turn enables magnetic resonance imaging devices to be constructed employing a switchable magnetic flux density of the main magnetic field.
  • This allows intra-examination magnetic flux density switching of the main magnetic field to a further magnetic flux density of a further main magnetic field of the magnetic resonance imaging device.
  • imaging is carried out at a main magnetic field of 1.5 T.
  • Subsequent switching of the high-temperature superconductor would cause a change in magnetic flux density to a further main magnetic field of 0.6 T and enable acquisition of further magnetic resonance images at this lower magnetic flux density.
  • the shimming is still set so that the main magnetic field at, for example, 1.5 T is homogeneous.
  • the further main magnetic field shows inhomogeneities due to the change in magnetic flux density, because the shimming was not performed for the further magnetic flux density different from the first magnetic flux density.
  • Shimming in general is the process that provides the maximum possible homogenization of the main magnetic field.
  • Passive shimming is carried out by placing and positioning ferromagnetic materials on the inside of the magnet forming the main magnetic field and serves to minimize the inhomogeneities caused by the design of the magnetic resonance imaging device.
  • Active shimming can be carried out using shim coils provided for this purpose. Active shimming homogenizes the inhomogeneities in the main magnetic field caused by the patient load.
  • a magnetic resonance imaging system is known that is configured for imaging near a metal implant and in which the magnetic field can be ramped in a clinically acceptable time.
  • a method of processing an acquired magnetic resonance image comprising the following steps: providing a first magnetic resonance image which has been acquired with a magnetic resonance imaging device at a first main magnetic field with a first magnetic flux density at shimming conditions for shimming a second main magnetic field to uniformity, wherein the second main magnetic field has a second magnetic flux density which is different from the first magnetic flux density, providing a second magnetic resonance image which has been acquired with the magnetic resonance imaging device at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, providing a first operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity into a magnetic resonance image which would be expected at the second magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, providing a second operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device at a first main magnetic field with a first magnetic
  • An operator is generally a mapping or function that acts on elements of a space to produce elements of another space.
  • the term can be used as a synonym for functions or for the application of a machine learning algorithm to the elements of a space.
  • the actual elements in this space are the first magnetic resonance image and the second magnetic resonance image with the respective first and second operator.
  • shimming a main magnetic field to uniformity means that such a shimming is applied with which image distortions due to field inhomogeneities are reduced. However, while such shimming may help to greatly reduce field inhomogeneities, such shimming will never be able to achieve a completely uniform main magnetic field.
  • the second operator is used together with the first operator, the first magnetic resonance image and the second magnetic resonance image in order to reconstruct the magnetic resonance image which would have been expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at correct shimming conditions for shimming the first main magnetic field to uniformity. Because the ground truth formed by the second magnetic resonance image which has been acquired with the magnetic resonance imaging device at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity is provided, i.e. acquired, the operation of the first operator provides for access to consistency of the operation of the second operator, that can be accordingly corrected for or adapted. The operation of the first operator may further be employed as a constraint for the operation of the second operator to improve the consistency of the result of the action of the first operator.
  • the transformation of the first magnetic resonance image with the two operators provides a time efficient method to examine a patient intra examination with the magnetic resonance imaging device being switchable into two different modes, two magnetic flux densities, without the need to adjust the passive or active shimming between the two magnetic flux densities.
  • passive shimming which involves changing the main magnetic field by introducing ferromagnetic materials, is a resource-intensive shimming method to achieve the homogenization of the magnetic field.
  • the method for performing the last step comprises a trained machine learning algorithm representing the action of the first operator and the second operator which, upon input of the first magnetic resonance image and the second magnetic resonance image, generates the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity.
  • the intermediate result may not necessarily be available to a user of the system when the first operator is applied, but remains internal to the trained machine learning algorithm and only the final result is accessible.
  • these operators can be execution steps in the trained machine learning algorithm that cannot be accessed.
  • these operators may be the activation functions in the nodes of the respective levels of this trained artificial neural network.
  • the training data set further comprises the training data set for developing the first operator in the trained machine learning algorithm including the magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
  • Magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity are thereby fed in pairs building the training data set to the machine learning algorithm.
  • in situ and in vivo magnetic resonance images of humanoid subjects, animal subjects or special magnetic resonance imaging phantoms intended as substitute for in vivo magnetic resonance imaging can be acquired at the different magnetic flux densities.
  • the training data set can include the respective magnitude images as well as the phase images.
  • the method further comprises that the training dataset for developing the first operator in the trained machine learning algorithm is generated with a generative adversarial network formed from two artificial neural networks.
  • the magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity can be generated by employing the generative adversarial network formed from two artificial neural networks to generate the training data for the first operator without the need for additional in vivo or in situ acquisition of magnetic resonance images.
  • the method further comprises that the training dataset for developing the first operator in the trained machine learning algorithm is generated with electromagnetic simulations.
  • Maxwell's equations for a given problem can be solved in a simulation environment to derive the first operator accordingly.
  • the method further comprises that the training data set for developing the second operator in the trained machine learning algorithm includes the magnetic resonance images of a magnetic resonance imaging phantom acquired with the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, and the magnetic resonance images of the magnetic resonance imaging phantom acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
  • the training data set contains the differences between the first magnetic resonance image and the second magnetic resonance image and by this a direct measure for the magnetic field inhomogeneities generated by the change of the magnetic flux density.
  • the machine learning algorithm can be trained to compensate for these inhomogeneities using the second operator. This effect is magnetic resonance imaging device-specific from the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from.
  • the trained machine learning algorithm for the second operator is a trained artificial neural network.
  • the method further comprises that in the last step the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is retrieved by minimizing a loss function which uses the first magnetic resonance image and the second magnetic resonance image and the first operator and the second operator.
  • the first magnetic resonance image converges to the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity. This is a process that can never be perfect, and therefore regularly proceeds in an iterative asymptotic approximation.
  • this loss function stops the calculation of the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, when reaching a certain iteration threshold which can be defined by a user.
  • the minimization of the loss function yields local minima instead of global minima. Basically, however, the artifacts in the magnetic resonance image that would have to be expected are reduced compared to the first magnetic resonance image.
