Two-state denoising for magnetic resonance imaging

Abstract

Disclosed herein is a medical system (100, 300) comprising a memory (110) storing machine executable instructions (120). The medical system also includes a computing system (104). Execution of machine executable instructions causes the computing system to: receive (200) a series (126) of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view of an object; receiving (202) the series (128) of complex valued de-noised images by inputting the series of complex valued magnetic resonance images to a first noise filtering module (122); calculating (204) a correction map (130) from the series of complex valued denoised images; reconstructing (206) a corrected magnetic resonance image (132) using the correction map and the series of complex valued magnetic resonance images; and receiving (208) a de-noised magnetic resonance image (134) by inputting the corrected magnetic resonance image to the second noise filtering module (124).

Description

Technical Field

[0001] The present invention relates to magnetic resonance imaging, and in particular to denoising of magnetic resonance images. Background Art

[0002] Magnetic resonance imaging (MRI) generally suffers from poor signal-to-noise ratio (SNR) and various types of image artifacts, which are caused by object motion, inhomogeneities of the main magnetic field B0, and imperfections of the gradient system, etc. Various methods aim to reduce noise, suppress artifacts, or both.

[0003] For example, U.S. Patent Application Publication US2022026516A1 discloses a computer-implemented method for correcting phase and reducing noise in a magnetic resonance (MR) phase image. The method includes running a neural network model for analyzing an MR image, wherein the neural network model is trained using a pair of original images and damaged images, wherein the damaged image contains damaged phase information, the original image is a damaged image with reduced damaged phase information, and the target output image of the neural network model is the original image. The method also includes receiving an MR image including damaged phase information, and analyzing the received MR image using the neural network model. The method also includes deriving an original phase image of the received MR image based on the analysis, wherein the derived original phase image includes reduced damaged phase information compared to the received MR image, and outputting the MR image based on the derived original phase image. Summary of the invention

[0004] The invention provides a medical system, a computer program and a method in the independent claims. Embodiments are given in the dependent claims.

[0005] Many magnetic resonance imaging techniques rely on multiple acquisitions or sub-acquisitions of k-space data. For various reasons, the k-space data of these acquisitions or sub-acquisitions are often inconsistent, which makes it difficult to combine them into a resulting image later. In addition, many magnetic resonance imaging techniques provide k-space data that is very noisy, which hinders any attempt to make the k-space data of these acquisitions or sub-acquisitions consistent, let alone reading the resulting image. An embodiment can provide an improved method to reduce noise and suppress artifacts caused by inconsistent k-space data in the resulting image. To achieve this purpose, a series of complex-valued magnetic resonance images is placed into a first noise filtering module to provide a series of complex-valued denoised images. Correction mapping is then performed using the series of complex-valued denoised images. For example, this can be a per-image phase correction map or a motion correction map. The correction map is then applied to the original series of complex-valued magnetic resonance images to reconstruct a corrected magnetic resonance image. The denoised magnetic resonance image is then received by inputting the corrected magnetic resonance image into a second noise filtering module.

[0006] In the framework of the present invention, the concept of a denoised (complex-valued) image involves that the denoised image is formed from the original noisy image by filtering out (irrelevant or other small-scale) changes that negatively affect the image content. This can be achieved by a conventional noise filter or by a denoising network trained on a large number of corresponding noisy, noise-free or noise-reduced images. Such noise filtering can preserve image content details on a small spatial scale. Since the high spatial frequency structure of the image content is still contained in the denoised complex-valued magnetic resonance images, the correction mapping (e.g. for the phase component of the complex-valued image) also takes into account the small high spatial frequency details. The correction mapping is then applied to the original series of complex-valued magnetic resonance images.

[0007] The first and second noise filtering modules may function in different ways. The first noise filtering module may remove uncorrelated noise. The second noise filtering may remove noise from a corrected magnetic resonance image in which the phase values ​​of the voxels or pixels have been corrected. The second noise filtering model may be different, for example in the case of a propeller (or multi-blade) acquisition, uncorrelated noise is first removed from the individual blades in the k-space obtained by the Cartesian acquisition by the first noise filtering module, and the second noise filtering module may remove or reduce spatially correlated noise generated by the blades of the Cartesian acquisition together forming a non-Cartesian (having a radial aspect in the k-space) acquisition.

[0008] The first and second noise filtering modules can function in the same manner. For example, in diffusion weighted imaging, the denoising performed by the first noise filtering module can be very error prone when operating on a signal average with a very poor signal-to-noise ratio. Performing the noise filtering first can make the correction map more robust despite the poor signal-to-noise ratio. Then, the phase-corrected average generated using the relatively robust correction map results in a diffusion weighted image with an improved signal-to-noise ratio because the noise is reduced while the signal is hardly affected. Then, the results of the second noise filtering module are more reliable because the operation of the first noise filtering module can prevent or at least reduce the propagation of artifacts in the final diffusion weighted image.

[0009] The first denoising module can operate on a complex-valued magnetic resonance image having a spatial resolution equal to or similar to the spatial resolution of a final denoised magnetic resonance image of diagnostic quality, excluding any interpolation or super-resolution steps that may be performed in post-processing or integrated in image reconstruction. When used in conjunction with individual blades of a propeller (MultiVane) acquired at different rotations in k-space, the denoising of the present invention can be applied to the (intermediate) reconstructed images of each blade. That is, the corresponding intermediate images are reconstructed from the k-space data of the corresponding blade. For each blade, the k-space may have a reduced limited spatial resolution because the k-space extension for each k-space blade is limited to the direction of the shorter axis of the blade, while the set of final magnetic resonance images reconstructed from all k-space blades has a full spatial resolution compatible with diagnostic image quality.

[0010] In one aspect, the present invention provides a medical system. The medical system includes a memory storing machine executable instructions. The memory also stores a first noise filtering module. The memory also stores a second noise filtering module. The medical system also includes a computing system. Execution of the machine executable instructions causes the computing system to receive a series of complex-valued magnetic resonance images, each image having a field of view of an object and acquired according to a magnetic resonance imaging protocol.

[0011] The magnetic resonance imaging protocol may be specified or identified by the context of the complex-valued magnetic resonance images. For example, the format of the images may implicitly specify the magnetic resonance imaging protocol. The series of complex-valued magnetic resonance images may be accompanied by data or metadata identifying the magnetic resonance imaging protocol and / or other acquisition details, or there may be metadata identifying the magnetic resonance imaging protocol accompanying the series of complex-valued magnetic resonance images.

