WO2013025487A1 - Spin echo sequences for diffusion weighted imaging of moving media - Google Patents

Abstract

Magnetic resonance imaging methods using application of single-sided bipolar pulsed gradient fields to specimens produce diffusion weightings with reduced sensitivity to bulk specimen motion. Single-sided pulse sequences can be arranged to produce a variety of diffusion weightings including trace weightings to produce images corresponding to average diffusion coefficients.

Description

SPIN ECHO SEQUENCES FOR DIFFUSION WEIGHTED IMAGING OF

MOVING MEDIA

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Patent Application 61/523,108, filed August 12, 2011 which is incorporated by reference herein.

FIELD

The disclosure pertains to pulse sequences for MRI of moving media. BACKGROUND AND SUMMARY

Motion artifacts are a serious confound for in vivo phase and amplitude MRI studies. In phase MRI, motion causes phase offsets and phase wrap-around, distorting measured displacement profiles and velocity maps. In amplitude MRI, velocity distributions and shearing motion lead to signal loss. For example, in diffusion MRI, velocity shear within a voxel causes signal attenuation that appears like diffusion (pseudo-diffusion) but is not caused by Brownian motion [16].

Diffusion MRI in tissues and organs is especially problematic. For example, the brainstem and spinal cord move significantly during the cardiac cycle, and are particularly troublesome to image. Other methods have been developed to try to freeze tissue motion, but these are not entirely successful, suffering from susceptibility and other problems that cause signal dropout. Although stunning diffusion tensor image (DTI) data has been collected for fixed human and animal hearts

[12, 27, 7, 10, 26, 21, 28], DTI in the beating heart remains an elusive goal. Whole body diffusion imaging is becoming increasingly important with the recognition that one can detect and possibly stage tumors using diffusion MRI [19] . However, significant motion in the abdomen and gut can hamper the interpretation of diffusion weighted image (DWI) data in these soft tissues.

There are a number of promising high angular resolution diffusion imaging (HARDI) sequences that can provide additional information about tissue microstructure and microarchitecture based on acquiring a large number of DWIs. In these methods, there is a presumption that the tissue occupying each voxel is the same in each DWI, an assumption that can seldom be satisfied with existing DWI acquisition methods. This is one reason why HARDI based methods have been applied largely to relatively low motion organs and tissue, such as the brain, and not more generally, particularly to moving organs such as the beating heart. Representative methods of imaging disclosed herein comprise applying a single sided bipolar pulsed gradient magnetic field to a specimen so as to produce diffusion weighted specimen magnetization. Based on the single sided bipolar pulsed gradient magnetic field, a diffusion weighted image of the specimen is formed. In some examples, the single sided bipolar pulsed gradient magnetic field is a balanced pulsed magnetic field or a symmetric pulsed magnetic field. In some examples, the bipolar pulsed gradient magnetic field has a duration that is less than about 1, 10, 20, or 50 ms. In further examples, a duration of the single sided bipolar pulsed gradient magnetic field is selected based on bulk specimen motion so as to reduce contributions to the diffusion weighted image from the bulk specimen motion. In further examples, a series of single sided bipolar pulsed gradient magnetic fields are applied to the specimen so as to produce a series of diffusion weighted specimen magnetizations. Based on the series of single sided bipolar pulsed gradient magnetic fields, a series of diffusion weighted images of the specimen is formed. In other examples, images are displayed based on application of one single sided pulsed gradient. In some embodiments, the diffusion weighted images are combined so as to form a combined image, and the combined image is displayed. In some examples, the single sided bipolar pulsed gradient magnetic field is a trace weighted gradient magnetic field, and the diffusion weighted image is a specimen trace map. In representative examples, the specimen is in vivo heart, brainstem, spinal cord, liver, and organs such as kidney tissue, or fetal organ tissue. Computer readable medium having computer-executable instructions are provided for performing any of the disclosed methods.

Additional disclosed magnetic resonance methods comprise establishing a longitudinal magnetization in a specimen, and applying a 90 degree radio-frequency pulse to the specimen. A time TE/2 is allowed to elapse and a refocusing pulse is applied to the specimen. A specimen magnetization is detected at a time TE/2 after the application of the refocusing pulse so as to obtain detected data values. A bipolar gradient pulse is applied between only one of the 90 degree pulse and the refocusing pulse or the refocusing pulse and the detection of the specimen magnetization. In particular examples, the detected data values are processed to determine at least one of an apparent diffusion constant or a trace of a diffusion tensor. In other examples, the bipolar gradient pulse is applied between the 90 degree pulse and the refocusing pulse or between the 90 degree pulse and the detection of the specimen magnetization. In some examples, the bipolar gradient pulse is a balanced symmetric bipolar gradient pulse. In some applications, an image based on the detected signal is produced and displayed. Computer readable medium comprise computer executable instructions for any of these methods.

Magnetic resonance imaging apparatus comprise a sequencer configured to apply single sided bipolar pulsed gradient (SS PG) magnetic fields to a specimen. A signal processor is configured to receive a detected signal based on the applied SS PG fields and produce a diffusion based specimen image. In some examples, the diffusion based specimen image is an apparent diffusion coefficient map. In other examples, the signal processor is configured to establish a trace weighted specimen map. In some

embodiments, the sequencer is configured to apply a single sided isotropic trace weighted pulsed gradient field.

These and other features of the disclosed technology are set forth below with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a Stejskal-Tanner sequence for diffusion weighted imaging.

FIG. 2a illustrates a single bipolar gradient pulse sequence.

FIG. 2b illustrates a pulse sequence that includes two bipolar diffusion sensitizing gradients.

FIG. 2c illustrates a pulse sequence associated with reversed bipolar diffusion sensitizing gradients.

FIGS. 3a-3b illustrate single-sided bipolar sequences applied before and after refocusing 180° RF pulses, respectively.

FIG. 3c is schematic diagram of a representative magnetic resonance apparatus configured to apply single sided bipolar pulsed gradients.

