Off-center ecc

The method addresses the challenge of higher order eddy current fields at off-center MRI positions by combining 1st and Oth order ECC, optimizing ECC model parameters for off-center regions, thereby reducing image artifacts and enhancing image quality.

WO2026234450A1PCT designated stage Publication Date: 2026-11-12CANON KK
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/IB2026/000269
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-07
Filing Date
2026-05-07
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing MRI systems struggle to correct higher order eddy current fields effectively, especially when imaging is performed at off-center positions, leading to image artifacts such as incomplete fat suppression.

Method used

A method and apparatus for off-center eddy current correction (ECC) that applies both 1st order gradient ECC and Oth order B0 ECC, accounting for the induced B0 offset by the 1st order ECC, using spatiotemporal magnetic field information to determine ECC model parameters, and optimizing these terms for off-center regions.

Benefits of technology

Substantially reduces residual eddy current fields at off-center positions, improving image quality by minimizing artifacts like incomplete fat suppression and enhancing the similarity of images acquired with different phase encoding directions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2026000269_12112026_PF_FP_ABST
    Figure IB2026000269_12112026_PF_FP_ABST
Patent Text Reader

Abstract

A method for performing off-center ECC in an MRI system includes: acquiring k-space data by scanning an object using the MRI system while ECC is applied, a target imaging region of the object being positioned off-center; and reconstructing an image of the object, based on the acquired k-space data. The ECC is applied based on an ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term. The ECC model is established by: acquiring spatiotemporal magnetic field information in response to an input gradient waveform, determining, based on the spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters.
Need to check novelty before this filing date? Find Prior Art

Description

OFF-CENTER ECCCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims the benefit of priority to U.S. provisional Application No. 63 / 801,533, filed on May 7, 2025, the entire contents of which are incorporated herein by reference.BACKGROUND FIELD

[0002] The present disclosure is related to eddy current correction (ECC) in magnetic resonance imaging (MRI) systems. In particular, the disclosure is related to ECC optimized for an off-center position in MRI imaging.DESCRIPTION OF THE RELATED ART

[0003] The background description provided herein is for the purpose of generally presenting the context of the disclosure. Work of the presently named inventors, to the extent the work is described in this background section, as well as aspects of the description that may not otherwise qualify as prior art at the time of filing, are neither expressly nor impliedly admitted as prior art against the present disclosure.

[0004] In a magnetic resonance imaging (MRI) system, magnetic field gradient switching can induce eddy currents (ECs) in conducting structures of the MRI system. The eddy currents can give rise to magnetic field changes that cause deviations from desired field gradients.

[0005] As a result, eddy currents can affect image acquisition, including the k-space trajectory of magnetic resonance data prior to image reconstruction. Therefore, eddy currentcorrection (ECC) is typically used in modern MRI systems to counteract unwanted effects from eddy currents.

[0006] Eddy current fields can be of Oth order, 1st order, and higher order in space. In most MRI systems, gradient ECC (1st order ECC) and BO ECC (Oth order ECC) are implemented. Higher order ECC by actuating higher order shim coils is possible, but is not commonly done on clinical scanners.

[0007] Generally, gradient ECC pre-emphasis can be calculated from an input gradient and an ECC model. The ECC model is often implemented using convolutions and the input (or target) gradient. For example, an ECC kernel can be implemented as a convolution of multiexponential terms with a derivative of the nominal input gradient, though purely numerical kernels can also be used. The ECC kernel can include a self-term component and a crossterm component.

[0008] B0 ECC (i.e., Oth order) for compensating Oth order cross-terms is often implemented by adjusting receiver and transmit frequency or phase. In general, ECC can be performed at run-time of a sequence, for example, on an FPGA.

[0009] To calculate or calibrate the ECC model, measurements that are sensitive to eddy currents usually need to be acquired. Such measurements can involve applying an eddy current-inducing input gradient, followed by readout of a spatiotemporal field response during and / or following the input gradient. The measurements can be acquired using magnetic field sensors or a spatiotemporally resolved MRI imaging sequence, for example.SUMMARY

[0010] In one embodiment, the present disclosure is directed to a method for performing off-center eddy current correction (ECC) in a magnetic resonance imaging (MRI) system. The method includes: acquiring k-space data by scanning an imaging object using the MRIsystem while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; and reconstructing an image of the imaging object, based on the acquired k-space data. The ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term. The ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0011] In one embodiment, the present disclosure is related to an apparatus for performing off-center ECC in an MRI system. The apparatus includes processing circuitry configured to: acquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system, and reconstruct an image of the imaging object, based on the acquired k-space data. The ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zeroorder ECC term is determined by taking into account a zero-order component induced by the first-order ECC term. The ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters thatdefine the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0012] In one embodiment, the present disclosure is related to an MRI system including: first processing circuitry, configured to acquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; and second processing circuitry, configured to reconstruct an image of the imaging object, based on the acquired k-space data. The ECC is applied based on an established ECC model, and comprises a zeroorder ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term. The ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0013] Note that this summary section does not specify every embodiment and / or incrementally novel aspect of the present disclosure or claimed invention. Instead, the summary only provides a preliminary discussion of different embodiments and corresponding points of novelty. For additional details and / or possible perspectives of the invention and embodiments, the reader is directed to the Detailed Description section and corresponding figures of the present disclosure as further discussed below.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:

[0100] FIG. 1 shows spatial distributions of eddy current field components in a target off-center region, and the concept of combined 1st order and Oth order eddy current correction in accordance with embodiments of the present disclosure;

[0015] FIG. 2 provides an overview of exemplary processing for determining off-center optimized eddy current correction (ECC) parameters;

[0016] FIG. 3 shows exemplary off-resonance field distributions measured in a transversal plane (z = 0) at 3 ms after the end of an input gradient lobe;

[0017] FIGs. 4A and 4B illustrate how using opposite readout polarities can reduce inaccuracy caused by local off-resonance in phantom-based field response measurement;

[0018] FIGs. 5A-5D show exemplary residual eddy-current-related field effects at a 20 cm off-center position, as measured using a dynamic field camera (DFC);

[0019] FIG. 6 shows exemplary fat-suppressed T2 Fast Spin Echo (FSE) images acquired at an off-center position with two different phase encoding directions, posterior-to-anterior (PA) and anterior-to-posterior (AP);

[0020] FIG. 7 shows the effects of off-center optimized ECC determined from phantom data;

[0021] FIG. 8 illustrates a flowchart of an exemplary image reconstruction method in accordance with embodiments of the present disclosure; and

[0022] FIG. 9 is a schematic block diagram of a magnetic resonance imaging (MRI) system, according to one embodiment of the present disclosure.DETAILED DESCRIPTION

[0023] The following disclosure provides different embodiments, or examples, for implementing different features of the provided subject matter. Specific examples of components and arrangements are described below to simplify the present disclosure. These are, of course, merely examples and are not intended to be limiting.

[0024] For example, the order of discussion of the different steps as described herein has been presented for clarity’s sake. In general, these steps can be performed in any suitable order. Additionally, although each of the different features, techniques, configurations, etc. herein may be discussed in different places of this disclosure, it is intended that each of the concepts can be executed independently of each other or in combination with each other. Accordingly, the present disclosure can be embodied and viewed in many different ways.