[0012] Execution of the machine executable instructions further causes the computing system to receive a series of complex-valued denoised images by inputting the series of complex-valued magnetic resonance images into a first noise filtering module. Execution of the machine executable instructions further causes the computing system to calculate a correction map based on the series of complex-valued denoised images. Execution of the machine executable instructions further causes the computing system to reconstruct a corrected magnetic resonance image using the correction map and the series of complex-valued magnetic resonance images.

[0013] The operation of the machine executable instructions also causes the computing system to receive the denoised magnetic resonance image by inputting the corrected magnetic resonance image into the second noise filtering module. This embodiment may be beneficial because it can provide a method for producing high quality denoised magnetic resonance images. In an embodiment, a series of complex-valued magnetic resonance images are first processed using a first noise filtering module to provide a series of complex-valued denoised images. A correction map is calculated based on the series of complex-valued denoised images. The correction map is then used to reconstruct the corrected magnetic resonance image based on the original series of complex-valued magnetic resonance images. Therefore, the corrected magnetic resonance image is not degraded by the first noise filtering module. These two steps result in a denoised magnetic resonance image that may contain fewer artifacts.

[0014] In some examples, the first noise filtering module and the second noise filtering module can be the same. In other examples, the first noise filtering module and the second noise filtering module can be different. The noise filtering module used here includes a software module that receives an image as input and provides a denoised image as output. The first noise filtering module and the second noise filtering module can be implemented using an algorithm or a neural network.

[0015] Examples of neural networks that can be used to denoise images include: DnCNN, Recurrent CNN, UNet, and SimplifiedDenseNet. Specifically, Mohan et al. describe a deep convolutional neural network in "Robust and interpretable blind image denoising via bias-free convolutional neural networks" (arXiv:1906.05478), which will be suitable for implementing the first noise filtering neural network and / or the second noise filtering neural network.

[0016] In another embodiment, the correction map comprises a phase correction map calculated individually for each complex-valued denoised image in the series of complex-valued denoised images. Execution of the machine-executable instructions further causes the computing system to calculate a series of phase-corrected images by applying the phase correction map to each complex-valued magnetic resonance image in the series of complex-valued magnetic resonance images. The corrected magnetic resonance image is calculated from the series of phase-corrected images. This embodiment may be beneficial when the phases of the series of complex-valued magnetic resonance images are inconsistent.

[0017] In another embodiment, a phase correction map is calculated individually for each complex-valued denoised image in the series of complex-valued denoised images by fitting a parameterized polynomial to the spatially correlated phase of each of the series of complex-valued denoised images. This embodiment may be beneficial because it may provide a robust method to individually calculate a phase correction map for each of the series of complex-valued denoised images.

[0018] For example, a parameterized polynomial p′(x,y) of order N is used to individually describe the spatial correlation phase p(x,y) of each image in a series of complex-valued denoised images, and is defined as follows:

[0019]

[0020] Among them, parameter c m,n Determined by minimizing:

[0021] ∑x,ym(x,y)(p(x,y)-mod(p′(x,y)+π,2π)-π),

[0022] With spatial correlation magnitude m(x,y).

[0023] In another embodiment, the magnetic resonance imaging protocol is a diffusion weighted magnetic resonance imaging protocol. The corrected magnetic resonance images are calculated from the series of phase corrected images by averaging the series of phase corrected images. The embodiment may be beneficial when complex averaging the series of complex valued magnetic resonance images results in signal cancellation due to phase changes in the series of complex valued magnetic resonance images.

[0024] In another embodiment, the magnetic resonance imaging protocol is a bipolar readout magnetic resonance imaging protocol. The execution of the machine executable instructions also causes the computing system to calculate a corrected magnetic resonance image from the series of phase-corrected images by applying a phase correction map to the series of complex-valued magnetic resonance images. The execution of the machine executable instructions also causes the computing system to calculate phase-corrected k-space data by transforming the series of phase-corrected images into k-space, thereby calculating the corrected magnetic resonance image. The corrected magnetic resonance image is then reconstructed based on the combination of the phase-corrected k-space data. In essence, each phase-corrected image in the series of phase-corrected images is transformed into k-space. The k-space data is then combined and used to calculate the corrected magnetic resonance image. The combination can be achieved in different ways. In some instances, the k-space data is simply combined and merged. In other cases, the k-space data is very similar and can be averaged.

[0025] In another embodiment, the correction map comprises a main field inhomogeneity correction map calculated from the series of complex-valued denoised images. Execution of the machine-executable instructions further causes the computing system to calculate a series of phase-corrected images by applying the main field inhomogeneity correction map to each complex-valued magnetic resonance image in the series of complex-valued magnetic resonance images. A corrected magnetic resonance image is calculated from the series of phase-corrected images.

[0026] In another embodiment, the magnetic resonance imaging protocol is a Dickson magnetic resonance imaging protocol and the corrected magnetic resonance image may be a Dickson magnetic resonance image, such as a water-only image or a fat-only image. This embodiment may be beneficial because it may provide an improved Dickson imaging method.

[0027] In another embodiment, the magnetic resonance imaging protocol is an echo planar magnetic resonance imaging protocol configured to acquire a series of complex-valued magnetic resonance images with two blip gradient polarities. The echo planar magnetic resonance imaging protocol includes a series of readout gradients with alternating polarities, between which blip gradients are applied perpendicular to the readout gradients to acquire k-space data along multiple lines in opposite directions after a single excitation. The polarity of the blip gradient determines the direction in which distortion occurs when the main field is inhomogeneous. The correction map includes a distortion correction map. The execution of the machine executable instructions also causes the computing system to calculate the distortion correction map by classifying the series of complex-valued magnetic resonance images into a first group with a first blip gradient polarity and a second group with a second blip gradient polarity. The execution of the machine executable instructions also causes the computing system to calculate the distortion correction map by registering the first group of images with the second group of images.

[0028] Execution of the machine executable instructions further causes the computing system to calculate a series of distortion corrected images by applying the distortion correction map to the series of complex valued magnetic resonance images. The corrected magnetic resonance images are reconstructed by averaging the series of distortion corrected images. The average of the distortion corrected images may be complex valued or real valued. For example, the average may be an average of the amplitude values ​​of the series of distortion corrected images.