FIG. 3d is a block diagram of a representative imaging method using single sided bipolar gradient pulses.

FIGS. 4a-4b illustrate alternative single-sided bipolar pulse sequences for Trace weighted imaging.

FIG. 5 illustrates a stepper motor/controller on a Bruker 7T imaging system. FIG. 6 illustrates a touch screen control module for the stepper motor controller of FIG. 5.

FIG. 7 illustrates an experimental arrangement for acquiring diffusion weighted data in the presence of specimen motion.

FIGS. 8a-8c are A(0), FA, and Trace (Tr) x lO^-3 mm²/sec maps of the stationary excised pig spinal cord sample obtained with pulsed gradient spin echo (PGSE), conventional bipolar spin echo (BPSE), and single-sided bipolar spin echo (SS-BPSE) sequences, respectively.

FIGS. 9a-9c are A(0), FA, and Tr x lO^-3 mm²/sec maps of the stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and, SS-BPSE sequences, respectively. FIGS. lOa-lOc are diffusion weighted images of stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively.

FIGS, l la-llc are diffusion-weighted images of the rotating excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively.

FIGS. 12a- 12c are color maps of the stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively.

FIGS. 13a- 13c are color maps of the rotating excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively.

FIGS. 14a- 14b show clusters of homogeneous tissue obtained from the measured DTI field maps obtained with PGSE and SS-BPSE sequences, respectively, for the stationary sample.

FIGS. 15a- 15b show clusters of homogeneous tissue obtained from the measured DTI field maps obtained with PGSE and SS-BPSE sequences, respectively, for the rotating sample.

DETAILED DESCRIPTION

As used in this application and in the claims, the singular forms a, an, and the include the plural forms unless the context clearly dictates otherwise. Additionally, the term includes means comprises.

The described systems, apparatus, and methods described herein should not be construed as limiting in any way. Instead, the present disclosure is directed toward all novel and non-obvious features and aspects of the various disclosed embodiments, alone and in various combinations and sub-combinations with one another. The disclosed systems, methods, and apparatus are not limited to any specific aspect or feature or combinations thereof, nor do the disclosed systems, methods, and apparatus require that any one or more specific advantages be present or problems be solved.

Although the operations of some of the disclosed methods are described in a particular, sequential order for convenient presentation, it should be understood that this manner of description encompasses rearrangement, unless a particular ordering is required by specific language set forth below. For example, operations described sequentially may in some cases be rearranged or performed concurrently. Moreover, for the sake of simplicity, the attached figures may not show the various ways in which the disclosed systems, methods, and apparatus can be used in conjunction with other systems, methods, and apparatus. Additionally, the description sometimes uses terms like produce and provide to describe the disclosed methods. These terms are high-level abstractions of the actual operations that are performed. The actual operations that correspond to these terms will vary depending on the particular implementation and are readily discernible by one of ordinary skill in the art.

Theories of operation, scientific principles, or other theoretical descriptions presented herein in reference to the apparatus or methods of this disclosure have been provided for the purposes of better understanding and are not intended to be limiting in scope. The apparatus and methods in the appended claims are not limited to those apparatus and methods which function in the manner described by such theories of operation.

In the following, representative methods and apparatus for diffusion weighting imaging (DWI) of moving specimens are described, typically with reference to particular tissues and organs of interest. It will be appreciated that these are examples only, and other tissues and organs can be similarly evaluated. In addition, both in vivo and in vitro specimens can be imaged, including specimens that do or do not exhibit specimen motion. While the disclosed methods and apparatus are especially targeted to provide DWI sequences that exhibit reduced motion sensitivity, the disclosed methods and apparatus are broadly applicable.

In some disclosed examples, one or more series of magnetic field pulses are applied and one or more corresponding images displayed on a computer display or as a paper or other copy. However, for convenient description, image refers to processed detected signals responsive to applied magnetic field pulses, and suitable for providing a displayed image using a conventional image viewer capable of displaying JPEG, TIFF, bitmap, or other types of image data. A displayed image is a viewable image suitable for viewing by a user on, for example, a computer display or as a printed copy. In addition, an acquired magnetic resonance (MR) signal refers to captured signal data responsive to one or more or a series of applied magnetic field pulses. Such MR signals can be retained as stored data, or can be processed so as to form image data for production of a displayed image. Data processing can be performed in conjunction with MR signal detection and acquisition so as to produce images, or image processing can be performed later either locally with image acquisition or at a remote location. Signal acquisition and processing can be performed locally or via a LAN or WAN using a computer, a general purpose or dedicated processor, or other processing system that is configured to execute suitable computer-executable instructions that can be stored in RAM, ROM, on a CD or DVD, or stored in one or more devices such as hard disk drives or flash drives.

Generally, it has been difficult to compare and contrast different motion control strategies in MRI and be able to reliably and reproducibly produce different motion artifacts and assess their relative severity. To that end, disclosed herein is a controllable MRI phantom that can exhibit a range of complex motions to test the susceptibility of MRI sequences to various motion artifacts and evaluate the efficacy of different correction strategies to mitigate them.

Stejskal-Tanner DWI sequences

One method of sensitizing magnitude MRI data to the effects of water diffusion is by incorporating the Stejskal-Tanner pulsed gradient NMR sequence into a spin-echo (PGSE) MRI sequence [23, 24, 13, 3] . Specifically, the spin echo is formed by applying a 90° RF pulse followed by a 180° RF pulse. Diffusion weighting is obtained by applying a pair of identical unipolar gradient pulses around the slice selective 180° RF pulse as shown in FIG. 1. These unipolar diffusion-sensitizing gradients produce intravoxel dephasing resulting in signal attenuation in tissues as described in Eq 1. Signal attenuation caused by phase dispersion from the diffusion or random motion of incoherently moving spins enables estimation of the water diffusivity in each voxel of tissues as:

S(b) = S(0)e-^bD, (1) wherein S(b) is the observed signal, S(0) is a signal in the absence of the

diffusion-sensitizing gradients, and b is given by:

'2 JT2 δ

b = G²5 A (2)

3 wherein G is the magnitude of the diffusion gradient pulse with duration δ, A is a diffusion time, and is gyromagnetic ratio [30] . For the pulse sequence of FIG. 1, the diffusion time Δ is a time between application of gradient pulses.