  • the method further comprises that for the loss function LF the following formula applies:
  • LF Operator l(Operator2(Imagel)) - Image2, with Operator 1 being the first operator, Operator 2 being the second operator, Image 1 being the first image and Image 2 being the second image.
  • This loss function iteratively minimizes the distance between the second magnetic resonance image and the first magnetic resonance image on which the first operator and the second operator were applied.
  • the first operator and the second operator are entered as initial values, which can be changed during the iterative minimization using the loss function.
  • the method further comprises that the first operator has been obtained by including the magnetic resonance image properties of magnetic resonance images acquired at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and of magnetic resonance images acquired at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, wherein the magnetic resonance images acquired at the first main magnetic field and the second main magnetic field, respectively, have not been acquired with the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from.
  • the first operator is independent of the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from and it is not necessary to determine the operator for each specific magnetic resonance imaging device. That is, this first operator needs to be determined only once.
  • the method further comprises that the second operator is obtained by acquiring magnetic resonance images from the magnetic resonance imaging phantom simulating a patient load at the first main magnetic field and at the second main magnetic field, and the difference between the acquired magnetic resonance images of the magnetic resonance imaging phantom at the first main magnetic field and the second main magnetic field includes the effect of the inhomogeneities of the first main magnetic field with the first magnetic flux density at shimming conditions for shimming the second main magnetic field to uniformity.
  • a computer program for magnetic resonance imaging modification comprising instructions which, when the program is executed by a computer, cause the computer to perform a method as described above.
  • the invention further provides a system of a magnetic resonance imaging device and a computing unit, the magnetic resonance imaging device with a switch with which the main magnetic field is switchable, with a first main magnetic field comprising a first magnetic flux density with shimming conditions for shimming a second main magnetic field to uniformity, with the second main magnetic field comprising a second magnetic flux density with shimming conditions for shimming the second main magnetic field to uniformity, wherein the second magnetic flux density is different from the first magnetic flux density, wherein the computing unit is adapted to execute the method as described above.
  • a superconducting magnet is provided for generating the first main magnetic field and the second main magnetic field, and the superconducting magnet is a dry magnet.
  • the superconducting magnet used to generate the first main magnetic field and the second main magnetic field is a high-temperature superconductor.
  • High-temperature superconductors allow the superconducting magnet to switch rapidly between the first magnetic flux density and the second magnetic flux density.
  • these first operator and second operator can be trained on the basis of ground truth annotated training image pairs that are converted into another by the respective first operator and second operator or on the basis of a ground truth developed by computer- simulations.
  • These operators can be implemented as neural networks that are trained to return the desired image form the input image.
  • the second operator may be trained on a large collection of image pairs e.g., of phantoms at respective field rampings and correct associated shim settings or by Bloch simulations .
  • the first operator may be implemented as trained neural network by training of image at one field ramp and correct/incorrect shim settings of a phantom model. Alternatively, it may be sufficient only to gather data at the incorrect shim settings and from knowledge of the induced encoding distortions owing the ‘wrong’ shim settings the correspond image at the equal field ramp and the correct shim settings may be generated.
  • TH - transmission between contrasts at different field strengths
  • US- unscrambling
  • real data may be employed.
  • the more data is provided for training the better the results.
  • GAN generative adversarial network
  • the GAN itself needs to be trained, but once that is done e.g. on the basis of a limited initial training set, it can generate virtually infinitely many TH and US feature/label pairs that can be used to train the above TH - and US -networks.
  • the superconducting magnet used to generate the first main magnetic field and the second main magnetic field is a magnesium diboride magnet.
  • Fig. 1 schematically depicts a scheme of a method according to a preferred embodiment of the invention.
  • Fig. 2 schematically depicts a system of a magnetic resonance imaging device and a computing unit according to a preferred embodiment of the invention.
  • Fig. 1 schematically depicts a scheme of a method according to a preferred embodiment of the invention. This method comprises steps SI to S5.
  • SI First, a first magnetic resonance image which has been acquired with a magnetic resonance imaging device 1 at a first main magnetic field with a first magnetic flux density at shimming conditions for shimming a second main magnetic field to uniformity is provided, wherein the second main magnetic field has a second magnetic flux density which is different from the first magnetic flux density, S2: Second, a second magnetic resonance image which has been acquired with the magnetic resonance imaging device 1 at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity is provided,
  • S5 Fifth, a magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is retrieved by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image.
  • the first operator is generated by a trained machine learning algorithm that has been trained with magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity. These images have not been acquired with the magnetic resonance imaging device 1 where the first magnetic resonance image and the second magnetic resonance image originate from.
  • This first operator is only deduced once and is valid for the combination of the first magnetic flux density and the second magnetic flux density for the given set of scan parameters for which they have been determined. In case the scan parameters were changed, e.g. if repetition time (TR) or echo time (TE) would change and, therefore, a different image contrast would be achieved, a new determination would be necessary.
  • the second operator is also generated by a trained machine learning algorithm that has been trained with a training data including the magnetic resonance images of a magnetic resonance imaging phantom 5 for simulating a patient load acquired with the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, and the magnetic resonance images of the magnetic resonance imaging phantom 5 acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity at the respective set of given scan parameters.
  • the first and the second operator are then implemented into a loss function in order to retrieve the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by minimizing this loss function which, uses the first magnetic resonance image and the second magnetic resonance image and the first operator and the second operator.
  • Fig. 2 schematically depicts a system of the magnetic resonance imaging device 1 and a computing unit 2 according to a preferred embodiment of the invention.
  • the main magnetic field in the bore of the magnetic resonance imaging device 1 is generated by a superconducting magnet 6, which is a high temperature superconductor.
  • the magnetic flux density of the main magnetic field generated by the superconducting magnet 6 is switchable by a switch 7 between the first main magnetic field with the first magnetic flux density at shimming conditions for shimming the second main magnetic field to uniformity and the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity having the second magnetic flux density different from the first magnetic flux density.
  • the magnetic resonance imaging phantom 5 is arranged on a treatment couch 4 of a patient positioning system 3.
  • the switch 7 is connected to the computing unit 2 and can be controlled by it.
  • the computing unit 2 is adapted to perform the method of steps S1-S5 and thus obtain the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image, then calculates the magnetic resonance image according to the first operator onto the first magnetic resonance image and the second operator onto the second magnetic resonance image.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • High Energy & Nuclear Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
  • Image Processing (AREA)