[0029] In another embodiment, the correction map includes a motion correction map calculated from a series of complex-valued denoised images relative to a reference image. In one example, the reference image can be selected from the series of complex-valued denoised images. In another example, the reference image can be another image from another source, such as a previous image acquired from the subject. One of the complex-valued denoised images is selected as a reference for all other images. A motion correction map is then determined that maps these images onto the reference image. This embodiment may be beneficial because motion correction may be more accurate when denoising is applied to these individual images.

[0030] In another embodiment, execution of the machine executable instructions further causes the computing system to calculate a series of motion-corrected complex-valued magnetic resonance images by applying the motion correction map to the series of complex-valued magnetic resonance images. Execution of the machine executable instructions further causes the computing system to calculate motion-corrected k-space data by transforming the series of motion-corrected magnetic resonance images into k-space. Corrected magnetic resonance images are reconstructed based on a combination of the motion-corrected k-space data.

[0031] In this embodiment, the series of motion corrected magnetic resonance images are first converted to k-space data to form motion corrected k-space data. The k-space data from the individual images are then combined together. This may involve simply aggregating the k-space data, or it may involve averaging the k-space data.

[0032] In another embodiment, the magnetic resonance imaging protocol is a segmented multi-shot magnetic resonance imaging protocol.

[0033] In another embodiment, the magnetic resonance imaging protocol is a propeller magnetic resonance imaging protocol.For a propeller magnetic resonance imaging protocol, the series of complex-valued magnetic resonance images are images reconstructed from individual "blades" of k-space data.

[0034] In another embodiment, the motion correction map is a rigid body map. In this embodiment, the transformation from image to image is a rigid body map without any deformation.

[0035] In another embodiment, the motion correction map is an affine map. In this embodiment, the transformation from image to image is an affine map.

[0036] In another embodiment, the motion correction map is a displacement field.

[0037] For rigid body mapping, affine mapping, and displacement fields, various algorithms can be used. For example, an anatomical atlas can be used to map points between different images. Alternatively, a shape deformable model can be used.

[0038] In another embodiment, the medical imaging system further comprises a magnetic resonance system. The memory further comprises pulse sequence commands configured to control the magnetic resonance imaging system to acquire measured k-space data according to a magnetic resonance imaging protocol. Execution of the machine executable instructions further causes the computing system to control the magnetic resonance imaging system using the pulse sequence commands to acquire measured k-space data. The k-space data is composed of a plurality of discrete acquisitions, each acquisition describing a field of view of the object.

[0039] Execution of the machine executable instructions further causes the computing system to reconstruct a series of complex-valued magnetic resonance images of the object field of view from the measured k-space data.In some examples, each discrete acquisition may be used to reconstruct a different image in the series of complex-valued magnetic resonance images.

[0040] In another aspect, the present invention provides a computer program comprising machine executable instructions. For example, the computer program may be stored on a non-transitory storage medium. The computer program also includes a first noise filtering module and a second noise filtering module. The execution of the machine executable instructions causes a computing system to receive a series of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view of an object.

[0041] Execution of the machine executable instructions further causes the computing system to receive a series of complex-valued denoised images by inputting the series of complex-valued magnetic resonance images into a first noise filtering module. Execution of the machine executable instructions further causes the computing system to calculate a correction map based on the series of complex-valued denoised images. Execution of the machine executable instructions further causes the computing system to reconstruct a corrected magnetic resonance image using the correction map and the series of complex-valued magnetic resonance images. Execution of the machine executable instructions further causes the computing system to receive a denoised magnetic resonance image by inputting the corrected magnetic resonance image into a second noise filtering module.

[0042] In another aspect, the present invention provides a medical imaging method. The method includes receiving a series of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view of an object. The method also includes receiving the series of complex-valued magnetic resonance images by inputting the series of complex-valued magnetic resonance images into a first noise filtering module. The method also includes calculating a correction map based on the series of complex-valued denoised images. The method also includes reconstructing a corrected magnetic resonance image using the correction map and the series of complex-valued magnetic resonance images. The method also includes receiving a denoised magnetic resonance image by inputting the corrected magnetic resonance image into a second noise filtering module.

[0043] It should be understood that one or more of the aforementioned embodiments of the present invention may be combined as long as the combined embodiments are not mutually exclusive.

[0044] As will be appreciated by those skilled in the art, several aspects of the present invention may be implemented as an apparatus, method, or computer program product. Thus, aspects of the present invention may take the form of a fully hardware embodiment, a fully software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects, which may be collectively referred to herein as a "circuit," "module," or "system." Additionally, aspects of the present invention may take the form of a computer program product implemented in one or more computer-readable media having a computer executable code implemented thereon.

[0045] Any combination of one or more computer-readable media can be used. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. As used herein, "computer-readable storage medium" includes any tangible storage medium that can store instructions that can be executed by a processor or a computing system of a computing device. The computer-readable storage medium can be referred to as a "computer-readable non-transient storage medium". The computer-readable storage medium can also be referred to as a tangible computer-readable medium. In some embodiments, a computer-readable storage medium can also store data that can be accessed by the computing system of the computing device. Examples of computer-readable storage media include, but are not limited to: a floppy disk, a magnetic hard disk drive, a solid-state hard disk, a flash memory, a USB thumb drive, a random access memory (RAM), a read-only memory (ROM), an optical disk, a magneto-optical disk, and a register file of a computing system. Examples of optical disks include compact discs (CDs) and digital versatile discs (DVDs), such as CD-ROMs, CD-RWs, CD-Rs, DVD-ROMs, DVD-RWs, or DVD-R discs. The term computer-readable storage medium also refers to various types of recording media that can be accessed by the computer device via a network or a communication link. For example, data may be retrieved via a modem, via the Internet, or via a local area network. Computer executable code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0046] Computer readable signal media may include a propagated data signal having computer executable code implemented therein, for example, in baseband or as part of a carrier wave. Such propagated signals may take any of a variety of forms, including, but not limited to, electromagnetic, optical, or any suitable combination thereof. Computer readable signal media may be any computer readable medium that is not a computer readable storage medium and that is capable of transmitting, propagating or transmitting a program for use by or in conjunction with an instruction execution system, device or apparatus.

[0047] "Computer memory" or "memory" is an example of a computer-readable storage medium. Computer memory is any memory that can be directly accessed by a computing system. "Computer storage" or "storage device" is another example of a computer-readable storage medium. A computer storage device is any non-volatile computer-readable storage medium. In some embodiments, a computer storage device may also be computer memory, or vice versa.