Coherently moving spins such as associated with bulk motion with a uniform velocity produce a constant phase shift in the Stejskal-Tanner PGSE sequence:

S(a)_v = S(a)e^~*² , (3) wherein σ is a function of the gradient strength G, the diffusion timing parameters δ and Δ, spin velocity v, and is calculated according to: σ = ^GSAv. (4)

As shown in Eq 4, as the diffusion time Δ increases, it is less likely for the same family of spins tagged by the first diffusion-sensitizing gradient to be refocused by the second gradient, thus introducing a constant phase shift in the phase-encode direction. Motion correction strategies for DWI in moving media

A number of techniques can minimize signal attenuation due to bulk motion during DWI acquisition. A gradient echo MRI sequence such as illustrated in FIG. 2a can be considered as a special case of the Stejskal-Tanner pulse sequence, and consists of a single bipolar gradient block applied after the 90° RF pulse, wherein an interval between diffusion-sensitizing gradients Δ is set to δ. Since the duration of the

diffusion-sensitizing gradients is typically shorter than the diffusion time (i < Δ), the effect of coherently moving spins is reduced in comparison to conventional PGSE sequences. The 5-value for a single bipolar gradient is given in Eq 5: b = H₇W, (5) and the velocity sensitive phase variable in Eq. 3 reduces to: σ = Οδ²ν. (6)

However, despite a shorter echo time and the reduced effect of the diffusion time, this technique is highly sensitive to magnetic field inhomogeneity. Thus, it suffers from a significant signal loss due to shorter T₂ relaxation time.

Bipolar sequences other than the single bipolar sequence shown in FIG. 2a can be used. Hong and Dixon [11] demonstrated how replacing the unipolar diffusion-sensitizing gradients in the conventional Stejskal-Tanner sequence with two bipolar

diffusion-sensitizing gradient blocks as shown in FIG. 2b can decrease spin dephasing due to magnetic field inhomogeneity, as well as reduce sensitivity to bulk motion. The 5-value for the two consecutive bipolar diffusion-sensitizing gradient blocks is:

Although, the bipolar diffusion-sensitizing (BP) sequence appears to be useful for compensating for coherent spin motion, in practice it is impossible to reduce the diffusion times between two consecutive bipolar gradients to δ due to the duration of the 180° RF pulse, as well as the duration of slice selective crusher pulses before and after the RF pulse. Thus, this sequence suffers from a prolonged diffusion time between two consecutive bipolar gradients and the possibility of specimen movement between the application of the two diffusion gradient pulse blocks.

A pulse sequence based on reversed bipolar-sensitizing gradients [11] as shown in Fig. 2c is similar to the PGSE sequence since it is not cross-term free. Hong and Dixon [11] showed that in the presence of cross-terms, the error in the diffusion coefficient calculations is increased.

Cardiac and respiratory triggering techniques [5, 25] are relatively successful for suppressing artifacts arising from periodic bulk motion such as cardiac and cerebrospinal fluid pulsation and respiratory motion. However, triggering is not as effective for diffusion- weighted imaging of cardiac tissue itself and/or spinal cord, due to the uncertainty of the organ's position after each cycle.

Other approaches are based on navigator echo correction [18, 1], where phase correction is performed during post-processing. Such approaches are not efficient for cardiac and spinal imaging due to a number of assumptions about tissue location during acquisition. Thus, the most efficient and logical way to reduce the influence of coherently moving spins is during acquisition itself.

Single-sided bipolar gradient spin echo DWI sequences

In contrast to conventional approaches, disclosed herein are DWI sequences based on single-sided bipolar diffusion-sensitizing gradient as shown in FIGS. 3a-3b, rather than one on each side of a 180° RF pulse such as shown in FIGS. 1 and 2b-2c. The approach disclosed herein tends to reduce or minimize the effects of the magnetic field

inhomogeneity observed in application of a gradient echo DWI sequence, and tends to reduce the influence of the coherently moving spins by shortening the time between consecutive diffusion-sensitizing gradients. Although commonly used bipolar diffusion weighting (DW) sequences such as those of FIGS. 2a-2c have effective diffusion times Δ that are equal to the duration of diffusion-sensitizing pulse (i.e., δ) , the time between consecutive bipolar diffusion-sensitizing gradients in the moving media can result in motion artifacts. Since the diffusion-sensitizing gradients are the leading source of phase dispersion, it is preferable to apply them to the same population of spins. Single-sided bipolar spin echo (SS-BPSE) DWI sequences such as those of FIGS. 3a-3b are based on reduced or minimum diffusion times, and tend to exhibit reducing signal attenuation due to T₂ relaxation.

Typically, single bipolar diffusion-sensitizing gradients are balanced in that the integral of gradient magnitude over the gradient durations is approximately zero, or less than about 10%, 5%, or 1% of the product of peak gradient magnitude and total gradient duration. Thus, positive and negative gradient pulse portions substantially balance. It is convenient to provide symmetric diffusion sensitizing gradient pulses in which the positive and negative portions are substantially the same, absent the gradient direction. Such diffusion sensitizing gradient pulses are referred to herein as symmetric.

As in the case of conventional gradient echo sequences, 5-values can be calculated according to Eq 5. For other than the rectangular shaped diffusion-sensitizing gradients, the calculations of 5-values can be modified according to the shape of the pulses.

Although, applying a single bipolar gradient pulse block significantly reduces diffusion weighting in comparison with conventional "two-sided" bipolar or unipolar

diffusion-sensitizing sequences, current clinical gradient hardware can generate adequate 5-values for cardiac DWI ( b f» 400 s/mm²) [28] and spinal cord DWI ( b f» 500 s/mm²) [22] .

