EP4689692A1

EP4689692A1 - Systems and methods for patterned multi-slice excitation in magnetic resonance imaging

Info

Publication number: EP4689692A1
Application number: EP24725979.9A
Authority: EP
Inventors: Jozef Henricus DUYN; Peter Van Gelderen; Jacobus Adrianus De Zwart
Original assignee: US Department of Health and Human Services
Current assignee: US Department of Health and Human Services
Priority date: 2023-04-03
Filing date: 2024-04-03
Publication date: 2026-02-11
Also published as: WO2024211423A1

Abstract

Various methods and systems are provided for patterned multi-slice excitation (PME) magnetic resonance image (MRI). In one example, a method for imaging a patient according to a PME MRI slice acquisition protocol includes, during a first pulse interval, applying a first radiofrequency (RF) pulse together with a first slice selection gradient to excite a first slice and a second slice, the second slice following the first slice in the slice acquisition protocol, and acquiring MR signals of only the first slice; and during a second pulse interval following the first pulse interval, applying a second RF pulse together with a second slice selection gradient to excite the second slice and a third slice, the third slice following the second slice in the slice acquisition protocol, and acquiring MR signals of only the second slice.

Description

SYSTEMS AND METHODS FOR PATTERNED MULTI-SLICE EXCITATION IN

MAGNETIC RESONANCE IMAGING

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority to U.S. Provisional Application No. 63/493,962 entitled “SYSTEMS AND METHODS FOR PATTERNED MULTI-SLICE EXCITATION IN MAGNETIC RESONANCE IMAGING,” and filed April 03, 2023. The entire contents of the above-listed application is hereby incorporated by reference for all purposes.

TECHNICAL FIELD

[0002] Embodiments of the subject matter disclosed herein relate to medical imaging, and more specifically to patterned multi-slice excitation for magnetic resonance imaging.

BACKGROUND AND SUMMARY

[0003] Magnetic resonance imaging (MRI) is a medical imaging modality that can create images of the inside of a human body without using x-rays or other ionizing radiation. MRI uses the nuclear magnetic resonance (NMR) phenomenon to produce images. When a substance such as human tissue is subjected to a uniform magnetic field (Bo) in the z-direction, the individual magnetic moments of the nuclei in the tissue attempt to align with this magnetic field, but precess about the field in random order at their characteristic Larmor frequency. If the tissue is subjected to a radio-frequency (RF) excitation magnetic field (Bi) that is in the x-y plane and that is near the Larmor frequency, the net aligned moment may be rotated, or “tipped,” into the x-y plane to produce a net transverse magnetic moment. A signal is emitted by the excited nuclei or “spins,” after the excitation signal Bi is terminated, and this signal may be received and processed to form an image.

[0004] MRI can flexibly generate a variety of contrasts to accentuate different aspects of tissue composition, physiology, and structure. Contrast may be generated by a series of RF excitation pulses and magnetic field gradient pulses that are played out with specific timings and in a specific sequence. Upon preparation of the contrast, spatial information may be encoded into the signal to generate an image. BRIEF DESCRIPTION

[0005] In one embodiment, a method for imaging a patient according to a patterned multislice excitation (PME) magnetic resonance imaging (MRI) slice acquisition protocol includes, during a first pulse interval, applying a first radiofrequency (RF) pulse together with a first slice selection gradient to excite a first slice and a second slice, the second slice following the first slice in the slice acquisition protocol, and acquiring MR signals of only the first slice; and during a second pulse interval following the first pulse interval, applying a second RF pulse together with a second slice selection gradient to excite the second slice and a third slice, the third slice following the second slice in the slice acquisition protocol, and acquiring MR signals of only the second slice.

[0006] It should be understood that the brief description above is provided to introduce in simplified form a selection of concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The present disclosure will be better understood from reading the following description of non-limiting embodiments, with reference to the attached drawings, wherein below: [0008] FIG. 1 is a block diagram of an MRI system according to an embodiment of the disclosure.