Abstract

According to the invention a method of processing an acquired magnetic resonance image is provided, with the following steps: S1) providing a first magnetic resonance image, S2) providing a second magnetic resonance image, S3) providing a first operator, S4) providing a second operator for, S5) retrieving a magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image.

Description

METHOD AND SYSTEM FOR MAGNETIC RESONANCE IMAGING PROCESSING
FIELD OF THE INVENTION
The invention relates to the field of magnetic resonance imaging. More particularly, the present invention relates to the field of processing a magnetic resonance image acquired with a magnetic resonance imaging device.
BACKGROUND OF THE INVENTION
High-temperature superconductors enable magnets to be ramped very quickly, which in turn enables magnetic resonance imaging devices to be constructed employing a switchable magnetic flux density of the main magnetic field. This allows intra-examination magnetic flux density switching of the main magnetic field to a further magnetic flux density of a further main magnetic field of the magnetic resonance imaging device. For example, imaging is carried out at a main magnetic field of 1.5 T. Subsequent switching of the high-temperature superconductor would cause a change in magnetic flux density to a further main magnetic field of 0.6 T and enable acquisition of further magnetic resonance images at this lower magnetic flux density. However, the shimming is still set so that the main magnetic field at, for example, 1.5 T is homogeneous. The further main magnetic field shows inhomogeneities due to the change in magnetic flux density, because the shimming was not performed for the further magnetic flux density different from the first magnetic flux density.
Shimming in general is the process that provides the maximum possible homogenization of the main magnetic field. A distinction is made between active and passive shimming. Passive shimming is carried out by placing and positioning ferromagnetic materials on the inside of the magnet forming the main magnetic field and serves to minimize the inhomogeneities caused by the design of the magnetic resonance imaging device. Active shimming can be carried out using shim coils provided for this purpose. Active shimming homogenizes the inhomogeneities in the main magnetic field caused by the patient load.
The inhomogeneities caused by the change in the magnetic flux density of the main magnetic field cause image artifacts in the magnetic resonance imaging device that cannot be sufficiently reduced by active shimming.
The article ‘Feasibility study of novel rapid ramp-down procedure in MgB2 MRI magnet using persistent current switch with high off-resistivity’ by Kodama et. al. in Superconductor Science and Technology, 34 (2021) 074003 ( 13pp); describes dry magnets with high-temperature superconductors and MgB2, which are provided with a novel fast shutdown method that can replace controlled shutdown in emergencies. For this purpose, it is provided to establish a power supply by means of a persistent current switch, which is switched off when heated and the power supply is interrupted by means of a breaker. The energy stored in the solenoid is consumed at an external resistor.
Accurate homogenization of the main magnetic field is critical in magnetic resonance imaging to reduce inhomogeneity artifacts in the magnetic resonance image. However, with existing rampable magnetic resonance imaging scanners and the ability to change the main magnetic field, homogenization of the main magnetic field by shimming is not sufficient.
From the US-patent application US2020/0400764. a magnetic resonance imaging system is known that is configured for imaging near a metal implant and in which the magnetic field can be ramped in a clinically acceptable time.
SUMMARY OF THE INVENTION
It is an objective of the invention to provide a time and memory efficient method for reducing the effects of magnetic field inhomogeneities in magnetic resonance images.
According to the invention, this object is addressed by the subject matter of the independent claims. Preferred embodiments of the invention are described in the dependent claims.
Therefore, according to the invention, a method of processing an acquired magnetic resonance image is provided, the method comprising the following steps: providing a first magnetic resonance image which has been acquired with a magnetic resonance imaging device at a first main magnetic field with a first magnetic flux density at shimming conditions for shimming a second main magnetic field to uniformity, wherein the second main magnetic field has a second magnetic flux density which is different from the first magnetic flux density, providing a second magnetic resonance image which has been acquired with the magnetic resonance imaging device at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, providing a first operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity into a magnetic resonance image which would be expected at the second magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, providing a second operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity into a magnetic resonance image which would be expected at the first magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, retrieving a magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image. An operator is generally a mapping or function that acts on elements of a space to produce elements of another space. The term can be used as a synonym for functions or for the application of a machine learning algorithm to the elements of a space. The actual elements in this space are the first magnetic resonance image and the second magnetic resonance image with the respective first and second operator.
The term “shimming a main magnetic field to uniformity” means that such a shimming is applied with which image distortions due to field inhomogeneities are reduced. However, while such shimming may help to greatly reduce field inhomogeneities, such shimming will never be able to achieve a completely uniform main magnetic field.