FIG. 3c is a schematic illustration of a representative apparatus 300 configured to control application of the disclosed single sided bipolar pulse (SS-BP) sequences, detection of the associated signals, and process the detected signals to provide images or image data. As shown in FIG. 3c, the MR apparatus includes a personal computer 301 or other computing device such as a laptop, workstation, or tablet computer configured to select one or more SS-BP sequences for the acquisition of diffusion based images. The computer 301 can provide a user interface for controlling data acquisition, analysis, and storage. A sequence/analyzer 302 is coupled to the computer 301 and is configured to establish suitable SS-BP sequences including pulse duration, pulse strength, and pulse orientation. The sequencer 302 is coupled to an RF generator 304 that can produce RF pulses that are coupled into specimen by an RF transmit coil 305. An RF receiver 306 is coupled to detect signals from the specimen via an RF receive coil 307, and a gradient controller 308 is configured to apply gradient magnetic fields to the specimen with a plurality of gradient coils 309. An axial magnetic field controller is coupled to one or more axial magnet coils 311. The sequencer 302 is configured to apply the selected single sided pulsed gradients and to process the received data to determine specimen properties of interest such as providing an apparent diffusion coefficient (ADC) map. Detected MR signals can be processed with computer executable instructions stored in one or more computer readable media for execution in a dedicated processor, in the sequencer 302, or at the computer 301.

A representative imaging method 330 is illustrated in FIG. 3d. At 332, a longitudinal magnetization is established in a specimen and at 334, a 90 degree radio-frequency (RF) pulse is applied. At 336, selection of a time for application of a single sided bipolar gradient pulse (SS BP) is determined so as to be either before or after application of a refocusing pulse. If selected to be prior to the refocusing pulse, the single sided bipolar gradient pulse (SS BP) is applied at 338. After a time TE/2 from application of the 90 degree pulse at 334, a refocusing pulse is applied at 340. If the SS BG pulse is to be applied after the refocusing pulse, an SS BG pulse is applied at 342. An associated signal S(G) is detected at 344, and at 336 it is determined if additional SS BP pulses are to be applied. If so, the method returns to 402. If acquisition is complete, at 350 an ADC map or other specimen image or characterization is produced based on S(G). As noted above, a variety of images can be produced including ADC maps, Trace maps, and diffusion tensor images.

Single-sided bipolar spin echoes for isotropically or trace-weighted DWIs

Isotropically weighted DWIs [14, 31] provide an attractive means for obtaining a diffusion Trace map, a mean Apparent Diffusion Coefficient (ADC) map, or a

Trace-weighted MRI using a single shot DWI acquisition, wherein the Trace of the diffusion tensor is given by Tr(D)= (D_xx + D_yy + D_xx) . The Trace or the mean ADC (1/3 of the Trace value) is an intrinsic property of tissue [2] that can change significantly due to development [15] , disease, such as stroke [29] , and degeneration and aging [20, 17] . Conventional diffusion weighting pulse sequences generally are unable provide reliable specimen Trace maps in rapidly moving media, like the heart, as well as maps of other DTI parameters. In media undergoing irregular, irreproducible or aperiodic motion, it is problematic to assemble a Trace map or Trace-weighted DWI using several different DWIs obtained at different time points. In such cases, tissue has generally rotated, shifted or deformed between acquisitions, rendering Trace measurements unreliable.

Wong showed that Trace-weighted DWIs can be generated by applying diffusion-sensitizing gradients in patterns that satisfy the following orthogonality conditions on the gradient waveforms:

This expression results in the off-diagonal elements of the b-matrix vanishing. Another requirement of obtaining Trace-weighted DWIs is that the diagonal elements of the b-matrix are all equal, i.e., b_xx = b_yy = b_zz, Then the signal attenuation can be written as:

Isotropic or trace-weighted DWIs have been obtained using conventional DWI schemes described above by applying a series of different gradient patterns with different diffusion-sensitizing gradient directions on either side of the refocusing 180° RF pulse [14, 31] . FIG. 4a shows an example of diffusion-sensitizing gradient pattern (described in [14] ) placed before refocusing 180° RF pulse. However, this pattern suffers from inefficient use of the gradients. A way to overcome this limitation, is to apply

diffusion-sensitizing gradient along all axes simultaneously, provided that the

off-diagonal elements are canceled by alternating polarities of the bipolar gradients in Eq 8. Conturo et al. [6] proposed obtaining a single trace weighted image to by playing out a combination of gradients in a set of four separate DWI acquisitions, and multiplying them together. Their particular tetrahedral gradient pattern was quite efficient, producing the maximum diffusion attenuation possible. Using the single-sided sequences disclosed above, a tetrahedral gradient pattern can be based on a single one-sided bipolar block within a single DWI, as shown in FIG. 4a. Such a single-sided gradient application permits efficient trace weighting with reduced bulk motion artifacts.

Similarly, a number of simultaneously applied gradient patterns (e.g., patterns presented in[14, 31]) can be played out before or after refocusing 180° RF pulse in order to minimize attenuation due to T₂ relaxation, but it is important to note that one does need to apply such gradient pulses on both sides of the refocusing 180° RF pulse, which had been assumed previously. Thus, in contrast to conventional approaches, single sided application of gradient patterns can produce satisfactory trace weighted images with reduced motion artifacts.

Representative imaging demonstrations

In order to demonstrate motion suppression in DWI using the disclosed sequences, a Bruker Rheo-NMR[4] unit was modified. This unit comprises an MR compatible rotating shaft driven by an integrated stepper motor/controller unit that can impart a continuous angular motion to a "cone and plate" or Couette flow cell within a Micro2.5 microscopy probe (25mm solenoid coil). The motor and controller unit were replaced as shown in FIG. 5 so that the shaft and fixture can exhibit arbitrarily complex, jerky motions, like those seen in vivo. A custom touch screen control module (FIG. 6) allows a user to prescribe arbitrary shaft rotation waveforms based on a trigger input, including motionless periods to correspond to those that occur in specimens, such as in the heart during end-diastole. In addition, motion can be triggered to MRI sequence application. Fixtures were provide to hold different tissues to be scanned as shown in FIG. 6. The specimen holder allows for a tissue plug to be inserted. Susceptibility effects can be minimized by potting the specimen in perffuoropolyethers such as Fomblin perffuoropolyethers. The fixture itself is constructed from ULTEM, a susceptibility matched plastic.