[0025] Furthermore, as used herein, the words “a,” “an,” and the like generally carry a meaning of “one or more,” unless stated otherwise.

[0026] Magnetic resonance imaging (MRI) involves the generation of switched gradients in the magnetic field during an imaging process, so as to determine spatial information from a magnetic resonance (MR) signal. The switched gradients can induce eddy currents (ECs), which in turn result in a secondary magnetic field superimposed on the gradient pulses.

[0027] The eddy currents present during image acquisition can affect how the MR data is acquired and processed. Specifically, the eddy currents can distort the actual gradient fields experienced during the image acquisition, because the resulting gradient waveform represents a superposition of the intended gradients generated by the MRI system and the fields arising from the eddy currents. These deviations in the spatiotemporal field evolution can affect how the MR signal is mapped into k-space and how signal and contrast are generated, for example. Thus, the resulting image may include artifacts, unless the deviations caused by the eddy currents can be characterized and corrected.

[0028] Eddy current-related fields often manifest in spatially higher orders. However, typical eddy current correction (ECC) implemented on MRI systems only adjusts the Oth order and 1st order components, while higher order eddy current fields are generally not corrected.

[0029] When imaging is performed at an off-center position, such higher order eddy current fields can become substantial and can lead to image artifacts. For example, in fat-suppressed Fast Spin Echo (FSE) shoulder MRI imaging, higher order eddy current fields can contribute to incomplete fat suppression at off-center positions.

[0030] To address this problem, the present disclosure provides a method and apparatus for performing MRI imaging at a target off-center region, in which both 1st order gradient ECC and Oth order B0 ECC are specifically determined to minimize higher order eddy current fields at the target off-center region.

[0031] An important observation is that, when a 1st order ECC gradient is applied at an off-center position, an additional Oth order (B0) field offset is induced at that position. When determining the Oth order ECC, this induced B0 offset should be accounted for. Failure to do this will result in a residual B0 error at the off-center region, which cannot be corrected by the 1st order ECC alone.

[0032] Accordingly, the Oth order ECC determined by the approaches provided in the disclosure involves a compensation for both the direct B0 eddy current component at the off-center region and the B0 offset induced at that region by the application of the 1st order ECC.

[0033] FIG. 1 illustrates spatial distributions of eddy current field components in a target off-center region, and the concept of combined 1st order and Oth order eddy current correction in accordance with embodiments of the present disclosure. The vertical axis B represents the magnetic field strength, and the horizontal axis X represents the spatial position. The isocenter corresponds to the origin, while the target off-center imaging region is located at aposition displaced from the isocenter along the X axis. The field components shown in FIG.1 represent the spatial field distributions after an input gradient waveform is applied.

[0034] In FIG. 1, the solid curve labeled “Higher order EC” represents the higher order eddy current field. This field is spatially nonlinear. In the target off-center region along the X axis, two distinct components are shown. One is a gradient component (i.e., a local slope), indicated by the solid line labeled “EC (1st order) Gradient at off-center,” and the other is a Oth order component (i.e., a local offset), indicated by the upward arrow labeled “BO (Oth order) EC at off-center.”

[0035] In FIG. 1, the dashed line labeled “ECC (1st order) Gradient to compensate EC gradient at off-center” represents a 1st order ECC applied to compensate for the gradient component of the higher order eddy current field in the target off-center region. For example, the slope of this line can be opposite to the local slope of the higher order eddy current field in the target off-center region, thereby canceling the gradient component.

[0036] However, this compensating gradient field has a nonzero value in the target off-center region, and thus its application introduces a B0 offset in the region, indicated by the downward arrow labeled “1st order ECC induced B0 (Oth order) off-set induced by (1st order) gradient at off-center.” As shown in FIG. 1, this induced B0 offset is larger in magnitude than the direct B0 eddy current component, and acts in the opposite direction.

[0037] Accordingly, the lower dashed curve labeled “Higher order EC with 1st order ECC” in FIG. 1 represents the residual field remaining after the application of the 1st order ECC. As shown in FIG. 1, a net B0 offset remains in the target off-center region, which corresponds to the difference between the induced B0 offset and the direct B0 eddy current component. If uncorrected, this residual B0 offset can cause off-resonance effects in the off-center region, leading to problems such as incomplete fat suppression.

[0038] To correct for this residual BO offset, the present approaches apply a Oth order BO ECC, represented by the arrow labeled “Combined ECCBO to correct at off-center” in FIG. 1. The magnitude of this Oth order ECC can be determined to account for both the direct BO eddy current component and the BO offset induced by the 1st order ECC.

[0039] In FIG. 1, the upper dashed curve labeled “Target: Higher order EC with Oth and 1st order ECC” represents the residual field after the combined application of the 1st order ECC and the Oth order ECC. As shown in FIG. 1, the field in the target off-center region can be substantially reduced and approaches 0.

[0040] The following provides a specific example illustrating how ECC parameters for a target off-center region can be determined in accordance with embodiments of the present disclosure. FIG. 2 is a diagram illustrating an overview of exemplary processing steps for determining off-center optimized ECC parameters. For example, spatiotemporal magnetic field response can be acquired while the MRI system is operating under a default ECC configuration (referred to as “default ECC”) that has been optimized for imaging at the isocenter region. Thus, the measured response under this condition represents the residual eddy current field response remaining after application of the default ECC. As will be described below, the off-center optimized ECC parameters can be determined based on this measured magnetic field data.

[0041] The following description provides complete mathematical expressions for the blocks shown in FIG. 2. In the schematic figure of FIG. 2, certain expressions can be depicted in a simplified form for ease of illustration.

[0042] As discussed previously, spatiotemporal magnetic field data can be acquired using magnetic field sensors, e.g., a dynamic field camera (DFC). To characterize the eddy current response of the MRI system, a calibration sequence comprising a long trapezoidal gradient lobe can be applied along each input gradient axis I, where I 6 {X, Y, Z}. For example, sucha calibration sequence can be a ramp-down sequence, as shown in FIG. 2. In this example, the target off-center imaging region is displaced from the isocenter in the X direction.Accordingly, for each input axis, the resulting spatiotemporal field evolution is recorded by the DFC at a plurality of off-center positions along the X axis, so as to cover the entire range of the target off-center imaging region. The measurement can be repeated multiple times with shifted time delays.

[0043] For each input axis I, the raw field data acquired by the DFC can be expanded into spherical harmonic basis functions bm(u) to obtain a set of k-space coefficients km(t), where u denotes a spatial position vector (x, y, z), and t denotes time. The spatially resolved phase evolution cp(u,t) can be obtained by evaluating the k-space coefficients km(t) on the spatial position u, according to:<p(u,t) = Sm=l km(t)bm(u) (1)where M denotes a total number of the spherical harmonic basis functions used in the expansion. In one embodiment, up to 64 basis functions can be used.

[0044] From the phase evolution cp(u,t), the spatially resolved magnetic field offset B(u,t) can be obtained via time differentiation, according to:B(u,t) = = E^i^(t)hm(u) / y (2)where y denotes a gyromagnetic ratio.