Further, it appears that in practice it will not be possible to generate operators which work perfectly, i.e. which are able to exactly calculate the expected image from a given image. If this were the case, the second operator alone would be sufficient to reconstruct a correctly shimmed image from a badly shimmed image, i.e. an image which has been acquired at shimming conditions for another main magnetic field with higher or lower magnetic flux density. Therefore, according to the invention, the second operator is used together with the first operator, the first magnetic resonance image and the second magnetic resonance image in order to reconstruct the magnetic resonance image which would have been expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at correct shimming conditions for shimming the first main magnetic field to uniformity. Because the ground truth formed by the second magnetic resonance image which has been acquired with the magnetic resonance imaging device at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity is provided, i.e. acquired, the operation of the first operator provides for access to consistency of the operation of the second operator, that can be accordingly corrected for or adapted. The operation of the first operator may further be employed as a constraint for the operation of the second operator to improve the consistency of the result of the action of the first operator.
This allows a time efficient and resource saving approach to unlock the full functionality of a magnetic resonance imaging device comprising a switchable magnetic flux density. The transformation of the first magnetic resonance image with the two operators provides a time efficient method to examine a patient intra examination with the magnetic resonance imaging device being switchable into two different modes, two magnetic flux densities, without the need to adjust the passive or active shimming between the two magnetic flux densities. In particular, passive shimming, which involves changing the main magnetic field by introducing ferromagnetic materials, is a resource-intensive shimming method to achieve the homogenization of the magnetic field. The use of operators for the transformation of the magnetic resonance images enables new magnetic resonance imaging possibilities, for example the reduction of susceptibility artifacts or the like by magnetic resonance image acquisition at different magnetic flux densities. Furthermore, by forming the operators, a computationally and memory efficient approach is chosen in particular. Once the operators have been constructed, the training data for the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from are no longer required, thus achieving a synergistic effect with the first and the second operator, since the first operator does not necessarily have to be generated locally at the magnetic resonance imaging device in question.
According to the invention, various algorithms may be used for the last step in which the operators are applied. However, according to a preferred embodiment of the invention, the method for performing the last step comprises a trained machine learning algorithm representing the action of the first operator and the second operator which, upon input of the first magnetic resonance image and the second magnetic resonance image, generates the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity.
This means in particular that the intermediate result may not necessarily be available to a user of the system when the first operator is applied, but remains internal to the trained machine learning algorithm and only the final result is accessible. Thus, these operators can be execution steps in the trained machine learning algorithm that cannot be accessed. For example, in a trained artificial neural network, these operators may be the activation functions in the nodes of the respective levels of this trained artificial neural network.
In principle, it is possible for the training data set to include various data. According to a preferred embodiment of the invention, however, the method further comprises the training data set for developing the first operator in the trained machine learning algorithm including the magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
Magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity are thereby fed in pairs building the training data set to the machine learning algorithm. For this purpose, in situ and in vivo magnetic resonance images of humanoid subjects, animal subjects or special magnetic resonance imaging phantoms intended as substitute for in vivo magnetic resonance imaging can be acquired at the different magnetic flux densities. The training data set can include the respective magnitude images as well as the phase images.
In general, it is possible to generate the training data set for the first operator in various ways. According to a preferred embodiment of the invention, however, the method further comprises that the training dataset for developing the first operator in the trained machine learning algorithm is generated with a generative adversarial network formed from two artificial neural networks. The magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity can be generated by employing the generative adversarial network formed from two artificial neural networks to generate the training data for the first operator without the need for additional in vivo or in situ acquisition of magnetic resonance images.
According to a preferred embodiment of the invention the method further comprises that the training dataset for developing the first operator in the trained machine learning algorithm is generated with electromagnetic simulations.
With these electromagnetic simulations, for example, Maxwell's equations for a given problem can be solved in a simulation environment to derive the first operator accordingly.
It is possible to generate the training data set for the second operator in various ways. According to a preferred embodiment of the invention, however, the method further comprises that the training data set for developing the second operator in the trained machine learning algorithm includes the magnetic resonance images of a magnetic resonance imaging phantom acquired with the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, and the magnetic resonance images of the magnetic resonance imaging phantom acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
Thus, the training data set contains the differences between the first magnetic resonance image and the second magnetic resonance image and by this a direct measure for the magnetic field inhomogeneities generated by the change of the magnetic flux density. The machine learning algorithm can be trained to compensate for these inhomogeneities using the second operator. This effect is magnetic resonance imaging device-specific from the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from.