[0048] As used herein, "computing system" encompasses an electronic component capable of executing a program or machine executable instructions or computer executable code. References to computing systems that include examples of "computing system" should be interpreted as possibly including more than one computing system or processing core. A computing system may be, for example, a multi-core processor. A computing system may also refer to a collection of computing systems within a single computer system or distributed across multiple computer systems. The term computing system should also be interpreted as possibly referring to a collection or network of computing devices, each of which includes a processor or multiple computing systems. Machine executable code or instructions may be executed by multiple computing systems or processors, which may be within the same computing device, or may even be distributed across multiple computing devices.

[0049] Machine executable instructions or computer executable code may include instructions or programs that cause a processor or other computing system to perform an aspect of the present invention. Computer executable code for performing operations for various aspects of the present invention may be written in any combination of one or more programming languages ​​(including object-oriented programming languages ​​such as Java, Smalltalk, C++, etc. and conventional procedural programming languages ​​such as "C" programming language or similar programming languages) and compiled into machine executable instructions. In some cases, the computer executable code may be used in the form of a high-level language or in a precompiled form and in conjunction with an interpreter that generates machine executable instructions on the fly. In other cases, machine executable instructions or computer executable code may be in the form of programming for a programmable logic gate array.

[0050] The computer executable code may run entirely on the user's computer, partially on the user's computer, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server as a stand-alone software package. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or a connection that may be made to an external computer (e.g., through the Internet using an Internet service provider).

[0051] Aspects of the present invention are described with reference to the flowchart illustration and / or block diagram of the method, device (system) and computer program product according to an embodiment of the present invention.It should be understood that each frame or a part of a frame of a flowchart, a diagram and / or a block diagram can be implemented by a computer program instruction in the form of a computer executable code when applicable.It should also be understood that, when not mutually exclusive, in different flowcharts, the combination of blocks in the diagram and / or block diagram can be combined.These computer program instructions can be provided to the memory of a general-purpose computer, a special-purpose computer or other programmable data processing devices to produce a machine, so that the instructions run via the memory of a computer or other programmable data processing devices create a unit for implementing the function / action specified in one or more frames of a flowchart and / or a block diagram.

[0052] These machine-executable instructions or computer program instructions may also be stored in a computer-readable medium, which is capable of directing a computer, other programmable data processing apparatus, or other device to work in a specific manner, so that the instructions stored in the computer-readable medium produce an article of manufacture including instructions for implementing the functions / actions specified in the flowchart and / or one or more block diagram blocks.

[0053] The machine-executable instructions or computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operating steps to be performed on the computer, other programmable apparatus, or other device to produce a computer-implemented process, such that the instructions running on the computer or other programmable apparatus provide a process for implementing the functions / actions specified in the flowchart and / or one or more block diagram blocks.

[0054] As used in this article, "user interface" is an interface that allows a user or operator to interact with a computer or computer system. "User interface" can also be referred to as "human-machine interface device". The user interface can provide information or data to the operator and / or receive information or data from the operator. The user interface can enable the input from the operator to be received by the computer, and the output can be provided to the user from the computer. In other words, the user interface can allow the operator to control or manipulate the computer, and the interface can allow the computer to indicate the effect of the control or manipulation of the operator. The display of data or information on a display or a graphical user interface is an example of providing information to the operator. Receiving data by keyboard, mouse, tracking ball, touch pad, pointing stick, graphic input board, joystick, game board, web camera, helmet, pedal, wired gloves, remote controller and accelerometer are all examples of user interface components that realize receiving information or data from the operator.

[0055] As used herein, "hardware interface" encompasses an interface that enables a computing system of a computer system to interact with or control an external computing device and / or apparatus. A hardware interface may allow a computing system to send control signals or instructions to an external computing device and / or apparatus. A hardware interface may also enable a computing system to exchange data with an external computing device and / or apparatus. Examples of hardware interfaces include, but are not limited to: a universal serial bus, an IEEE 1394 port, a parallel port, an IEEE 1284 port, a serial port, an RS-232 port, an IEEE-488 port, a Bluetooth connection, a wireless LAN connection, a TCP / IP connection, an Ethernet connection, a control voltage interface, a MIDI interface, an analog input interface, and a digital input interface.

[0056] As used herein, a "display" or "display device" encompasses an output device or user interface suitable for displaying images or data. A display may output visual, audio, and tactile data. Examples of displays include, but are not limited to, computer monitors, television screens, touch screens, tactile electronic displays, Braille screens, cathode ray tubes (CRTs), memory tubes, bi-stable displays, electronic paper, vectorscopes, flat panel displays, vacuum fluorescent displays (VFs), light emitting diode (LED) displays, electroluminescent displays (ELDs), plasma display panels (PDPs), liquid crystal displays (LCDs), organic light emitting diode displays (OLEDs), projectors, and head mounted displays.

[0057] K-space data is defined herein as the Fourier transform of a magnetic resonance image. The k-space data in one example may be a recorded measurement of radio frequency signals emitted by atomic spins using an antenna of a magnetic resonance device during a magnetic resonance imaging scan. In another example, the k-space data is formed by performing an inverse Fourier transform on an existing magnetic resonance image.

[0058] A magnetic resonance image is defined herein as a reconstructed two- or three-dimensional visualization of anatomical data contained within magnetic resonance imaging data.Such a visualization may be performed using a computer. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In the following, preferred embodiments of the present invention will be described by way of example only and with reference to the accompanying drawings, in which:

[0060] Figure 1 An example of a medical system is illustrated;

[0061] Figure 2 It is used as a graphic Figure 1 A flowchart of a method of a medical instrument;

[0062] Figure 3 Another example of a medical system is illustrated;

[0063] Figure 4 It is used as a graphic Figure 3 A flowchart of a method of a medical instrument;

[0064] Figure 5 A flow chart illustrating another method is shown;

[0065] Figure 6 A flow chart illustrating another method is shown;

[0066] Figure 7 A flow chart illustrating another method is shown;

[0067] Figure 8 A flow chart illustrating another method is shown;

[0068] Fig. 9 A flow chart illustrating another method is shown;

[0069] Fig.10 A flow chart illustrating another method is shown;

[0070] Fig.11 The calculation of diffusion-weighted magnetic resonance images is illustrated;

[0071] Fig.12 illustrates denoising of individual images used to construct a diffusion-weighted magnetic resonance image;

[0072] Fig.13 Picture shows Fig.12 Phase correction of the image in ; and

[0073] Fig.14 A complex combination of individual images of diffusion-weighted magnetic resonance imaging is shown.