[0009] FIG. 2 schematically shows a first example set of pulse sequences that may be carried out by an MRI system to perform two-slice excitation patterned multi-slice excitation (PME) MRI. [0010] FIG. 3 shows an example pulse sequence for one interval of the set of pulse sequences of FIG. 2.

[0011] FIG. 4 schematically shows a second example set of pulse sequences that may be carried out by an MRI system to perform two-slice excitation PME MRI.

[0012] FIG. 5 schematically shows a first example set of pulse sequences that may be carried out by an MRI system to perform three-slice excitation PME MRI. [0013] FIG. 6 schematically shows a second example set of pulse sequences that may be carried out by an MRI system to perform three-slice excitation PME MRI.

[0014] FIG. 7 shows an example pulse sequence for one interval of the set of pulse sequences of FIG. 5 or FIG. 6.

[0015] FIG. 8 is a flow chart illustrating an example method for performing PME MRI.

[0016] FIG. 9 schematically shows an example set of pulse sequences that may be carried out by an MRI system to perform four-slice excitation PME MRI.

[0017] FIG. 10A shows an example RF pulse set that may be applied during the pulse sequences of FIG. 9.

[0018] FIG. 10B schematically shows example gradient moments that may be observed during the RF pulse sets of the pulse sequences of FIG. 9.

[0019] FIG. 11 is a flow chart illustrating an example method for performing SMS-PME MRI.

[0020] FIG. 12 shows example PME-based measurement of brain tissue pulsations with the cardiac cycle.

[0021] FIG. 13 shows example PME-based measurement of water diffusion.

[0022] FIGS. 14A and 14B show histograms of SNR and tSNR for four subj ects scanned with PME and SMS-PME.

[0023] FIG. 15 shows histograms depicted SNR and tSNR averaged over 12 high-b directions for three subjects, comparing PME, SMS-PME, and PME without a time shift.

[0024] FIG. 16 shows images of MD, FA, and dominant diffusion directions for PME and SMS-PME scans.

DETAILED DESCRIPTION

[0025] The following description relates to magnetic resonance imaging (MRI) pulse sequences for simultaneously exciting a target slice for imaging and generating contrast in one or more upcoming slices in the slice acquisition protocol. Conventional approaches to generate Ti and T2 contrast, particularly in clinical brain imaging, may use a radiofrequency (RF) inversion recovery or 90°-180° RF spin echo (SE) preparation period to generate contrast in a slice. Diffusion-weighted MRI techniques similarly use a preparation period that combines a 90°-180° RF spin echo preparation with strong gradient pulses to create diffusion contrast. Typically, preparation periods are immediately followed by a localization segment where spatial information is encoded and a magnetic resonance (MR) signal is collected (referred to as an image acquisition segment). These methods allow control of contrast but may be time-inefficient as delays integral to signal preparation can be significant. As a result, a substantial fraction of the scan time is spent preparing contrast and not acquiring data. On the other hand, steady state free precession (SSFP) is a class of techniques that perform both functions concurrently and have improved timeefficiency; however, generated contrast is often more difficult to interpret and generally sub- optimal, at least in part due to the various types of NMR echoes that contribute to the SSFP signal and determine image contrast.

[0026] Thus, embodiments for patterned multi-slice excitation (PME) MRI are disclosed herein, to allow time-efficient generation of contrasts that require more than one RF pulse to generate, including spin echo (SE)- and stimulated echo (STE)-based techniques. PME changes the excitation pattern for each RF pulse to not just excite a target slice, but also manipulate (e.g., prepare) the magnetization of one or more upcoming slices in the multi-slice acquisition protocol. Of note, here and below use the term “excite” liberally to include broadly its effect on the NMR spin magnetization, including any rotation with a specified flip angle around a certain axis in the transverse plane. The entire PME spatial excitation pattern shifts with the slice order dictated by the slice acquisition protocol. Therefore, while acquiring data for a specific slice, the magnetization of to-be-imaged slices is altered to provide a way of generating contrast in a timeefficient manner. The slice pattern excited with each RF pulse, together with the slice acquisition protocol, dictates the sequence of RF pulses that each slice experiences in succession and provides flexibility in generating the desired contrast. As each RF excitation is followed by image acquisition, time efficiency is increased compared to standard SE and STE MRI by reducing the number of RF pulse intervals required for each slice. Unlike SSFP, PME MRI allows strict control of the types and number of echo signals that contribute to each image.