Preferably, the trained machine learning algorithm for the second operator is a trained artificial neural network.
In principle, different methods may be provided to approximate the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field under shimming conditions for shimming the first main magnetic field for uniformity. According to a preferred embodiment of the invention, however, the method further comprises that in the last step the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is retrieved by minimizing a loss function which uses the first magnetic resonance image and the second magnetic resonance image and the first operator and the second operator. With the minimization of the loss function, the first magnetic resonance image converges to the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity. This is a process that can never be perfect, and therefore regularly proceeds in an iterative asymptotic approximation. Therefore, it can be provided that this loss function stops the calculation of the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, when reaching a certain iteration threshold which can be defined by a user. Similarly, it is possible that the minimization of the loss function yields local minima instead of global minima. Basically, however, the artifacts in the magnetic resonance image that would have to be expected are reduced compared to the first magnetic resonance image.
It is possible to employ different implementations of minimizing a loss function. According to a preferred embodiment of the invention, however, the method further comprises that for the loss function LF the following formula applies:
LF = Operator l(Operator2(Imagel)) - Image2, with Operator 1 being the first operator, Operator 2 being the second operator, Image 1 being the first image and Image 2 being the second image.
This loss function iteratively minimizes the distance between the second magnetic resonance image and the first magnetic resonance image on which the first operator and the second operator were applied. In this minimization of the loss function, the first operator and the second operator are entered as initial values, which can be changed during the iterative minimization using the loss function.
In general, it is possible for different magnetic resonance imaging devices to be used to acquire the magnetic resonance images for the first operator. According to a preferred embodiment of the invention, however, the method further comprises that the first operator has been obtained by including the magnetic resonance image properties of magnetic resonance images acquired at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and of magnetic resonance images acquired at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, wherein the magnetic resonance images acquired at the first main magnetic field and the second main magnetic field, respectively, have not been acquired with the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from. Thus, the first operator is independent of the magnetic resonance imaging device where the first magnetic resonance image and the second magnetic resonance image originate from and it is not necessary to determine the operator for each specific magnetic resonance imaging device. That is, this first operator needs to be determined only once.
According to a preferred embodiment of the invention the method further comprises that the second operator is obtained by acquiring magnetic resonance images from the magnetic resonance imaging phantom simulating a patient load at the first main magnetic field and at the second main magnetic field, and the difference between the acquired magnetic resonance images of the magnetic resonance imaging phantom at the first main magnetic field and the second main magnetic field includes the effect of the inhomogeneities of the first main magnetic field with the first magnetic flux density at shimming conditions for shimming the second main magnetic field to uniformity.
Further, according to the invention, a computer program for magnetic resonance imaging modification is provided, comprising instructions which, when the program is executed by a computer, cause the computer to perform a method as described above.
The invention further provides a system of a magnetic resonance imaging device and a computing unit, the magnetic resonance imaging device with a switch with which the main magnetic field is switchable, with a first main magnetic field comprising a first magnetic flux density with shimming conditions for shimming a second main magnetic field to uniformity, with the second main magnetic field comprising a second magnetic flux density with shimming conditions for shimming the second main magnetic field to uniformity, wherein the second magnetic flux density is different from the first magnetic flux density, wherein the computing unit is adapted to execute the method as described above.
By allowing the magnetic flux density to change to a lower or higher flux density, artifacts in magnetic resonance images that depend on the magnetic flux density can be compensated.
In principle, different superconducting magnets can be employed. According to a preferred embodiment of the invention, however, a superconducting magnet is provided for generating the first main magnetic field and the second main magnetic field, and the superconducting magnet is a dry magnet.
Preferably, the superconducting magnet used to generate the first main magnetic field and the second main magnetic field is a high-temperature superconductor.
High-temperature superconductors allow the superconducting magnet to switch rapidly between the first magnetic flux density and the second magnetic flux density.
As to details of the training of (i) the first operator to transfer images at field/shim settings of one field strength to an image at field/shim settings of another field strength and (ii) the second operator to transfer an image at a field setting with a non-corresponding shim setting into an image the same field setting and its corresponding shim setting, these first operator and second operator can be trained on the basis of ground truth annotated training image pairs that are converted into another by the respective first operator and second operator or on the basis of a ground truth developed by computer- simulations. These operators can be implemented as neural networks that are trained to return the desired image form the input image. The second operator may be trained on a large collection of image pairs e.g., of phantoms at respective field rampings and correct associated shim settings or by Bloch simulations . The first operator may be implemented as trained neural network by training of image at one field ramp and correct/incorrect shim settings of a phantom model. Alternatively, it may be sufficient only to gather data at the incorrect shim settings and from knowledge of the induced encoding distortions owing the ‘wrong’ shim settings the correspond image at the equal field ramp and the correct shim settings may be generated.