Excised pig spinal cord DTI imaging

The disclosed single-sided bipolar spin echo pulsed field gradient (SS-BPSE PFG) sequences were applied to representative moving media to demonstrate artifact reduction in comparison with conventional PGSE and BP sequences. DWI data were obtained from the same excised pig spinal cord specimen fixed with a 4% paraformaldehyde solution. Prior to DWI data collection, the pig spinal cord was washed in

phosphate-buffered saline (PBS) doped with Gd-DTPA. Gadolinium was used in order to decrease the Ti relaxation time of the spinal cord tissue. The sample was imaged in a modified " cone and place" cell from a RheoNMR cell kit filled with FOMBLIN perfluoropolyetfier oil, within a Micro2.5 microscopy probe (25 mm solenoid coil) with 1450 mT/m 3-axis gradients. The common parameters for PGSE, conventional BP, and SS-BPSE sequences include repetition time (TR) = 800ms, bandwidth = 100kHz, field-of-view (FOV) = 20 x 20mm, matrix = 64 x 64 with a one 1-mm thick axial slices. PGSE echo time (TE) was set to 26ms, while both BP and SS-BPSE had TE set to 37ms. One DWI per slice was acquired with b f» 0 s/mm², followed by acquisition of 21 DWIs with b = 1200 s/mm² without rotation and with rotation at 1 Hz and a 15 ms motor pause. The diffusion gradient duration δ was 8 ms and the gradient separation Δ was 10.2 ms for PGSE and 8 ms and an 8 ms gradient separation for conventional BP and SS-BPSE, respectively. The experimental arrangement is illustrated in FIG. 7.

At each voxel location in the raw image, the apparent diffusion tensor, D, was estimated from the acquired DWIs and tensor-derived parameters, such as the eigenvectors or principal directions, e_{1 ?} C2 , and £3 , and the corresponding eigenvalues or principal diffusivities, λ_{1 ?} λ2 , and X₃, were estimated. These were passed to

parsimonious model selection[9] and multivariate hypothesis testing clustering[8] algorithms but other signal processing methods can be used.

Comparison of SS-BPSE with PGSE and conventional BP

FIGS. 8a-8c show A(0), FA, and Tr x lO^-3 mm²/sec maps of the stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and, SS-BPSE sequences, respectively. As can be seen from FIGS. 8a-8c, image results were consistent for all three evaluated sequences for the stationary sample of the excised pig spinal cord, although signal-to-noise ratio (SNR) was better for PGSE sequences due to a shorter echo time. For this particular sample, the average fractional anisotropy FA in white matter was around 0.48, while trace of the diffusion tensor Tr was approximately

FIGS. 9a-9c show A(0), FA, and Tr x lO^-3 mm²/sec maps of the rotating excised pig spinal cord sample obtained with PGSE, conventional BP, and, SS-BPSE sequences, respectively. As is apparent, the resulting images differ from those of FIGS. 8a-8c.

However, for the rotating sample, results obtained with BP sequence (FIG. 3b) were quite different from the results obtained with PGSE and SS-BPSE (FIG. 3a and FIG. 3c, respectively), i.e., FA f» 0.71 and Tr f» 4.5 x 10^~3 mm²/sec for BP sequence, while for both PGSE and SS-BPSE FA « 0.49 and Tr « 1.3 x 10^"3 mm²/sec.

Although results for the rotating sample obtained with PGSE and SS-BPSE sequences appear to be similar, after further examination of the individual DWIs (the examples of the DWIs for one direction are given in FIGS. 8a-8c and FIGS. 9a-9c for stationary and rotating samples, respectively), motion artifacts are more pronounced in the PGSE images of FIG. 9a in comparison with the SS-BPSE images of FIG. 9c.

FIGS. lOa-lOc are DWIs of stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively. FIGS, lla-llc are DWIs of the rotating excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE sequences, respectively.

FIGS. 12a- 12c are Color Direction Encoded maps of the stationary excised pig spinal cord sample obtained with PGSE, conventional BP, and SS-BPSE, respectively and FIGS. 13a- 13c are corresponding direction encoded maps with rotating excised pig spinal cord. Color coded maps show that the results obtained with PGSE sequences (FIGS. 10a and 11a) and BP sequences (FIGS, lib and lib) are not consistent between stationary and rotating measurements. However, SS-BPSE sequences produced compatible results for both stationary and rotating experiments as shown in FIGS. 10c and 11c.

FIGS. 14a- 14b show clusters of homogeneous tissue obtained from the measured DTI field maps obtained with PGSE and SS-BPSE sequences, respectively, for the stationary sample. FIGS. 15a-15b show clusters of homogeneous tissue obtained from the measured DTI field maps obtained with PGSE and SS-BPSE sequences,

respectively, for the rotating sample. As it can be seen, both sequences produced consistent results in stationary white matter. However, in the rotating experiment, due to averaging caused by motion, clustering algorithm failed to distinguish white and gray matter in the spinal cord (see FIG. 15a). The SS-BPSE sequence successfully separated white and gray matter, however showed slight variations in the degree of homogeneity in white matter (different colors within white matter identify three separate clusters).

Additional Examples

In the presence of complex motion, SS-BPSE sequences outperform both PGSE and conventional BP sequences. Conventional BP sequences do not properly compensate for jerky motion. The advantages of SS-BPSE are especially pronounced as the resolution gets lower. However, PGSE can have a better SNR due to shorter TE and requires less diffusion gradient strength for achieving diffusion weighting comparable to that obtained with SS-BPSE and conventional BP.