[0045] The local field gradients at the spatial position u can be described by the spatial partial derivatives of B(u,t) with respect to each output axis O 6 {X, Y, Z}:H(u,t) = S^=1^(t) / K^(u) (3)

[0046] As an alternative to partial derivation of the basis functions, new coefficients P can be numerically calculated and fitted to the spherical harmonic basis functions according to: dB= Sm=l 0)^00 (4)

[0047] The measured local gradient for an input axis I and an output axis O can be expressed as:W / K = S"=l (5) where pm i0(t') are coefficients that describe the measured spatiotemporal eddy current field response for each input-output axis combination, and constitute the primary input to a subsequent optimization phase. The quantities Go, Gx, Gy, and Gzin FIG. 2 correspond to B dxB(u,t), dyB(u,t), and dzB(u,t), respectively. The measurement processing phase described above correspond to the upper portion of FIG. 2 labeled “Measurement Processing.”

[0048] The nominal gradient waveform G7vom(t) denotes the intended gradient waveform applied along the input axis I, prior to an ECC pre-emphasis. The ECC gradient produced on the MRI system for the input axis I and the output axis O can be expressed as:(6)where aI O i(u) denotes a spatially varying gain, rz o :idenotes a time constant, and N denotes a total number of r per input-output axis combination.

[0049] The gain aI O i(u) can be expressed as a linear combination of spatial basis functions pm,(u) and coefficientsasfollows:« / ,O,i(w) Sm,_ i I.O.rru.i (7)where M' denotes a total number of the spatial basis functions used to express the gain a. In one embodiment, spherical harmonic basis functions can be chosen as pm,(u), as these are motivated by electrostatics and are also used for the k-space coefficients described above. Time-dependent values can be introduced in a term c as follows:= (8)

[0050] The temporal evolution associated with one or more ECC terms, such as cI 0 Tt'), can be calculated and stored for reuse in later steps of the ECC optimization process. For example, the stored temporal evolution can be reused when fitting the coefficients f and / or when evaluating the loss for different candidate time constant combinations.

[0051] In summary, an ECC forward model can be expressed in the following form:GE~CC(.^ = SiU a / ,o,i(w)c / ,o,Ti(t) = S^;=1(pm,(u)^,0,mU) c / j0,T.(t) (9)

[0052] Note that in the present example, T is chosen to be independent of u, which is consistent with standard ECC pre-emphasis implementations. However, in general, T could also be a function of u.

[0053] The optimization phase determines, for each input-output axis combination I— >0, the time constants Ti,o,i and coefficients < ,o,m that best fit the measured field dataThe optimization phase corresponds to the middle portion of FIG. 2 labeled “Optimization.”

[0054] Firstly, a set of candidate time constants Ti,o,icanbe selected. In practice, a total number N of the time constants is limited by the capacity of the MRI system’s ECC subsystem, for example. Typically, the ECC subsystem only accepts a certain number of terms.

[0055] For each candidate set Ti:o,b the corresponding coefficients ^i:o,m,iaredetermined by minimizing a difference between the simulated ECC response and the measured field coefficients as follows:argmin^O m|| $m,t ■ Q.o.rXO) - || (10) where cI 0 Tt') is as defined in Equation (8). This minimization can be a linear least squares problem, which can be solved efficiently for each basis function index m independently.

[0056] The quality of the fit achieved by a given candidate set rz ocan be evaluated by a loss function:argminTF(Sm=i((E1N=i U,i ’ Q.O.TXO) ’ bm(u) - • bm(u)) (11) where F denotes a predetermined loss metric, and the loss is evaluated over spatial positions u within a target region of interest (ROI), rather than at a single spatial point. The ROI corresponds to a predefined off-center imaging region for which the ECC parameters are being optimized. For instance, in the case of shoulder MRI imaging, the ROI can be a region displaced approximately 20 cm from the isocenter in the X direction.

[0057] For example, the loss F can be defined as a temporal norm, such as an integral over time, a peak-to-peak amplitude over time, a 2-norm over time, etc. As another example, F can be defined as a spatial metric over the ROI, such as an integral over space in the ROI, a peak-to-peak amplitude in the ROI, a 2-norm over the ROI, etc. Other forms of the metric F can also be used, e.g., a spatiotemporal combination.

[0058] By evaluating the loss over the target ROI rather than at a single spatial point, the optimization yields ECC parameters that minimize residual eddy current fields across an entire target off-center imaging region.

[0059] The optimization can proceed by exhaustive search over a predefined set of candidate time constant combinations Ti,o,i- Foreach candidate combination, the least squares fit ofcanbe performed, and the resulting loss value can be calculated and stored. This process is repeated for each candidate combination in the set, until all combinations have been evaluated. Upon completion of the exhaustive search, the combination rz o :iyielding the minimum loss value is selected.

[0060] Optionally, the time constants selected through the above exhaustive search can be further refined by a non-linear local optimization algorithm, using the exhaustive search result as the initial starting point:argminT(12)Compared with what is achievable by the exhaustive search alone, this optional refinement can further improve the accuracy of the final ECC model parameters.

[0061] Accordingly, a set of optimized time constantsO iand coefficientscanbe determined for each input-output axis combination I— >0. These parameters are illustrated as the output of the “Optimization” portion of FIG. 2, constituting the spatiotemporal ECC model used in the subsequent ECC parameter calculation phase.

[0062] The ECC parameter calculation phase can use the optimized time constants Ti,o,i and coefficients fobtained from the optimization phase to determine the ECC parameters to be applied by the MRI system for a given target off-center position u. This phase corresponds to the lower portion of FIG. 2 labeled “ECC.”

[0063] The objective of the ECC calculation is to determine ECC parameters, such that the ECC field cancels the measured eddy current field as follows:^cc(^t) = - G^s(u,t) (13)

[0064] Substituting the ECC forward model from Equation (9) and the measured field response from Equation (5):^1=1 = - ^m=l Pmj,o(t)bm(u) (14)

[0065] Substituting the expression for cr / 0,(zz) from Equation (7) into Equation (14):Sr=lK,'=1(15) According to embodiments of the disclosure, spherical harmonic basis functions bm(ii) can be chosen as the spatial basis functions pm,(u) of the gain cr / 0, ( i) . Spherical harmonics are motivated by electrostatics and are also used for the k-space coefficients described above. In addition, spherical harmonics can more compactly model ECCs, compared to single-axis polynomial basis functions typically used in conventional ECC approaches.

[0066] While spherical / solid harmonic basis functions are described above as one example, other harmonic basis functions can also be used, for example, cylindrical harmonic, etc.

[0067] With this choice, Equation (15) becomes:^=i(Sm=i bm(u^IiOimijQ,0,Ti(t) = - Hi ?mj,o(t)bm(u) (16) which can be rearranged as:^=iter=i c / ,0,Ti(t)^,0,m,i)bm(u) = - Hi ?mj,o(t)bm(u) (17)

[0068] Since this must hold for each spherical harmonic basis function bm(u), T and S, need to be found such that:^=1 (18)This condition needs to be satisfied once per ECC calibration, and can be achieved through the optimization phase.

[0069] Once the optimized Ti,o,i andhavebeen determined, the ECC gain cr / 0,(zz) for a given target off-center position u can be calculated as:« / ,O,i(w) ^m=l bm(u)^i Qmi (19)Together with the time constants Ti:o,b these gains cr / ,o,i(w) constitute the ECC parameters to be applied by the MRI system’s real-time ECC subsystem for a target off-center position u.