For training of the described TH -(transformation between contrasts at different field strengths) and US- (unscrambling) networks, real data may be employed. As a general rule, the more data is provided for training, the better the results. In order to improve the training without having to measure more real data, one can think of enhancing the training data set artificially by virtue of a GAN (generative adversarial network). The GAN itself needs to be trained, but once that is done e.g. on the basis of a limited initial training set, it can generate virtually infinitely many TH and US feature/label pairs that can be used to train the above TH - and US -networks.
Even more preferably, the superconducting magnet used to generate the first main magnetic field and the second main magnetic field is a magnesium diboride magnet.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. Such an embodiment does not necessarily represent the full scope of the invention, however, and reference is made therefore to the claims and herein for interpreting the scope of the invention.
In the drawings:
Fig. 1 schematically depicts a scheme of a method according to a preferred embodiment of the invention; and
Fig. 2 schematically depicts a system of a magnetic resonance imaging device and a computing unit according to a preferred embodiment of the invention.
DETAILED DESCRIPTION OF EMBODIMENTS
Fig. 1 schematically depicts a scheme of a method according to a preferred embodiment of the invention. This method comprises steps SI to S5.
SI: First, a first magnetic resonance image which has been acquired with a magnetic resonance imaging device 1 at a first main magnetic field with a first magnetic flux density at shimming conditions for shimming a second main magnetic field to uniformity is provided, wherein the second main magnetic field has a second magnetic flux density which is different from the first magnetic flux density, S2: Second, a second magnetic resonance image which has been acquired with the magnetic resonance imaging device 1 at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity is provided,
S3: Third, a first operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity into a magnetic resonance image which would be expected at the second magnetic field at shimming conditions for shimming the second main magnetic field to uniformity is provided,
S4: Fourth, a second operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity into a magnetic resonance image which would be expected at the first magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is provided,
S5: Fifth, a magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is retrieved by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image.
The first operator is generated by a trained machine learning algorithm that has been trained with magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity. These images have not been acquired with the magnetic resonance imaging device 1 where the first magnetic resonance image and the second magnetic resonance image originate from. This first operator is only deduced once and is valid for the combination of the first magnetic flux density and the second magnetic flux density for the given set of scan parameters for which they have been determined. In case the scan parameters were changed, e.g. if repetition time (TR) or echo time (TE) would change and, therefore, a different image contrast would be achieved, a new determination would be necessary.
The second operator is also generated by a trained machine learning algorithm that has been trained with a training data including the magnetic resonance images of a magnetic resonance imaging phantom 5 for simulating a patient load acquired with the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, and the magnetic resonance images of the magnetic resonance imaging phantom 5 acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity at the respective set of given scan parameters. The first and the second operator are then implemented into a loss function in order to retrieve the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device 1 at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by minimizing this loss function which, uses the first magnetic resonance image and the second magnetic resonance image and the first operator and the second operator.
Fig. 2 schematically depicts a system of the magnetic resonance imaging device 1 and a computing unit 2 according to a preferred embodiment of the invention. The main magnetic field in the bore of the magnetic resonance imaging device 1 is generated by a superconducting magnet 6, which is a high temperature superconductor. The magnetic flux density of the main magnetic field generated by the superconducting magnet 6 is switchable by a switch 7 between the first main magnetic field with the first magnetic flux density at shimming conditions for shimming the second main magnetic field to uniformity and the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity having the second magnetic flux density different from the first magnetic flux density.
The magnetic resonance imaging phantom 5 is arranged on a treatment couch 4 of a patient positioning system 3. The switch 7 is connected to the computing unit 2 and can be controlled by it. Further, the computing unit 2 is adapted to perform the method of steps S1-S5 and thus obtain the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image, then calculates the magnetic resonance image according to the first operator onto the first magnetic resonance image and the second operator onto the second magnetic resonance image.
While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope. Further, for the sake of clearness, not all elements in the drawings may have been supplied with reference signs. REFERENCE SYMBOL LIST
1. Magnetic resonance imaging device
2. Computing unit 3 Patient positioning system
4. Treatment couch
5. Magnetic resonance imaging phantom
6. Superconducting magnet
7. Switch