The disclosed sequences can be used to perform diffusion MRI studies in the beating heart to, for example, assess cardiac muscle and obtain estimates of the mean ADC and other DTI derived parameters within the muscle tissue. The mean ADC could be calculated from the estimated diffusion tensor itself, from several ADCs obtained in an isotropically organized DWI acquisition, or by using isotropically weighted DWI sequences as disclosed herein. Data obtained in this manner can be used in identifying abnormal or ischemic areas of the heart.

DWI data with reduced motion artifacts based on the disclosed SS sequences can also be used in cancer screening, diagnosis, tumor staging and determining therapeutic effectiveness. Respiratory and cardiac cycles produce complex organ motions in the liver, kidney, prostate and other organs and this lack of rigid mechanical tethering makes internal organs susceptible to unavoidable movement and rearrangement, even if a subject is asked to remain still in a supine position within an MRI system magnet. The lack of reproducibility of tissue coordinates makes it problematic to estimate quantitative diffusion parameters based on conventional DWI sequences. DWI and ADC maps have been shown to be remarkably effective in detecting tumors from the surrounding normal tissue, and the application of the disclosed DWI and isotropically weighted DWI sequences permits reduction in artifacts associated with bulk organ and tissue motion that can be a confound in the radiological interpretation of tumors.

In some living tissues, there have been reports of intravoxel incoherent motion (IVIM), particularly in highly perfused tissues like kidney and liver. IVIM effects have been reported in b-value ranges of 0 to approximately 100 sec/mm². Some of the observed IVIM effect may be due to other types of complex organ and tissue motion and deformation other than the presumed capillary and extracellular fluid flow that has the appearance of Brownian motion. The disclosed methods can reduce such motion artifacts to aid in evaluation of IVIM perfusion effects. The disclosed methods can also be applied to the assessment of the spinal cord and the brainstem. Motion of these structures includes significant side to side "whip" of the spinal cord and the plunging motion of the brainstem. The disclosed methods based on SS sequences can reduce or eliminate the associated motion artifacts.

In another example, DTI and other diffusion MRI methods based on the disclosed sequences can be used for fetal assessment in utero as well as assessment of the uterine tissue surrounding the infant. Unpredictable fetal movement, as well as movement of the mother's uterus during a scan, can make conventional DWI acquisitions challenging. The ability to freeze the diffusive motion to a short window as disclosed herein can provide superior estimates of ADCs, and superior image quality. Once fetal position is known as well as the orientation of the diffusion gradients applied within the laboratory coordinate system, fetal images can be re-registered to a common template and gradients or b-matrices can be transformed to be able to obtain a mutually co-registered set of DWI volume data. Diffusion coefficients in the fetus tend to be higher than in adult organs, and closer to free water, making the demands on a diffusion gradient set less severe, since lower b-values are required for fetal DWI than in adult DWI. There are several embodiments of the disclosed DWI sequences that have advantages for particular applications. One could apply the diffusion spectroscopic SE sequence as a prefilter prior to the application of an imaging block. One could apply the diffusion gradient pulse within a convention SE sequence with slice select and phase encode gradients applied. For isotropically weighted DWIs the same principle holds. Once can apply an isotropically weighted spectroscopic SE sequence as a filter to an imaging block or incorporate the isotropically weighted SE sequence within an imaging block.

In general, there are a large family of DWI applications that do not use the ADC or DTI models to obtain useful information about tissue structure and architecture. High Angular Resolution Diffusion Imaging (HARDI) acquisitions arise in many higher order diffusion MRL Examples include Q-ball MRI, PAS MRI, DOT, GDTI etc. Other methods of analyzing DWI data that are not model specific, such as k- and q-space MRI, Diffusion Spectrum MRI (DSI), SHORE 3D etc., which attempt to measure the average propagator directly from the DWI data, can also employ the disclosed SS-based DWIs to improve data quality and fidelity. Such applications should become more feasible as gradient hardware improves, particularly as gradient strength and slew-rate increases.

There are clearly several potential embodiments of this DWI sequence that have advantages for particular applications. A diffusion spectroscopic SE NMR sequence can be applied as a prefilter prior to the application of an imaging (MRI) block. Diffusion gradient pulses can be applied within a conventional SE sequence with slice select and phase encode gradients applied. There are similar applications to isotropically weighted DWI. An isotropically weighted spectroscopic SE sequence can be applied as a filter to an imaging block or the isotropically weighted SE sequence can be incorporated within an imaging block.

The disclosed methods for obtaining DWIs can be applied to bipolar gradient acquisition in multiple wavevector or multiple pulsed gradient field (PFG) NMR and MRI measurements. In these applications at least two Stejskal- Tanner single PFG pulses are concatenated to produce a multiple PFG MR sequence. An example is a concatenation of two Stejskal- Tanner PFG sequences to produce a double PFG sequence. A single sided bipolar PFG sequence can be used to replace the two

Stejskal-Tanner gradient pulses to obtain diffusion weighting.

In other examples, the disclose single-side diffusion sensitizing gradients can be unbalanced so as to permit compensation or elimination of one or more image artifacts such as those associated with gradient artifacts or image background. In addition, such diffusion sensitizing gradients can be used with methods such as diffusion tensor imaging (DTI) or diffusion spectrum imaging (DSI) or other methods. A plurality of diffusion sensitized images can be obtained and registered with respect to each other or a common reference location or orientation. Such registered images provide self-consistent DWI data. Whereas the technology has been described in connection with several examples, it will be understood that the technology is not limited to these embodiments. On the contrary, the invention is intended to encompass all alternatives, modifications, and equivalents as may be included within the spirit and scope of the disclosure as defined by the appended claims.

REFERENCES

[1] A. W. Anderson and J. C. Gore. Analysis and correction of motion artifacts in diffusion weighted imaging. Magn Reson Med, 32(3):379-387, Sep 1994.