[0074] Reference numerals list

[0075] 100 Medical Systems

[0076] 102 Computer

[0077] 104 Computing System

[0078] 106 Hardware Interface

[0079] 108 User Interface

[0080] 110 Memory

[0081] 120 machine executable instructions

[0082] 122 First noise filtering module

[0083] 124 Second noise filtering module

[0084] Series of 126 complex-valued MRI images

[0085] Series of 128 complex-valued denoised images

[0086] 130 Correction Mapping

[0087] 132 Corrected MRI

[0088] 134 Denoised MRI

[0089] 200 receiving a series of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view of an object

[0090] 202 receiving a series of complex-valued magnetic resonance images by inputting the series of complex-valued magnetic resonance images into a first noise filtering module

[0091] 204 Compute a correction map based on a series of complex-valued denoised images

[0092] 206 reconstructing a corrected magnetic resonance image using the correction map and the series of complex-valued magnetic resonance images

[0093] 208 Receive the de-noised magnetic resonance image by inputting the corrected magnetic resonance image into a second noise filtering module

[0094] 300 Medical Systems

[0095] 302 Magnetic Resonance Imaging System

[0096] 304 magnet

[0097] 306 magnet bore

[0098] 308 Imaging Area

[0099] 309 Field of View

[0100] 310 magnetic field gradient coil

[0101] 312 Magnetic Field Gradient Coil Power Supply

[0102] 314 RF Coil

[0103] 316 transceiver

[0104] 318 objects

[0105] 320 object support

[0106] 330 pulse sequence command

[0107] 332 measured k-space data

[0108] 400 controls the magnetic resonance imaging system using a pulse sequence command to acquire measured k-space data;

[0109] 402 Reconstructing a series of complex-valued magnetic resonance images of the field of view of the object from the measured k-space data

[0110] 500 Discretely collected measured k-space data 332

[0111] Series of 502 phase-corrected images

[0112] 602 Motion Corrected Image

[0113] 902 Distortion Corrected Image

[0114] 1100 Diffusion Weighted Image

[0115] 1102 Orthogonal Direction

[0116] 1104 Amplitude Image

[0117] 1106 Phase Image

[0118] 1108 Summation of amplitude images

[0119] 1200 complex-valued MRI images

[0120] 1202 Complex-valued denoised image

[0121] 1204 real part

[0122] 1206 imaginary part

[0123] 1208 Amplitude

[0124] 1210 Phase

[0125] 1300 phase-corrected images

[0126] 1400Summation of complex-valued images

[0127] 1402 Summation of complex-valued denoised image

[0128] 1404 Denoised image DETAILED DESCRIPTION

[0129] In these figures, similarly numbered elements are equivalent elements or perform the same function. Elements that have been discussed previously will not necessarily be discussed in later figures if the function is equivalent.

[0130] Figure 1An example of a medical system 100 is illustrated. Medical system 100 is shown to include a computer 102. Computer 102 is intended to represent one or more computers that may be located at one or more locations. Computer 102 is shown to include a computing system 104. Similarly, computing system 104 is intended to represent one or more computing systems or cores located at one or more locations. Computing system 104 is shown to communicate with an optional hardware interface 106 and an optional user interface 108. Hardware interface 106, if present, can be used to communicate and control other components of medical system 100, if they exist. For example, hardware interface 106 can be used to control a magnetic resonance imaging system. User interface 108 can be used by a user or technician to control the operation and function of medical system 100.

[0131] Computing system 104 is also shown in communication with memory 110. Memory 110 is intended to represent one or more types of memory accessible by computing system 104. For example, memory 110 may be a non-transitory storage medium.

[0132] The memory 110 is shown as storing machine executable instructions 120. The machine executable instructions 120 enable the computing system 104 to perform basic tasks, such as computing and image processing tasks. The memory 110 is also shown as including a first noise filtering module 122 and a second noise filtering module 124. Both the noise filtering modules 122 and 124 are configured to receive an image as input and then output a denoised image. These noise filtering modules 122 and 124 can be algorithmic or based on a neural network. In some examples, the first noise filtering module 122 and the second noise filtering module 124 are the same noise filtering module.

[0133] The memory 110 is also shown as containing a series 126 of complex-valued magnetic resonance images. Each of these images 126 has a field of view of the object according to a magnetic resonance imaging protocol. The magnetic resonance imaging protocol specifies the technology used to acquire the series 126 of complex-valued magnetic resonance images. The magnetic resonance imaging protocol may also specify the relationship between the images. For example, the series 126 of complex-valued magnetic resonance images may be accompanied by metadata specifying the magnetic resonance imaging protocol. The series 126 of complex-valued magnetic resonance images may be encapsulated in a DICOM file that provides this information.

[0134] The memory 110 is further shown as containing a series of complex-valued denoised images 128. The series of complex-valued magnetic resonance images 126 is input to a first noise filtering module 122, resulting in a series of complex-valued denoised images.

[0135] The memory 110 is also shown as containing a correction map 130 calculated from the series 128 of complex-valued denoised images. The correction map 130 can take different forms in different examples. In some examples, it may be a motion correction or it may be a correction for distortion within a single image 128. The memory 110 is also shown as containing a corrected magnetic resonance image 132 calculated or reconstructed using the correction map 130 and the series 126 of complex-valued magnetic resonance images. The memory 110 is also shown as containing a denoised magnetic resonance image 134. The denoised magnetic resonance image 134 is constructed by inputting the corrected magnetic resonance image 132 into the second noise filtering module 124. It is worth noting that in some examples, the first noise filtering module 122 and the second noise filtering module 124 are the same. In other examples, the first noise filtering module 122 can be customized for the series 126 of complex-valued magnetic resonance images, and the second noise filtering module 124 can be customized for the corrected magnetic resonance image 132.

[0136] Figure 2 It is a graphic operation Figure 1 Flowchart of a method of a medical system 100 of the present invention. First, in step 200, a series of complex-valued magnetic resonance images 126 is received. Next, in step 202, a series of complex-valued denoised images 128 is received by inputting the series of complex-valued magnetic resonance images 126 into a first noise filtering module 122. Then, in step 204, a correction map 130 is calculated based on the series of complex-valued denoised images 128. Next, in step 206, a corrected magnetic resonance image 132 is reconstructed using the correction map 130 and the series of complex-valued magnetic resonance images 126. Finally, in step 208, a denoised magnetic resonance image 134 is received by inputting the corrected magnetic resonance image 132 into a second noise filtering module 124.