Claims

CLAIMS:
Claim 1. Method of processing an acquired magnetic resonance image, the method comprising the following steps:
51) providing a first magnetic resonance image which has been acquired with a magnetic resonance imaging device (1) at a first main magnetic field with a first magnetic flux density at shimming conditions for shimming a second main magnetic field to uniformity, wherein the second main magnetic field has a second magnetic flux density which is different from the first magnetic flux density,
52) providing a second magnetic resonance image which has been acquired with the magnetic resonance imaging device (1) at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity,
53) providing a first operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity into a magnetic resonance image which would be expected at the second magnetic field at shimming conditions for shimming the second main magnetic field to uniformity,
54) providing a second operator for transforming a magnetic resonance image which has been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity into a magnetic resonance image which would be expected at the first magnetic field at shimming conditions for shimming the first main magnetic field to uniformity,
55) retrieving a magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity by applying the second operator onto the first magnetic resonance image and applying the first operator onto the magnetic resonance image which has been received by applying the second operator onto the first magnetic resonance image.
Claim 2. Method according to claim 1, wherein for performing the step S5, a trained machine learning algorithm representing the action of the first operator and the second operator is provided which, upon input of the first magnetic resonance image and the second magnetic resonance image, generates the magnetic resonance image that would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity.
Claim 3. Method according to claim 2, wherein the training data set for developing the first operator in the trained machine learning algorithm includes the magnetic resonance images acquired with the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity, and the magnetic resonance images acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
Claim 4. Method according to claim 3, wherein the training dataset for developing the first operator in the trained machine learning algorithm is generated with a generative adversarial network formed from two artificial neural networks.
Claim 5. Method according to any of claims 2 to 4, wherein the training dataset for developing the first operator in the trained machine learning algorithm is generated with electromagnetic simulations.
Claim 6. Method according to any of claims 2 to 5, wherein the training data set for developing the second operator in the trained machine learning algorithm includes the magnetic resonance images of a magnetic resonance imaging phantom (5) for simulating a patient load acquired with the first main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, and the magnetic resonance images of the magnetic resonance imaging phantom (5) acquired at the second main magnetic field at shimming conditions for the second main magnetic field to uniformity.
Claim 7. Method according to any of the previous claims, wherein in step S5 the magnetic resonance image which would be expected instead of the first magnetic resonance image if the first magnetic resonance image had been acquired with the magnetic resonance imaging device (1) at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity is retrieved by minimizing a loss function which uses the first magnetic resonance image and the second magnetic resonance image and the first operator and the second operator.
Claim 8. Method according to claim 7, wherein for the loss function LF the following formula applies:
LF = Operator l(Operator2(Imagel)) - Image2, with Operator 1 being the first operator, Operator 2 being the second operator, Image 1 being the first image and Image 2 being the second image.
Claim 9. Method according to any of the previous claims, wherein the first operator has been obtained by including the magnetic resonance image properties of magnetic resonance images acquired at the first main magnetic field at shimming conditions for shimming the first main magnetic field to uniformity and of magnetic resonance images acquired at the second main magnetic field at shimming conditions for shimming the second main magnetic field to uniformity, wherein the magnetic resonance images acquired at the first main magnetic field and the second main magnetic field, respectively, have not been acquired with the magnetic resonance imaging device (1) where the first magnetic resonance image and the second magnetic resonance image originate from.
Claim 10. Method according to any of the previous claims, wherein the second operator is obtained by acquiring magnetic resonance images from the magnetic resonance imaging phantom (5) simulating a patient load at the first main magnetic field and at the second main magnetic field, and the difference between the acquired magnetic resonance images of the magnetic resonance imaging phantom (5) at the first main magnetic field and the second main magnetic field includes the effect of the inhomogeneities of the first main magnetic field with the first magnetic flux density at shimming conditions for shimming the second main magnetic field to uniformity.
Claim 11. Computer program for magnetic resonance imaging modification, comprising instructions which, when the program is executed by a computer, cause the computer to perform a method according to any one of claims 1 to 10.
Claim 12. System of a magnetic resonance imaging device (1) and a computing unit (2), the magnetic resonance imaging device (1) with a switch (7) with which the main magnetic field is switchable, with a first main magnetic field comprising a first magnetic flux density with shimming conditions for shimming a second main magnetic field to uniformity, with the second main magnetic field comprising a second magnetic flux density with shimming conditions for shimming the second main magnetic field to uniformity, wherein the second magnetic flux density is different from the first magnetic flux density, wherein the computing unit (2) is adapted to execute the method to any of claims 1 to 11.
Claim 13. System according to claim 12, wherein a superconducting magnet (6) is provided for generating the first main magnetic field and the second main magnetic field, and the superconducting magnet is a dry magnet.
EP23820855.7A 2022-12-19 2023-12-07 Method and system for magnetic resonance imaging processing Pending EP4639191A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP22214558.3A EP4390431A1 (en) 2022-12-19 2022-12-19 Method and system for magnetic resonance imaging processing
PCT/EP2023/084645 WO2024132543A1 (en) 2022-12-19 2023-12-07 Method and system for magnetic resonance imaging processing