[2] P. J. Basser, J. Mattiello, and D. LeBihan. MR diffusion tensor spectroscopy and imaging. Biophys J, 66(l):259-267, Jan 1994.

[3] D. Le Bihan, E. Breton, D. Lallemand, P. Grenier, E. Cabanis, and

M. Laval-Jeantet. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology, 161 (2) :401-407, Nov 1986.

[4] PT Callaghan. Rheo-NMR: nuclear magnetic resonance and the rheology of

complex fluids. REPORTS ON PROGRESS IN PHYSICS, 62(4):599-670, APR 1999.

[5] D. Chien, R. B. Buxton, K. K. Kwong, T. J. Brady, and B. R. Rosen. MR diffusion imaging of the human brain. J Comput Assist Tomogr, 14(4):514-520, 1990.

[6] T. E. Conturo, R. C. McKinstry, E. Akbudak, and B. H. Robinson. Encoding of anisotropic diffusion with tetrahedral gradients: a general mathematical diffusion formalism and experimental results. Magn Reson Med, 35(3):399-412, Mar 1996.

[7] Jiangang Dou, Timothy G Reese, Wen-Yih I Tseng, and Van J Wedeen. Cardiac diffusion MRI without motion effects. Magn Reson Med, 48(1): 105-114, Jul 2002.

[8] Raisa Z Freidlin, Evren Ozarslan, Yaniv Assaf, Michal E Komlosh, and Peter J Basser. A multivariate hypothesis testing framework for tissue clustering and classification of DTI data. NMR Biomed, 22(7):716-729, Aug 2009.

[9] Raisa Z Freidlin, Evren Ozarslan, Michal E Komlosh, Lin-Ching Chang,

Cheng Guan Koay, Derek K Jones, and Peter J Basser. Parsimonious model selection for tissue segmentation and classification applications: a study using simulated and experimental DTI data. IEEE Trans Med Imaging,

26(11):1576-1584, Nov 2007.

[10] Patrick Helm, Mirza Faisal Beg, Michael I Miller, and Raimond L Winslow.

Measuring and mapping cardiac fiber and laminar architecture using diffusion tensor MR imaging. Ann N Y Acad Sci, 1047:296-307, Jun 2005. [11] Xiaole Hong and W. Thomas Dixon. Measuring diffusion in inhomogeneous systems in imaging mode using antisymmetric sensitizing gradients. Journal of Magnetic Resonance (1969), 99(3):561 - 570, 1992.

[12] E. W. Hsu, A. L. Muzikant, S. A. Matulevicius, R. C. Penland, and C. S.

Henriquez. Magnetic resonance myocardial fiber-orientation mapping with direct histological correlation. Am J Physiol, 274(5 Pt 2):H1627-H1634, May 1998.

[13] KD Merboldt, Hanicke W, and J Frahm. Self-Diffusion NMR Imaging Using

Stimulated Echoes. Journal of Magnetic Resonance, 64(3):479-486, 1985.

[14] S. Mori and P. C. van Zijl. Diffusion weighting by the trace of the diffusion tensor within a single scan. Magn Reson Med, 33(l):41-52, Jan 1995.

[15] J. Neil, J. Miller, P. Mukherjee, and P. S. Hppi. Diffusion tensor imaging of normal and injured developing human brain - a technical review. NMR Biomed,

15(7-8) :543-552, 2002.

[16] Uri Nevo, Evren Ozarslan, Michal E Komlosh, Cheng Guan Koay, Joelle E Sarlls, and Peter J Basser. A system and mathematical framework to model shear flow effects in biomedical DW-imaging and spectroscopy. NMR Biomed, 23(7):734-744, Aug 2010.

[17] A. O. Nusbaum, C. Y. Tang, M. S. Buchsbaum, T. C. Wei, and S. W. Atlas.

Regional and global changes in cerebral diffusion with normal aging. AJNR Am J Neuroradiol, 22(1): 136-142, Jan 2001.

[18] R. J. Ordidge, J. A. Helpern, Z. X. Qing, R. A. Knight, and V. Nagesh. Correction of motional artifacts in diffusion-weighted MR images using navigator echoes. Magn Reson Imaging, 12(3):455-460, 1994.

[19] Anwar R Padhani, Guoying Liu, Dow Mu Koh, Thomas L Chenevert, Harriet C Thoeny, Taro Takahara, Andrew Dzik-Jurasz, Brian D Ross, Marc Van Cauteren, David Collins, Dima A Hammoud, Gordon J S Rustin, Bachir Taouli, and Peter L Choyke. Diffusion-weighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia, 11 (2): 102-125, Feb 2009.

[20] A. Pfefferbaum, E. V. Sullivan, M. Hedehus, K. O. Lim, E. Adalsteinsson, and M. Moseley. Age-related decline in brain white matter anisotropy measured with spatially corrected echo-planar diffusion tensor imaging. Magn Reson Med,

44(2):259-268, Aug 2000. [21] T. G. Reese, R. M. Weisskoff, R. N. Smith, B. R. Rosen, R. E. Dinsmore, and V. J. Wedeen. Imaging myocardial fiber architecture in vivo with magnetic resonance. Magn Reson Med, 34(6):786-791, Dec 1995.

[22] J. Renoux, D. Facon, P. Fillard, I. Huynh, P. Lasjaunias, and D. Ducreux. MR

diffusion tensor imaging and fiber tracking in inflammatory diseases of the spinal cord. AJNR Am J Neuroradiol, 27(9):1947-1951, Oct 2006.

[23] E. O. Stejskal and J. E. Tanner. Spin diffusion measurements: Spin echoes in the presence of a time-dependent field gradient. J Chem Phys, 42(l):288-292, January 1966.

[24] D. G. Taylor and M. C. Bushell. The spatial mapping of translational diffusion coefficients by the NMR imaging technique. Phys Med Biol, 30 (4): 345-349, Apr 1985.