[0137] Figure 3 Another example of a medical system 300 is illustrated. Figure 3 The medical system 300 depicted in FIG. Figure 1 The medical system 100 in FIG. 1 further comprises a magnetic resonance imaging system 302 controlled by the computing system 104 .

[0138] The magnetic resonance imaging system 302 includes a magnet 304. The magnet 304 is a superconducting cylindrical magnet having a bore 306 passing through it. Split cylindrical magnets and so-called open magnets can also be used. A split cylindrical magnet is similar to a standard cylindrical magnet, except that the cryostat has been split into two parts to allow access to the isoplane of the magnet so that the magnet can be used, for example, in conjunction with charged particle beam therapy. An open magnet has two magnet sections, one above the other, with a space in between that is large enough to receive a subject. The arrangement of the two section areas is similar to that of a Helmholtz coil. Open magnets are popular because the subject is less confined. Inside the cryostat of the cylindrical magnet there is a collection of superconducting coils.

[0139] Within the bore 306 of the cylindrical magnet 304, there is an imaging zone 308 in which the magnetic field is strong and uniform enough to perform magnetic resonance imaging. A field of view 309 within the imaging zone 308 is shown. K-space data is acquired for the field of view 309. The region of interest may be similar to the field of view 309, or it may be a sub-volume of the field of view 309. An object 318 is shown supported by an object support 320 so that at least a portion of the object 318 is within the imaging zone 308 and the field of view 309.

[0140] Also within the bore 306 of the magnet is a collection of magnetic field gradient coils 310 for collecting measured k-space data to spatially encode magnetic spins within the imaging zone 308 of the magnet 304. The magnetic field gradient coils 310 are connected to a magnetic field gradient coil power supply 312. The magnetic field gradient coils 310 are intended to be representative. Typically, the magnetic field gradient coils 310 include a collection of three discrete coils for spatially encoding in three orthogonal spatial directions. The magnetic field gradient power supply supplies current to the magnetic field gradient coils. The current supplied to the magnetic field gradient coils 310 is controlled according to time and can be ramped or pulsed.

[0141] Adjacent to the imaging zone 308 is a radio frequency coil 314 for manipulating the orientation of magnetic spins within the imaging zone 308 and for receiving radio frequency transmissions from spins also within the imaging zone 308. The radio frequency antenna may include multiple coil elements. The radio frequency antenna may also be referred to as a channel or antenna. The radio frequency coil 314 is connected to a radio frequency transceiver 316. The radio frequency coil 314 and the radio frequency transceiver 316 may be replaced by independent transmitting coils and receiving coils and independent transmitters and receivers. It is to be understood that the radio frequency coil 314 and the radio frequency transceiver 316 are representative. The radio frequency coil 314 is intended to also represent a dedicated transmitting antenna and a dedicated receiving antenna. Similarly, the transceiver 316 may also represent a separate transmitter and receiver. The radio frequency coil 314 may also have multiple receiving / transmitting elements, and the radio frequency transceiver 316 may have multiple receiving / transmitting channels. The transceiver 316 and the gradient controller 312 are shown as being connected to the hardware interface 106 of the computer system 102.

[0142] The memory 110 is also shown as containing pulse sequence commands 330 and the measured k-space data 124. A pulse sequence is a command or a collection of commands that can be converted into commands that can control the magnetic resonance imaging system 302 to acquire the measured k-space data 124.

[0143] Figure 4 It is a graphic operation Figure 3 Flow chart of a method of a medical system 300 . Figure 4 The method described in is similar to Figure 2 The method described in . Figure 4 The method in FIG. 4 starts at step 400. In step 400, the magnetic resonance imaging system 302 is controlled by the pulse sequence command 330 so that it acquires measured k-space data 332. Next, in step 402, a series of complex-valued magnetic resonance images are reconstructed based on the measured k-space data 332. After step 402 is performed, Figure 2 As shown in , steps 200 , 202 , 204 , 206 , and 208 are performed.

[0144] Figure 5Another example of a method for diffusion weighted imaging is illustrated. Several acquisitions 400 including discrete measurements 500 provide measured k-space data 332. These measured k-space data are then reconstructed 402 into a series 126 of complex valued magnetic resonance images. They are then denoised 202 by a first noise filtering module 122 to produce a series 128 of complex valued denoised images. The series 128 of complex valued denoised images is then used to calculate or estimate 204 a correction map 130. In the example of diffusion weighted imaging, the correction map 130 is a phase correction of a single image 128. This phase correction 130 is applied to the series 126 of complex valued magnetic resonance images to produce a series 502 of phase corrected images, to which an averaging process 206 is then applied to produce an averaged corrected image or a corrected magnetic resonance image 132. The corrected magnetic resonance image 132 is finally input into a second noise filtering module 124 to produce a denoised magnetic resonance image 134.

[0145] Figure 6 Another example of a method for propeller imaging is shown. The acquisition includes several individual blades 400, each blade providing an acquisition of measured k-space data 332. These measured k-space data acquisitions are reconstructed 402 into a series of complex-valued magnetic resonance images 126, so that one image is obtained for each blade. Each blade is then denoised 202 using a first noise filtering module 122 to obtain a series of complex-valued denoised images 128. A correction map 130 is then calculated 204, which in this case is a motion correction. The motion correction 130 is applied to the original series of complex-valued magnetic resonance images 126 to produce a series of motion-corrected images 602. They are then combined 206 to produce a combined or corrected magnetic resonance image 132. For example, the motion-corrected images 602 can be transformed into k-space and the resulting k-space data can be combined together. The combined k-space data is then reconstructed into a corrected magnetic resonance image 132. The corrected magnetic resonance image 132 is then denoised 208 to produce a denoised magnetic resonance image 134.

[0146] Figure 7Another method for segmented multi-shot magnetic resonance imaging is shown. The acquisition is segmented into several interleaved excitations 400, each providing an acquisition of measured k-space data 332. Again, these individual acquisitions are reconstructed 402 into a series 126 of complex-valued magnetic resonance images, so that each excitation produces an image. Each of these images 126 is then denoised 202 using a first noise filtering module 122 to produce a series 128 of complex-valued denoised images. Motion estimation 204 is then performed to calculate a correction map 130 based on the series 128 of complex-valued denoised images. In this case, it is again a motion correction. This motion correction 130 is then applied to the original series 126 of complex-valued magnetic resonance images, respectively, to produce a motion-corrected image 602.