[0070] In some embodiments, the target off-center position u can be provided as an input to the real-time ECC subsystem. The real-time ECC subsystem can calculate 1st order ECC samples for the target off-center position u, and further calculates, based on the 1st order ECC samples and the target off-center position u, a corresponding induced Oth order ECC contribution. The induced Oth order ECC contribution can then be added to other Oth order ECC terms to be applied by the real-time ECC subsystem.

[0071] As discussed previously, when a 1st order ECC gradient is applied at the off-center position u, an additional Oth order (B0) field offset can be induced at that position.According to embodiments of the present disclosure, this induced B0 offset should be accounted for when determining the Oth order ECC parameters.

[0072] Specifically, for each 1st order ECC output term along an output axis X, Y, or Z, the induced B0 term can be calculated as follows:a[°(u) = — alxux, = ilxa-°(u) = — a[Yuy, = r-ra|°(u) = -aiZuz> (20)where a °(u) and i(° denote the gain and time constant of the induced BO term, ux, uy, uzdenote the X, Y, Z components of the off-center position vector u, and ax, a1azdenote the 1st order ECC gains along each output axis X, Y, and Z as calculated in Equation (19).

[0073] Although Equations (5)-(l 9) are formulated for general input-output axis combinations I— >0, Equation (20) illustrates the calculation of induced Oth order ECC using same-axis 1st order ECC terms as an example. The same approach can also be applied to cross-terms. For example, where switching of a gradient along one axis gives rise to an ECC gradient along another axis, the resulting contribution to the induced Oth order ECC at an off-center position can likewise be included. Thus, where a Y gradient gives rise to an X-gradient cross-term, the corresponding contribution to the induced Oth order ECC at an X off-center position can also be included.

[0074] The induced Oth order terms can be assigned the same time constants as the corresponding inducing 1st order ECC terms. Accordingly, the induced Oth order correction can be achieved by adjusting an amplitude of an existing Oth order ECC term having the same time constant as the inducing 1st order ECC term, without introducing an additional term.

[0075] Further, the original Oth order ECC terms and the additional Oth order ECC terms induced by the 1st order ECC can be combined and represented by a reduced number of Oth order ECC terms. The reduced number of terms is smaller than the sum of the numbers of the original Oth order ECC terms and the induced Oth order ECC terms. For example, amplitudes and time constants of the reduced number of terms can be adjusted, so that ECC similar tothat achieved using the original Oth order ECC terms and the induced Oth order ECC terms is maintained.

[0076] Accordingly, the induced BO terms calculated from Equations (20) can be added to the Oth order ECC terms to yield a combined Oth order ECC correction in the target off-center region. The ECC parameter calculation phase described above correspond to the lower portion of FIG. 2 labeled “ECC.”

[0077] Although the above example describes acquiring spatiotemporal magnetic field data while the MRI system is operating under a default ECC configuration that has been optimized for imaging at the isocenter region, the present disclosure is not limited to this. One skilled in the art will appreciate that the methods can also be applied to systems in which field measurements are acquired under other ECC conditions, or in which a complete set of off-center optimized ECC parameters is determined without reference to a pre-existing ECC configuration.

[0078] In the example described above, the spatiotemporal magnetic field response is acquired using a DFC. In an alternative embodiment, the spatiotemporal magnetic field data used to determine the spatiotemporal ECC model can be acquired by performing MRI imaging on a phantom. The calculation of the spatiotemporal magnetic field response from the phantom image phase data is conceptually similar to the DFC-based approach described above. Additionally, the phantom-based approach may involve considerations specific to off-center imaging, which will be described below.

[0079] For off-center imaging applications such as shoulder MRI, where the imaging region is displaced from the isocenter in the X direction, the phantom should span a sufficient range along the X axis to cover the target off-center imaging region. In certain embodiments, the phantom can be of a rectangular shape spanning at least 40 cm in the X direction. Forexample, a phantom of dimensions 45 cm * 15 cm * 15 cm in the X, Y, and Z directions can be used.

[0080] While DFC measurements can provide accurate spatiotemporal field data, the DFC-based method requires specialized external hardware that may not be available in all settings. By using standard MRI equipment, the phantom-based approach can provide a practical alternative implementation.

[0081] To acquire the phantom-based field data, the phantom image data can be acquired using a time-resolved gradient echo sequence, such as a 2D or 3D fast field echo (FFE) sequence, following an input gradient applied to generate the eddy currents to be characterized. The eddy current-induced magnetic field response can be extracted from the phase information of the phantom images for each echo of multiple echoes, for example, after spatial and temporal unwrapping.

[0082] In the phantom-based approach, local static B0 off-resonance at off-center positions can cause spatial distortions in the phantom images. If uncorrected, these spatial distortions can affect the accuracy of the calculated eddy current gradient fields. In the following examples, the effects of the spatial distortions are corrected using different methods, so as to improve the accuracy of the phantom-based approach at the off-center positions.

[0083] In one embodiment, the effects of the spatial distortions can be reduced by acquiring phantom image data with opposite readout polarities. For example, field information can be derived from acquisitions performed with a first readout polarity and with a second, reversed readout polarity. The field information derived from the two acquisitions can then be combined, for example, by taking a mean, so as to reduce the distortion-related effects in the obtained eddy current fields. The reversed-polarity readout gradient can be played out in the same sequence block directly after the first-polarity readout gradient, without applying an radiofrequency (RF) pulse in between.

[0084] Additionally or alternatively, phantom image data can be acquired with opposite slice-selection polarities. Similarly, field information derived from the acquisitions with the opposite slice-selection polarities can be combined (e.g., by taking a mean), so as to reduce the effects of static BO-related spatial distortions on the obtained eddy current fields.

[0085] In another embodiment, the correction can be achieved using a static BO map acquired in a separate scan. The static BO map can be used to correct for the spatial distortions of the phantom images caused by static BO off-resonance, or to account for the effects of the spatial distortions on the obtained eddy current fields. For example, the static BO map can be acquired using a gradient echo scan with at least two different echo times. Alternatively, the static BO map can be obtained from measurements acquired for a static shimming procedure.

[0086] In either case, the corrected phantom-based field data can be subsequently processed in a similar manner as the DFC-based field data discussed above, so as to obtain the coefficients which can then be used as input to the optimization phase shown in FIG. 2.

[0087] Note that the spatial distortion correction methods described above can be used alone or in combination with other means for performing phase correction of local static off-resonance. For example, In some embodiments, for a given input axis, the phantom-based measurement can be performed twice, one with a positive input gradient, and the other with a negative input gradient. Data acquired with the positive and negative input gradients can be used to reduce phase effects caused by local static off-resonance. The correction of the spatial distortions described above can be used alone, or in combination with this phase correction. For example, even when phase-related off-resonance effects may be reduced using data acquired with positive and negative input gradients, spatial distortions of the phantom images can still affect the obtained eddy current fields. In this situation, theapproaches described above, such as combining field information derived from acquisitions with opposite readout polarities, combining field information derived from acquisitions with opposite slice-selection polarities, and / or using a static BO map, can be further used to reduce the distortion-related inaccuracy in the obtained eddy current fields.