[25] C. Thomsen, O. Henriksen, and P. Ring. In vivo measurements of relaxation

process in the human liver by MRI. the role of respiratory gating/triggering. Magn Reson Imaging, 6(4):431-436, 1988.

[26] Nicolas Toussaint, Maxime Sermesant, Christian T Stoeck, Sebastian Kozerke, and Philip G Batchelor. In vivo human 3D cardiac fibre architecture: reconstruction using curvilinear interpolation of diffusion tensor images. Med Image Comput Comput Assist Interv, 13(Pt l):418-425, 2010.

[27] W. Y. Tseng, T. G. Reese, R. M. Weisskoff, and V. J. Wedeen. Cardiac diffusion tensor MRI in vivo without strain correction. Magn Reson Med, 42(2):393-403, Aug 1999.

[28] P. Boesiger U. Gamper and S. Kozerke. Diffusion imaging of the in vivo heart using spin echoesconsiderations on bulk motion sensitivity. Magn Reson Med, 57:331-337, 2007.

[29] A. M. Ulug, N. Beauchamp, R. N. Bryan, and P. C. van Zijl. Absolute quantitation of diffusion constants in human stroke. Stroke, 28(3):483-490, Mar 1997.

[30] G. E. Wesbey, M. E. Moseley, and R. L. Ehman. Translational molecular

self-diffusion in magnetic resonance imaging, i. effects on observed spin-spin relaxation. Invest Radiol, 19(6):484-490, 1984.

[31] E. C. Wong, R. W. Cox, and A. W. Song. Optimized isotropic diffusion weighting.

Magn Reson Med, 34(2): 139-143, Aug 1995.

Claims

What is claimed is:

1. A method of magnetic resonance imaging, comprising:

applying a single-sided bipolar pulsed gradient magnetic field to a specimen so as to produce a diffusion weighted specimen magnetization; and

based on the single-sided bipolar pulsed gradient magnetic field, forming a diffusion weighted image of the specimen.

2. The method of claim 1, wherein the single sided bipolar pulsed gradient magnetic field is a balanced pulsed gradient magnetic field.

3. The method of claim 2, wherein the single sided bipolar pulsed gradient magnetic field is a symmetric pulsed gradient magnetic field.

4. The method of claim 1, wherein the single sided bipolar pulsed gradient magnetic field has a duration that is less than about 10 ms.

5. The method of claim 1, further comprising selecting a duration of the single sided bipolar pulsed gradient magnetic field based on bulk specimen motion so as to reduce contributions to the diffusion weighted image associated with bulk specimen motion.

6. The method of claim 1, further comprising:

applying a series of single bipolar pulsed gradient magnetic fields to the specimen so as to produce a series of diffusion weighted specimen magnetizations; and

based on the series of single bipolar pulsed gradient magnetic fields, forming a series of diffusion weighted images of the specimen.

7. The method of claim 6, further comprising combining the series of diffusion weighted images so as to form a combined image, and displaying the combined image.

8. The method of claim 1, wherein the single sided bipolar pulsed gradient magnetic field is a trace weighted pulsed gradient magnetic field, and the computed image is a specimen trace map.

9. The method of claim 1, wherein the specimen is in vivo heart, brainstem, spinal cord, liver, or kidney tissue.

10. At least one computer readable medium have computer-executable instructions thereon for the method of any of claims 1-9.

11. A magnetic resonance method, comprising:

establishing a longitudinal magnetization in a specimen; applying a 90 degree radio-frequency pulse to the specimen;

allowing a time TE/2 to elapse;

applying a refocusing pulse to the specimen after a time TE/2 has elapsed from the application of the 90 degree pulse;

detecting specimen magnetization at a time TE/2 after the application of the refocusing pulse so as to obtain detected data values; and

applying a bipolar gradient pulse between only one of the 90 degree pulse and the refocusing pulse or the refocusing pulse and the detection of the specimen magnetization.

12. The method of claim 11, further comprising processing the detected data values to determine at least one of an apparent diffusion coefficient or a trace of a diffusion tensor.

13. The method of claim 11. wherein the bipolar gradient pulse is applied between the 90 degree pulse and the refocusing pulse.

14. The method of claim 11, wherein the bipolar gradient pulse is applied between the 90 degree pulse and the detection of the specimen magnetization.

15. The method of claim 11, wherein the bipolar gradient pulse is a balanced symmetric bipolar gradient pulse.

16. The method of claim 11, wherein the bipolar gradient pulse is a trace weighted gradient pulse sequence.

17. The method of claim 1, wherein the bipolar gradient pulse is an unbalanced bipolar gradient pulse.

18. The method of claim 17, further comprising selecting the unbalanced bipolar gradient pulse to compensate for at least one image artifact.

19. The method of claim 11, further comprising producing an image based on the detected data values.

20. At least one computer readable medium comprising computer executable instructions for the method of any of claims 11-19.

21. A magnetic resonance imaging apparatus, comprising:

a sequencer configured to apply single sided bipolar pulse gradient (SS PG) magnetic fields to a specimen; and

a signal processor configured to receive a detected signal based on the applied SS PG fields and produce a diffusion based specimen image.

22. The apparatus of claim 21, wherein the diffusion based specimen image is a an apparent diffusion coefficient map.

23. The apparatus of claim 21, wherein the signal processor is configured to establish a trace weighted specimen map.

24. The apparatus of claim 21, wherein the sequencer is configured to apply a single sided isotropic trace weighted pulsed gradient field.

25. The method of claim 1, wherein the diffusion weighted image is a a diffusion tensor image.

26. The method of claim 1, wherein the diffusion weighted image is a diffusion spectrum image.

27. A method, comprising:

acquiring a plurality of diffusion weighted images based on single-sided diffusion sensitizing gradients; and

registering the diffusion weighted images so as to produce s self-consistent diffusion weighted image data.

28. The method of claim 27, wherein the single-sided diffusion sensitizing gradients are bipolar gradients, balanced gradients, unbalanced gradients, or symmetric gradients.