[0147] These motion corrected images 602 are again transformed into k-space and the resulting k-space data may be combined together 206. The combined k-space data is then reconstructed into a corrected magnetic resonance image 132. The corrected magnetic resonance image 132 is then denoised 208 using the second noise filtering module 124 to produce a denoised combined image or denoised magnetic resonance image 134.

[0148] Figure 8 Another method for bipolar readout imaging is illustrated. In this example, two acquisitions 400 are performed, which have different polarities of readout gradients. The k-space data 332 measured from the two acquisitions are reconstructed 402 again into two complex-valued magnetic resonance images 126. They are then denoised using a first noise filtering module 122 to produce a complex-valued denoised image 128. A correction map 130 is then calculated 204, which in this case is a phase correction. The correction map 130 is then applied to the complex-valued magnetic resonance image 126 to calculate a phase-corrected image 502. The two phase-corrected images 502 are then transformed into k-space, where they are combined 206 to reconstruct a corrected magnetic resonance image 132. The corrected magnetic resonance image 132 is then denoised 208 using a second noise filtering module 124 to produce a denoised magnetic resonance image 134.

[0149] Fig. 9Another method for echo planar imaging with two gradient polarities is illustrated. In this example, two acquisitions 400 are performed, which have different polarities of peak-like gradients. The k-space data 332 measured from the two acquisitions are reconstructed 402 into two complex-valued magnetic resonance images 126. They are processed into two complex-valued denoised images 128 using a first noise filtering module 122. The two complex-valued denoised images are registered to calculate a correction map 130. In this case, the correction map 130 is a distortion correction map. The correction map 130 is then applied to the complex-valued magnetic resonance image 126 to produce a distortion-corrected image 902, which is then combined 206 to obtain a corrected magnetic resonance image 132. The corrected magnetic resonance image 132 is then denoised 208 using a second noise filtering module 124 to produce a denoised magnetic resonance image 134.

[0150] Fig.10 An example of a method for Dickson magnetic resonance imaging is illustrated. In this example, two or more acquisitions 400 with different echo times are performed. The k-space data 332 measured from these acquisitions are reconstructed 402 as a series 126 of complex-valued magnetic resonance images. Each of these images is then denoised 202 using a first noise filtering module 122. This produces a series 128 of complex-valued denoised images. In this example, the calculation of the correction map 204 involves an estimate of a main field inhomogeneity correction map 130. This correction map 130 is then applied to the series 126 of complex-valued magnetic resonance images to produce a series 502 of phase-corrected images. These phase-corrected images 502 are then used to calculate individual Dickson images 132, such as a water-only image and a fat-only image. These individual Dickson images 132 are then denoised by inputting them into a second noise filtering module 124 to provide denoised magnetic resonance images 134, which in this case are denoised Dickson magnetic resonance images.

[0151] Figures 11 to 14 The method applied to diffusion-weighted magnetic resonance imaging (DWI) is shown, which mainly uses single-shot echo planar imaging (EPI) to reduce sensitivity to motion. However, it usually only provides a poor signal-to-noise ratio (SNR) even at moderate spatial resolution, so multiple averages need to be acquired. Due to motion, the phases of the individual averages are usually inconsistent, so immediate complex combination is not possible.

[0152] In this case, it is usually generated as Fig.11Diffusion-weighted magnetic resonance image 1100 is shown. In this prostate case, a total of 30 images of each slice were collected, each image having diffusion sensitization gradients (M, P, S) in three orthogonal directions. The first and second columns show the amplitude 1104 and phase 1106 of a selected image in each direction, respectively, where each row represents a direction. First, the amplitude image (M) is added by direction, while the phase image (P) is ignored. The third column shows the sum of the amplitude 1108 in each direction. Then, the trajectory is calculated as the geometric mean of each direction to obtain a trajectory image (T) or a diffusion-weighted magnetic resonance image.

[0153] Amplitude combination of individual averages can lead to undesirable accumulation of background signals, while straightforward complex combination risks canceling out foreground signals. Furthermore, the low SNR of individual averages often prevents reliable phase correction and good instantaneous denoising.

[0154] An example may use two-stage DWI denoising in the low SNR range. In the first stage, the individual averages are denoised for the sole purpose of deriving a reliable phase correction. The phase correction is then applied to the individual averages before complex combining for each direction. In the second stage, the SNR is further improved by denoising the resulting image before directional combining. In this way, an example may exploit the superior performance of complex averaging and denoising in the lower and higher SNR ranges, respectively, while avoiding potential signal cancellation.

[0155] Fig.12 Illustrated is a single complex valued magnetic resonance image 1200 input into a noise filter 202 to produce a single complex valued denoised image 1202. As this is a complex image, the real part 1204, imaginary part 1206, magnitude 1208 and phase 1210 are shown for both images.

[0156] It is worth noting that since the spatial resolution of the acquisition is only moderate and the image area where the signal can be perceived is very limited, it is not feasible to improve the signal-to-noise ratio simply by reducing the spatial resolution.

[0157] Fig.13 A magnitude image 1208 and a phase image 1210 of a single image 1202 of the series 128 of complex valued denoised images are shown. These are used to provide a phase correction applied to the single complex valued magnetic resonance image 1200. Also shown is the phase 1210 of that image and the resulting image 1300. In this case, the phase estimation involves a weighted least squares fit of a constant and a linear phase.

[0158] Fig.14The individual averaged complex combinations for each direction 1102 are shown. The magnitudes of the individual averaged complex combinations for each direction are shown. The images in column 1400 represent the series 126 of complex-valued magnetic resonance images. The images in column 1402 represent the series 128 of complex-valued denoised images. The images in column 1404 represent the denoised magnetic resonance images 134. Fig.11 In comparison, the image quality shows a great improvement.

[0159] 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.

[0160] Those skilled in the art can understand and realize other variations of the disclosed embodiments when practicing the claimed invention by studying the drawings, the disclosure and the claims. In the claims, the word "comprising" does not exclude other elements or steps, and the words "one" or "an" do not exclude multiple. A single processor or other unit can implement the functions of several items recorded in the claims. Although specific measures are recorded in mutually different dependent claims, this does not indicate that the combination of these measures cannot be used advantageously. The computer program can be stored / distributed on a suitable medium such as an optical storage medium or a solid-state medium provided together with other hardware or as part of other hardware, but can also be distributed in other forms such as via the Internet or other wired or wireless telecommunications systems. Any figure mark in the claims should not be interpreted as limiting the scope.