[0088] FIG. 3 shows exemplary off-resonance field distributions measured with a DFC and visualized in a transversal plane (z = 0) at 3 ms after the end of an input gradient lobe. In each row of FIG. 3, the left heat map shows a field distribution obtained using regular isocenter optimized ECC, and the right heat map shows a field distribution obtained using ECC optimized to reduce eddy-current effects at a target off-center region. The square shown in each heat map in the right column corresponds to the target off-center region. The top row corresponds to a situation where the input gradient lobe is in the X direction, the middle row corresponds to a situation where the input gradient lobe is in the Y direction, and the bottom row corresponds to a situation where the input gradient lobe is in the Z direction.

[0089] In the left column of FIG. 3, the off-resonance field distributions do not show visible spatial variation at the isocenter, indicating that the regular iso-center optimized ECC can provide satisfactory correction at the isocenter. However, in regions displaced from the isocenter, substantial off-resonance spatial variations exist.

[0090] In accordance with embodiments of the disclosure, ECC can be optimized for the target off-center region identified by the square, which is displaced by about +20 cm in the X direction. As shown in the right column of FIG. 3, the off-resonance spatial variations within the target off-center region are strongly reduced relative to those shown in the left column.

[0091] FIGs. 4A and 4B further illustrate how using opposite readout polarities can reduce inaccuracy caused by local off-resonance in phantom-based field response measurement.

[0092] FIG. 4A shows two magnitude images of a phantom acquired using opposite readout polarities. The upper image is acquired with a first readout polarity, and the lower image isacquired with a second, reversed readout polarity. As shown, local off-resonance causes the two images to appear with different spatial distortions.

[0093] FIG. 4B shows eddy-current gradients in the Z direction (Hz / m) measured shortly after an eddy-current-stimulating gradient is played out in the X direction. In particular, portion (a) of FIG. 4B shows reference eddy-current gradients measured using a DFC, and portion (b) shows eddy-current gradients obtained using a gradient echo sequence on a phantom with one readout polarity. Portion (c) shows a difference map between (a) and (b). Portion (e) shows eddy-current gradients obtained by taking the mean of the gradients acquired from the phantom with positive and negative readout polarities. Portion (f) shows a difference map between (a) and (e). Comparison of portions (c) and (f) shows that, after taking the mean of the eddy-current gradients acquired with positive and negative readout polarities, the phantom-based result more closely matches the DFC-based reference.

[0094] FIGs. 5A-5D show exemplary residual eddy-current-related field effects at a 20 cm off-center position, as measured using a DFC. FIG. 5A corresponds to a default ECC, FIG.5B corresponds to a DFC-based off-center optimized ECC, FIG. 5C corresponds to an initial phantom-data-based off-center ECC, and FIG. 5D corresponds to an improved phantom-data-based off-center ECC.

[0095] In each of FIGs. 5A-5D, the left graph shows a 1st order off-resonance gradient component in Hz / m at the off-center position, and the right graph shows a Oth order off-resonance component in Hz at the off-center position. In the left graph, dco / dX, dco / dY, and dco / dZ denote gradient components of the off-resonance in the X, Y, and Z directions, respectively, and in the right graph, co denotes the Oth order off-resonance component. The horizontal axis in each graph represents a delay in ms after ramp-down.

[0096] FIG. 5 A shows the residual eddy-current-related field effects when the default ECC, i.e., an iso-center optimized ECC, is applied. As shown, substantial residual 1st order off-resonance gradients and a substantial residual Oth order off-resonance remain at the off-center position. Especially, at that off-center position, the Oth order off-resonance can exceed about 100 Hz, and the 1st order off-resonance gradients can reach several hundred Hz / m.

[0097] FIG. 5B shows the residual eddy-current-related field effects when an off-center optimized ECC determined from DFC measurements is used. Compared with FIG. 5 A, both the residual 1st order off-resonance gradients and the residual Oth order off-resonance component are substantially reduced. Additionally, FIG. 5B provides a reference benchmark against which the phantom-data-based results of FIGs. 5C and 5D can be evaluated.

[0098] FIG. 5C shows the residual eddy-current-related field effects when an initial phantom-data-based off-center ECC is used. This initial phantom-data-based result corresponds to a phantom-based approach in which static B0 off-resonance-related spatial distortions of the phantom images are not corrected. Compared with the default ECC result shown in FIG. 5 A, the residual field effects are reduced. However, effects of the spatial distortions of the phantom images caused by static B0 off-resonance are still reflected in the determined ECC. As a result, the result shown in FIG. 5C still differs substantially from the DFC-based reference benchmark shown in FIG. 5B.

[0099] FIG. 5D shows the residual eddy-current-related field effects when an improved phantom-data-based off-center ECC is used. This improved phantom-data-based result corresponds to a phantom-based approach in which field information derived from acquisitions with opposite readout polarities is combined, for example, by taking a mean, so as to reduce effects of static B0 off-resonance-related spatial distortions of the phantom images on the obtained eddy current fields. Compared with FIG. 5C, the improved phantom-data-based off-center ECC further reduces the residual 1st order off-resonance gradients and the residual Oth order off-resonance component. As a result, FIG. 5D more closely approaches the DFC-based reference benchmark shown in FIG. 5B.1

[0100] As discussed above, effects of static BO off-resonance-related spatial distortions can be reduced using other methods. FIG. 5D illustrates one example using opposite readout polarities. In other examples, such effects can be reduced by combining field information derived from acquisitions with opposite slice selection polarities or by using a static BO map.

[0101] FIG. 6 shows exemplary fat-suppressed T2 FSE images acquired at an off-center position with two different phase encoding directions, posterior-to-anterior (PA) and anterior-to-posterior (AP). In FIG. 6, the top row corresponds to the PA phase encoding, and the bottom row corresponds to the AP phase encoding. The images in the left column are acquired using the default ECC, i.e., the iso-center optimized ECC. The images in the middle column are acquired using off-center optimized gradient ECC together with default B0 ECC. The images in the right column are acquired using off-center optimized gradient ECC together with off-center optimized B0 ECC. For the right column, the off-center optimized B0 ECC accounts for the Oth order contribution induced at the off-center position by the gradient ECC, as discussed previously.

[0102] In FIG. 6, the left column provides a baseline corresponding to the default ECC, which is not optimized for off-center imaging. In the left column, a larger circle identifies a region of signal drop-out, and a smaller circle identifies a region of incomplete fat suppression. In the middle column, a circle identifies a region of incomplete fat suppression. Thus, relative to the default ECC baseline, the middle and right columns show reduced image artifacts at the off-center position, and the right column shows even further reduction of the artifacts relative to the middle column.

[0103] The difference between the middle and right columns of FIG. 6 reflects the difference between optimizing only the gradient ECC at the off-center position and optimizing both the gradient ECC and the B0 ECC at the off-center position. In the middle column, the gradient ECC is optimized for the off-center position, but the B0 ECC remainsthe default BO ECC. As discussed above with reference to FIG. 1, application of a 1st order gradient ECC at the off-center position can induce an additional Oth order BO offset at that position. In the right column, the BO ECC is also optimized at the off-center position, so as to compensate both the direct Oth order eddy current component and the Oth order contribution induced by the gradient ECC. This additional correction leads to the further reduction of image artifacts at the off-center position.