Publications (1)

Publication Number Publication Date
EP4639191A1 true EP4639191A1 (en) 2025-10-29

Family

ID=84537903

Family Applications (2)

Application Number Title Priority Date Filing Date
EP22214558.3A Withdrawn EP4390431A1 (en) 2022-12-19 2022-12-19 Method and system for magnetic resonance imaging processing
EP23820855.7A Pending EP4639191A1 (en) 2022-12-19 2023-12-07 Method and system for magnetic resonance imaging processing

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP22214558.3A Withdrawn EP4390431A1 (en) 2022-12-19 2022-12-19 Method and system for magnetic resonance imaging processing

Country Status (4)

Country Link
EP (2) EP4390431A1 (en)
JP (1) JP2025539500A (en)
CN (1) CN120380362A (en)
WO (1) WO2024132543A1 (en)

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10914802B2 (en) * 2019-06-24 2021-02-09 Synaptive Medical Inc. Magnetic resonance imaging systems and methods

Also Published As

Publication number Publication date
JP2025539500A (en) 2025-12-05
WO2024132543A1 (en) 2024-06-27
EP4390431A1 (en) 2024-06-26
CN120380362A (en) 2025-07-25

Similar Documents

Publication Publication Date Title
EP3500869B1 (en) Corrected compressed sensing magnetic resonance imaging
CN109073723B (en) Modifying MRI Pulse Sequence Parameters Using the History Database
US12204007B2 (en) Estimation of B0 inhomogeneities for improved acquisition and/or reconstruction of magnetic resonance images
US10576302B2 (en) Method and apparatus for monitoring a radiation therapy of a patient
Price et al. Characterization and correction of gradient nonlinearity induced distortion on a 1.0 T open bore MR‐SIM
CN107743588A (en) Use B0 inhomogeneities figure and the bone MRI of object magnetic susceptibility figure
US12360190B2 (en) Realtime design of radio-frequency pulses and gradient pulses in magnetic resonance imaging
US10921412B2 (en) Intensity corrected magnetic resonance images
WO2024245746A1 (en) Machine learning based generation of synthetic contrast-weighted images
EP4390431A1 (en) Method and system for magnetic resonance imaging processing
KR102895196B1 (en) A method of designing a pulse sequence for parallel-transmission mri, and a method of performing parallel-transmission mri using such a pulse sequence
JP2022527509A (en) Correction of magnetic resonance images using simulated magnetic resonance images
KR102090690B1 (en) Apparatus and method for selecting imaging protocol of magnetic resonance imaging by using artificial neural network, and computer-readable recording medium storing related program
CN114761817A (en) Adaptive reconstruction of magnetic resonance images
EP4220212A1 (en) Simulated computer tomographic imaging
US12078697B1 (en) System and method for adaptive magnetic resonance imaging workflows from prescan data for subjects with metal
EP4431967A1 (en) Training of neural networks to generate synthetic three-dimensional magnetic resonance images
US12579643B2 (en) System and method for anatomical features transplantation for data variety augmentation on medical images
US20250155535A1 (en) Simulated computer tomographic imaging
WO2023148039A1 (en) Simulated computer tomographic imaging

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20250721

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)