Claims

1. A medical system (100, 300), comprising: A memory (110) storing: machine executable instructions (120), a first noise filtering module (122), and a second noise filtering module (124); A computing system (104), wherein execution of the machine executable instructions causes the computing system to: receiving (200) a series (126) of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view (309) of a subject (318); receiving (202) a series of complex-valued denoised images (128) by inputting the series of complex-valued magnetic resonance images into the first noise filtering module; computing (204) a correction map (130) from said series of complex-valued denoised images; reconstructing (206) a corrected magnetic resonance image (132) using the correction map and the series of complex-valued magnetic resonance images; and A de-noised magnetic resonance image (134) is received (208) by inputting the corrected magnetic resonance image into the second noise filtering module.

2. The medical system according to claim 1, wherein: The correction map comprises a phase correction map calculated individually for each complex-valued denoised image in the series of complex-valued denoised images, wherein execution of the machine-executable instructions further causes the computing system to calculate a series of phase-corrected images by applying the phase correction map to each complex-valued magnetic resonance image in the series of complex-valued magnetic resonance images (502), and wherein the corrected magnetic resonance images are calculated from the series of phase-corrected images.

3. The medical system according to claim 2, wherein: The phase correction map is computed individually for each complex-valued denoised image in the series of complex-valued denoised images by fitting a parameterized polynomial to the spatially correlated phases of the series of complex-valued denoised images.

4. The medical system according to claim 2 or 3, wherein: The magnetic resonance imaging protocol is a diffusion weighted magnetic resonance imaging protocol, wherein the corrected magnetic resonance image is calculated from the series of phase-corrected images by averaging the series of phase-corrected images.

5. The medical system according to claim 1, wherein: The correction map comprises a phase correction map computed together for the series of complex-valued denoised images, wherein the magnetic resonance imaging protocol is a bipolar readout magnetic resonance imaging protocol, wherein execution of the machine-executable instructions further causes the computing system to compute the corrected magnetic resonance image from the series of phase-corrected images by: computing a series of phase-corrected images by applying the phase correction map to the series of complex-valued magnetic resonance images; and Phase-corrected k-space data are calculated by transforming the series of phase-corrected images into k-space, wherein the corrected magnetic resonance image is reconstructed from a combination of the phase-corrected k-space data.

6. The medical system according to claim 1 or 5, wherein: The correction map comprises a main field inhomogeneity correction map calculated from the series of complex-valued denoised images, wherein execution of the machine-executable instructions further causes the computing system to calculate a series of phase-corrected images by applying the main field inhomogeneity correction map to each complex-valued magnetic resonance image in the series of complex-valued magnetic resonance images, wherein the corrected magnetic resonance images are calculated from the series of phase-corrected images.

7. The medical system according to claim 6, wherein: The magnetic resonance imaging protocol is a Dickson magnetic resonance imaging protocol, and wherein the corrected magnetic resonance image is a Dickson magnetic resonance image.

8. The medical system according to any one of claims 1, 2, 3 and 4, wherein: The magnetic resonance imaging protocol is an echo planar magnetic resonance imaging protocol configured to acquire the series of complex-valued magnetic resonance images utilizing two peak gradient polarities, wherein the correction map comprises a distortion correction map, wherein execution of the machine-executable instructions further causes the computing system to compute the distortion correction map by: classifying the series of complex-valued magnetic resonance images into a first group having a first peak gradient polarity and a second group having a second peak gradient polarity; computing the distortion correction map by registering images from the first set with images from the second set; and A series of distortion corrected magnetic resonance images is calculated by applying the distortion correction map to the series of complex-valued magnetic resonance images, wherein the corrected magnetic resonance images are reconstructed by averaging the series of distortion corrected magnetic resonance images.

9. The medical system according to claim 1, wherein: The correction map comprises a motion correction map computed from the series of complex-valued denoised images with respect to a reference image.

10. The medical system according to claim 9, wherein: Execution of the machine-executable instructions further causes the computing system to: computing a series of motion corrected magnetic resonance images by applying the motion correction map to the series of complex-valued magnetic resonance images, and, Motion corrected k-space data are calculated by transforming the series of motion corrected magnetic resonance images into k-space, wherein the corrected magnetic resonance images are reconstructed from a combination of the motion corrected k-space data.

11. The medical system according to claim 10, wherein: The magnetic resonance imaging protocol is a segmented multi-excitation magnetic resonance imaging protocol or a propeller magnetic resonance imaging protocol.

12. The medical system of claim 9, 10 or 11, wherein: The motion correction map is any of the following: a rigid body map, an affine map, and a displacement field.

13. The medical system according to any one of the preceding claims, wherein: The medical system further comprises a magnetic resonance imaging system (302), wherein the memory further comprises pulse sequence commands (330) configured to control the magnetic resonance imaging system to acquire measured k-space data (332) according to the magnetic resonance imaging protocol, wherein execution of the machine executable instructions further causes the computing system to: controlling (400) the magnetic resonance imaging system using the pulse sequence commands to acquire the measured k-space data, wherein the measured k-space data comprises a plurality of discrete acquisitions, each discrete acquisition describing the field of view of the object; and The series of complex-valued magnetic resonance images, each complex-valued magnetic resonance image having the field of view of the object, is reconstructed (402) from the measured k-space data.

14. A computer program comprising: A machine executable instruction (120), a first noise filtering module (122), and a second noise filtering module (124); wherein the execution of the machine executable instruction causes the computing system to: receiving (200) a series (126) of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view (309) of a subject (318); receiving (202) a series of complex-valued denoised images (128) by inputting the series of complex-valued magnetic resonance images into the first noise filtering module; computing (204) a correction map (130) from said series of complex-valued denoised images; reconstructing (206) a corrected magnetic resonance image (132) using the correction map and the series of complex-valued magnetic resonance images; and A de-noised magnetic resonance image (134) is received (208) by inputting the corrected magnetic resonance image into the second noise filtering module.

15. A method of medical imaging, wherein: The method comprises: receiving (200) a series (126) of complex-valued magnetic resonance images according to a magnetic resonance imaging protocol, each complex-valued magnetic resonance image having a field of view (309) of a subject (318); receiving (202) a series of complex-valued denoised images (128) by inputting said series of complex-valued magnetic resonance images into a first noise filtering module; computing (204) a correction map (130) from said series of complex-valued denoised images; reconstructing (206) a corrected magnetic resonance image (132) using the correction map and the series of complex-valued magnetic resonance images; and A de-noised magnetic resonance image (134) is received (208) by inputting the corrected magnetic resonance image into a second noise filtering module.

Images