[0104] As shown in the example of FIG. 6, when the off-center optimized gradient and BO ECC is used, the images acquired with the two different phase-encoding directions become more similar to each other. This increased similarity indicates improved and accurate ECC effects.

[0105] Whereas FIG. 6 illustrates an example showing the effects of off-center optimized ECC determined from DFC measurements, FIG. 7 provides an example to show the effects of off-center optimized ECC determined from phantom data. The exemplary fat suppressed T2 FSE images of FIG. 7 are acquired at an off-center position using two different phase encoding directions, PA and AP. The left column shows images obtained using the regular iso-center optimized ECC, and the right column shows images obtained using off-center optimized ECC calculated from phantom data.

[0106] In FIG. 7, the left column provides a baseline corresponding to the regular iso-center optimized ECC. A larger circle in the left column identifies a region of signal drop-out. Smaller circles in the left column identify regions of incomplete fat suppression. Relative to the baseline, the right column shows reduced image artifacts at the off-center position.

[0107] FIG. 7 also shows that, compared with when the regular iso-center optimized ECC is used, the images obtained with the two different phase encoding directions are more similar to each other when the off-center optimized ECC calculated from phantom data is used. This increased similarity between the images obtained with the two different phase encodingdirections also indicates that the phantom-based off-center ECC provides improved and accurate eddy current correction at the off-center position.

[0108] In summary, in embodiments described above, the data used to determine the ECC model can be derived from measurements of a magnetic field sensor (e.g., a DFC), or from a spatiotemporally resolved MRI imaging sequence (e.g., through MRI imaging performed on a phantom). Regardless of the measurement approach employed, the ECC parameters can be determined according to a similar principle.

[0109] The 1st order ECC parameters can be determined such that the ECC gradient field cancels the eddy current gradient field at the target off-center position. The Oth order ECC parameters can be determined to account for two contributions: the direct Oth order B0 eddy current component at the target off-center position, and the additional B0 offset induced at that position by the application of the 1st order ECC itself. The final Oth order ECC is determined as the combination of the direct B0 correction and an additional correction for the induced B0 offset, such that the total residual magnetic field at the target off-center position approaches zero.

[0110] By explicitly accounting for the B0 offset induced by the 1st order ECC when determining the Oth order ECC parameters, the total residual magnetic field at the target off-center region can be substantially reduced. As a result, the accuracy of eddy current correction at the off-center position can be improved, thereby enabling improved image quality in off-center MRI applications, such as fat-suppressed shoulder imaging.

[0111] In the above description, the ECC forward model does not explicitly include eddy-current effects generated by application of the ECC itself. In further implementations, the ECC forward model can be extended to include such effects. In some cases, the ECC-induced eddy-current effects can include components of higher spatial order.

[0112] For example, a first pre-emphasized gradient waveform can be calculated using an ECC forward model that does not explicitly include ECC-induced eddy-current effects, together with a nominal gradient waveform. Then, the resulting pre-emphasized gradient waveform can be used, together with an ECC forward model that includes the ECC-induced eddy-current effects, to calculate a second pre-emphasized gradient waveform that accounts for those effects.

[0113] Although the above described examples use a long trapezoidal gradient lobe, the input gradient waveform is not limited thereto. Other input gradient waveforms can also be used, including, but not limited to, a triangular waveform, a chirp waveform, a spiral waveform, and an echo-planar imaging (EPI) waveform.

[0114] In further implementations, the eddy-current correction terms are not limited to Oth order and 1st order terms. For example, one or more higher order correction terms can also be used, e.g., by actuating one or more coils that generate magnetic field patterns of higher spatial order.

[0115] In further implementations, the ECC forward model can additionally include concomitant field contributions associated with the input gradient fields.

[0116] In further implementations, different sets of ECC parameters can be stored and selectively used for different sequence types. For example, the ECC parameters can include gains a and corresponding time constants T determined for the respective sequence types.

[0117] Similarly, different sets of ECC parameters can be stored and selectively used for different hardware configurations of the MRI system. These hardware configurations can include different receive coil arrays or other auxiliary devices, for example.

[0118] In further implementations, a Oth order frequency-shift correction can be applied by digitally shifting an acquisition band.

[0119] In further implementations, the ECC applied by the MRI system can be adjusted based on ambient temperature. For example, the gains a can be adjusted based on an input from a temperature sensor. Such temperature-based adjustment is not limited to the gains a, and can additionally or alternatively be applied to other ECC parameters used by the MRI system.

[0120] FIG. 8 illustrates a flowchart of an exemplary image reconstruction method in accordance with embodiments of the present disclosure. The method includes two stages, an offline stage 810 and an online stage 850.

[0121] During the offline stage 810, an ECC model can be determined for use in subsequent online processing. In step S815, spatiotemporal magnetic field information is acquired in response to an input gradient waveform. The spatiotemporal magnetic field information represents a magnetic field evolution in space and time in an off-center region. In step S820, model parameters defining the ECC model can be determined based on the spatiotemporal magnetic field information. In step S825, the determined model parameters can be stored for subsequent use in the online stage 850.

[0122] During the online stage 850, the stored model parameters can be used to determine ECC parameters to be applied during an imaging scan. In step S855, k-space data are acquired by scanning an imaging object with the MRI system while the ECC is applied. In this manner, the scan may be performed using ECC adapted to a target off-center region. In step S860, an image of the imaging object is reconstructed based on the acquired k-space data. Accordingly, the illustrated method can mitigate eddy current effects in the target off-center region, thereby reducing image artifacts associated with inaccurate ECC. This can improve fat suppression, reduce local signal drop-out, and provide improved image quality in off-center MRI.

[0123] Referring now to FIG. 9, a non-limiting example of a magnetic resonance imaging (MRI) system 100 is shown. The MRI system 100 depicted in FIG. 9 includes a gantry 101 (shown in a schematic cross-section) and various related system components 103 interfaced therewith. At least the gantry 101 is typically located in a shielded room. The MRI system geometry depicted in FIG. 9 includes a substantially coaxial cylindrical arrangement of the static field B0 magnet 111, a Gx, Gy, and Gz gradient coil set 113, and a large whole-body RF coil (WBC) assembly 115. Along a horizontal axis of this cylindrical array of elements, an imaging volume 117 is shown as substantially encompassing a portion of a patient 119 supported by a patient table 120. In the illustrated example, the imaging volume 117 is shown as encompassing the head of the patient 119. However, this is merely a non-limiting example, and the MRI system 100 can be used to image other anatomical regions, including, but not limited to, off-center anatomical regions such as a shoulder.

[0124] One or more smaller array RF coils 121 can be more closely coupled to the patient’s head an anatomical region of interest of the patient (referred to herein, for example, as “imaging object” or “object”) in the imaging volume 117. As those in the art will appreciate, compared to the WBC (whole-body coil), relatively small coils and / or arrays, such as surface coils or the like, are often customized for particular body parts (e.g., arms, shoulders, elbows, wrists, knees, legs, chest, spine, etc.). Such smaller RF coils are referred to herein as array coils (AC) or phased-array coils (PAC). These can include at least one coil configured to transmit RF signals into the imaging volume, and a plurality of receiver coils configured to receive RF signals from an object, such as the patient’s head, in the imaging volume.

[0125] The MRI system 100 includes an MRI system controller 130 that has input / output ports connected to a display 124, a keyboard 126, and a printer 128. As will be appreciated, the display 124 can be of the touch-screen variety so that it provides control inputs as well. A mouse or other I / O device(s) can also be provided.

[0126] The MRI system controller 130 interfaces with an MRI sequence controller 140, which, in turn, controls the Gx, Gy, and Gz gradient coil drivers 132, as well as the RF transmitter 134, and the transmit / receive switch 136 (if the same RF coil is used for both transmission and reception). The RF transmitter 134 may be composed of two or more transmitter channels for driving two or more RF transmit coils or ports on coils, as is used for RF shimming. The MRI sequence controller 140 includes suitable program code structure 138 for implementing MRI imaging (also known as nuclear magnetic resonance, or NMR, imaging) techniques including Bl field shimming. MRI sequence controller 140 can be configured for MR imaging with or without parallel imaging. Moreover, the MRI sequence controller 140 can facilitate one or more preparation scan (pre-scan) sequences, and a scan sequence to obtain a main scan magnetic resonance (MR) image (referred to as a diagnostic image). MR data from pre-scans can be used, for example, to determine shimming parameters forRF coils 115 and / or 121.

[0127] The MRI system components 103 include an RF receiver 141 providing input to data processor 142 so as to create processed image data, which is sent to display 124. The MRI data processor 142 is also configured to access previously generated MR data, images, navigator data, system configuration parameters 146, and / or program code structures 144 and 150.

[0128] In one embodiment, the MRI data processor 142 includes processing circuitry. The processing circuitry can include devices such as an application-specific integrated circuit (ASIC), configurable logic devices (e.g., simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs), and other circuit components that are arranged to perform the functions recited in the present disclosure, such as described with respect to FIGs. 1-19.

[0129] The processor 142 executes one or more sequences of one or more instructions contained in the program code structures 144 and 150. Alternatively, the instructions can be read from another computer-readable medium, such as a hard disk or a removable media drive. One or more processors in a multi-processing arrangement can also be employed to execute the sequences of instructions contained in the program code structures 144 and 150. In alternative embodiments, hard-wired circuitry can be used in place of or in combination with software instructions. Thus, the disclosed embodiments are not limited to any specific combination of hardware circuitry and software. For example, the program code structure 150 can store instructions that when executed perform the methods described herein.

[0130] Additionally, the term “computer-readable medium” as used herein refers to any non-transitory medium that participates in providing instructions to the processor 142 for execution. A computer readable medium can take many forms, including but not limited to, non-volatile media or volatile media. Non-volatile media includes, for example, optical, magnetic disks, and magneto-optical disks, or a removable media drive. Volatile media includes dynamic memory.

[0131] Also illustrated in FIG. 9 is a generalized depiction of an MRI system program storage (memory) 150, where stored program code structures such as instructions to perform the method 200 are stored in non-transitory computer-readable storage media accessible to the various data processing components of the MRI system 100. As those in the art will appreciate, the program store 150 can be segmented and directly connected, at least in part, to different ones of the system 103 processing computers having most immediate need for such stored program code structures in their normal operation (i.e., rather than being commonly stored and connected directly to the MRI system controller 130).

[0132] Additionally, the MRI system 100 as depicted in FIG. 9 can be utilized to practice exemplary embodiments described herein. The system components can be divided intodifferent logical collections of “boxes” and typically comprise numerous digital signal processors (DSP), microprocessors and special purpose processing circuits (e.g., for fast A / D conversions, fast Fourier transforming, array processing, etc.). Each of those processors is typically a clocked “state machine” wherein the physical data processing circuits progress from one physical state to another upon the occurrence of each clock cycle (or predetermined number of clock cycles).

[0133] Furthermore, not only does the physical state of the processing circuits (e.g., CPUs, registers, buffers, arithmetic units, etc.) progressively change from one clock cycle to another during the course of operation, the physical state of associated data storage media (e.g., bit storage sites in magnetic storage media) is transformed from one state to another during operation of such a system. For example, at the conclusion of an image reconstruction process and / or sometimes an image reconstruction map (e.g., coil sensitivity map, unfolding map, ghosting map, a distortion map etc.) generation process, an array of computer-readable accessible data value storage sites in physical storage media will be transformed from some prior state to a new state wherein the physical states at the physical sites of such an array vary between minimum and maximum values to represent real world physical events and conditions. As those in the art will appreciate, such arrays of stored data values represent and also constitute a physical structure, as does a particular structure of computer control program codes that, when sequentially loaded into instruction registers and executed by one or more CPUs of the MRI system 100, causes a particular sequence of operational states to occur and be transitioned through within the MRI system 100.

[0134] Embodiments of the present application may also be as set forth in the following parentheticals.

[0135] (1) A method for performing off-center eddy current correction (ECC) in a magnetic resonance imaging (MRI) system, the method comprising: acquiring k-space data byscanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; and reconstructing an image of the imaging object, based on the acquired k-space data, wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, and wherein the ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0136] (2) The method of (1), wherein the spatiotemporal magnetic field information is derived from measurements acquired using a magnetic field sensor during and / or after application of the input gradient waveform.

[0137] (3) The method of (1), wherein the spatiotemporal magnetic field information is derived from phase information of magnetic resonance data acquired from a phantom by performing a time-resolved gradient echo sequence after application of the input gradient waveform.

[0138] (4) The method of (3), wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first readout gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second readout gradient polarity, and wherein the first readout gradient polarity and the second readoutgradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

[0139] (5) The method of (3), wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first slice-selection gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second slice-selection gradient polarity, and wherein the first slice-selection gradient polarity and the second slice-selection gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

[0140] (6) The method of (3), wherein the spatiotemporal magnetic field information is derived from the phase information with reference to a static B0 map, so as to correct for effects of image distortions caused by local off-resonance.

[0141] (7) The method of (6), wherein the static B0 map is acquired in a separate gradient echo scan with at least two different echo times.

[0142] (8) The method of (1), wherein the spatiotemporal magnetic field information is represented by a set of coefficients that are associated with a set of harmonic basis functions and have a temporal dimension.

[0143] (9) The method of (1), wherein the model parameters that define the ECC model comprise a set of time constants and a set of spatial coefficients, and are optimized over a plurality of off-center positions within a selected region of interest (ROI) based on a loss function that assesses a quality of fit between the spatiotemporal magnetic field information and magnetic field information predicted by the ECC model.

[0144] (10) The method of (9), wherein the loss function comprises: a temporal norm, selected from an integral over time, a peak-to-peak amplitude over time, or a 2-norm over time, a spatial norm over the selected ROI, selected from an integral over space in theselected ROI, a peak-to-peak amplitude in the selected ROI, or a 2-norm over the selected ROI, or a spatiotemporal combination of the temporal norm and the spatial norm.

[0145] (11) The method of (1), wherein the model parameters that define the ECC model comprise a set of temporal basis functions shared across a plurality of regions of interest (ROIs) within the region comprising the off-center region, the selected imaging region corresponding to one of the plurality of ROIs, and wherein the ECC parameters determined at the selected imaging region comprise gains, for the selected imaging region, of the temporal basis functions.

[0146] (12) An apparatus for performing off-center eddy current correction (ECC) in a magnetic resonance imaging (MRI) system, the apparatus comprising: processing circuitry configured to acquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system, and reconstruct an image of the imaging object, based on the acquired k-space data, wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, and wherein the ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0147] (13) The apparatus of (12), wherein the spatiotemporal magnetic field information is derived from measurements acquired using a magnetic field sensor during and / or after application of the input gradient waveform.

[0148] (14) The apparatus of (12), wherein the spatiotemporal magnetic field information is derived from phase information of magnetic resonance data acquired from a phantom by performing a time-resolved gradient echo sequence after application of the input gradient waveform.

[0149] (15) The apparatus of (14), wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first readout gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second readout gradient polarity, and wherein the first readout gradient polarity and the second readout gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

[0150] (16) The apparatus of (14), wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first slice-selection gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second slice-selection gradient polarity, and wherein the first slice-selection gradient polarity and the second slice-selection gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

[0151] (17) The apparatus of (14), wherein the spatiotemporal magnetic field information is derived from the phase information with reference to a static B0 map, so as to correct for effects of image distortions caused by local off-resonance.

[0152] (18) The apparatus of (17), wherein the static BO map is acquired in a separate gradient echo scan with at least two different echo times.

[0153] (19) The apparatus of (12), wherein the spatiotemporal magnetic field information is represented by a set of coefficients that are associated with a set of harmonic basis functions and have a temporal dimension.

[0154] (20) A magnetic resonance imaging (MRI) system, the system comprising: first processing circuitry, configured to acquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; and second processing circuitry, configured to reconstruct an image of the imaging object, based on the acquired k-space data, wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, and wherein the ECC model is established by: acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform, determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, and storing the model parameters for determining ECC parameters at a selected imaging region.

[0155] Numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.

Claims

CLAIMS1. A method for performing off-center eddy current correction (ECC) in a magnetic resonance imaging (MRI) system, the method comprising:acquiring k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; andreconstructing an image of the imaging object, based on the acquired k-space data, wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, andwherein the ECC model is established by:acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform,determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, andstoring the model parameters for determining ECC parameters at a selected imaging region.

2. The method of claim 1, wherein the spatiotemporal magnetic field information is derived from measurements acquired using a magnetic field sensor during and / or after application of the input gradient waveform.

3. The method of claim 1, wherein the spatiotemporal magnetic field information is derived from phase information of magnetic resonance data acquired from a phantom by performing a time-resolved gradient echo sequence after application of the input gradient waveform.

4. The method of claim 3, wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first readout gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second readout gradient polarity, andwherein the first readout gradient polarity and the second readout gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

5. The method of claim 3, wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first slice-selection gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second slice-selection gradient polarity, andwherein the first slice-selection gradient polarity and the second slice-selection gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

6. The method of claim 3, wherein the spatiotemporal magnetic field information is derived from the phase information with reference to a static BO map, so as to correct for effects of image distortions caused by local off-resonance.

7. The method of claim 6, wherein the static BO map is acquired in a separate gradient echo scan with at least two different echo times.

8. The method of claim 1, wherein the spatiotemporal magnetic field information is represented by a set of coefficients that are associated with a set of harmonic basis functions and have a temporal dimension.

9. The method of claim 1, wherein the model parameters that define the ECC model comprise a set of time constants and a set of spatial coefficients, and are optimized over a plurality of off-center positions within a selected region of interest (ROI) based on a loss function that assesses a quality of fit between the spatiotemporal magnetic field information and magnetic field information predicted by the ECC model.

10. The method of claim 9, wherein the loss function comprises:a temporal norm, selected from an integral over time, a peak-to-peak amplitude over time, or a 2-norm over time,a spatial norm over the selected ROI, selected from an integral over space in the selected ROI, a peak-to-peak amplitude in the selected ROI, or a 2-norm over the selected ROI, ora spatiotemporal combination of the temporal norm and the spatial norm.

11. The method of claim 1, wherein the model parameters that define the ECC model comprise a set of temporal basis functions shared across a plurality of regions of interest (ROIs) within the region comprising the off-center region, the selected imaging region corresponding to one of the plurality of ROIs, and wherein the ECC parameters determined at the selected imaging region comprise gains, for the selected imaging region, of the temporal basis functions.

12. An apparatus for performing off-center eddy current correction (ECC) in a magnetic resonance imaging (MRI) system, the apparatus comprising:processing circuitry configured toacquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system, andreconstruct an image of the imaging object, based on the acquired k-space data,wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, andwherein the ECC model is established by:acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform,determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, andstoring the model parameters for determining ECC parameters at a selected imaging region.

13. The apparatus of claim 12, wherein the spatiotemporal magnetic field information is derived from measurements acquired using a magnetic field sensor during and / or after application of the input gradient waveform.

14. The apparatus of claim 12, wherein the spatiotemporal magnetic field information is derived from phase information of magnetic resonance data acquired from a phantom by performing a time-resolved gradient echo sequence after application of the input gradient waveform.

15. The apparatus of claim 14, wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a first readout gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second readout gradient polarity, andwherein the first readout gradient polarity and the second readout gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

16. The apparatus of claim 14, wherein the spatiotemporal magnetic field information is derived based on first magnetic resonance data acquired by performing the time-resolvedgradient echo sequence with a first slice-selection gradient polarity and second magnetic resonance data acquired by performing the time-resolved gradient echo sequence with a second slice-selection gradient polarity, andwherein the first slice-selection gradient polarity and the second slice-selection gradient polarity are reversed relative to each other, so as to correct for effects of image distortions caused by local off-resonance.

17. The apparatus of claim 14, wherein the spatiotemporal magnetic field information is derived from the phase information with reference to a static BO map, so as to correct for effects of image distortions caused by local off-resonance.

18. The apparatus of claim 17, wherein the static BO map is acquired in a separate gradient echo scan with at least two different echo times.

19. The apparatus of claim 12, wherein the spatiotemporal magnetic field information is represented by a set of coefficients that are associated with a set of harmonic basis functions and have a temporal dimension.

20. A magnetic resonance imaging (MRI) system, the system comprising:first processing circuitry, configured to acquire k-space data by scanning an imaging object using the MRI system while ECC is applied, wherein a target imaging region of the imaging object is positioned off-center relative to an isocenter of the MRI system; and second processing circuitry, configured to reconstruct an image of the imaging object, based on the acquired k-space data,wherein the ECC is applied based on an established ECC model, and comprises a zero-order ECC term and a first-order ECC term that correct effects of eddy current fields having spatial components of an order higher than the first order, and the zero-order ECC term is determined by taking into account a zero-order component induced by the first-order ECC term, andwherein the ECC model is established by:acquiring spatiotemporal magnetic field information representing a magnetic field evolution in space and time in a region comprising an off-center region in response to an input gradient waveform,determining, based on the acquired spatiotemporal magnetic field information, model parameters that define the ECC model, andstoring the model parameters for determining ECC parameters at a selected imaging region.