[0027] FIG. 1 illustrates an MRI system 100 that includes a static magnetic field magnet 102, a gradient magnetic field coil 104, an RF receiver coil unit 106, an RF transmitter coil unit 108, a patient table 110, a transmit/receive (T/R) switch 112, an RF driver 114, a gradient coil driver 116, a data acquisition unit 118, a controller 120, and a computing system 130.

[0028] The static magnetic field magnet 102 includes, for example, a superconductive magnet, a permanent magnet, or the like. The magnet defines a cylindrical space surrounding a subject 122 and generates a constant primary static magnetic field B . [0029] The gradient magnetic field coil 104 forms a gradient magnetic field in an imaging space 124 so as to provide the magnetic resonance signals (which will be received by the RF receiver coil unit) with three-dimensional positional information. The gradient magnetic field coil 104 includes three gradient coil systems, each of which generates a gradient magnetic field along a respective one of three spatial axes perpendicular to each other. For example, the gradient magnetic field coil 104 is formed by combining three coils (an X-axis gradient magnetic field coil, a Y-axis gradient magnetic field coil, a Z-axis gradient magnetic field coil) corresponding to respective axes of X, Y, and Z that are perpendicular to each other. These three coils generate a gradient magnetic field, the magnetic field intensity of which vary along the respective axes of X, Y, and Z, each receiving a separate supply of an electric current from the gradient magnetic field power source. The Z-axis direction is the same direction as the static magnetic field. Moreover, the Y-axis direction is a vertical direction, and the X-axis direction is a direction perpendicular to the Z axis and the Y axis.

[0030] Thus, the gradient magnetic field coil 104 generates a gradient field in each of a frequency encoding direction (e.g., along the read-out axis and thus also referred to as Gr), a phase encoding direction (e.g., along the phase-encode axis and thus also referred to as Gp), and a slice selection direction (e.g., along the slice-select axis and thus also referred to as Gs) in accordance with a specified pulse sequence (which may be dictated by a scan protocol or prescription). More specifically, the gradient magnetic field coil 104 applies a gradient field in the slice selection direction (or scan direction) of the subject 122, to select the slice (e.g., the gradient magnetic field for slice selection is used to determine an imaging section); and the RF transmitter coil unit 108 may transmit an RF pulse to a selected slice of the subject 122. The gradient magnetic field coil 104 also applies a gradient field in the phase encoding direction of the subject 122 to phase encode the magnetic resonance signals from the slice excited by the RF pulse (e.g., the gradient magnetic field for phase encoding is used to change a phase of an MR signal according to a spatial position). The gradient magnetic field coil 104 also applies a gradient field in the frequency encoding direction of the subject 122 (also referred to as a readout direction) to frequency encode the magnetic resonance signals from the slice excited by the RF pulse (e.g., the magnetic field for readout is used to change a frequency of an MR signal according to a spatial position).

[0031] In some embodiments, the RF receiver coil unit 106 is a surface coil, which is a local coil typically placed proximate to the anatomy of interest of the subject 122. The RF receiver coil unit 106 may include one or more RF coil elements, e.g., an array of coil elements. Herein, the RF transmitter coil unit 108 is a transmit coil that transmits RF signals, and the local surface RF receiver coil unit 106 receives the MR signals. As such, the transmit coil and the surface receive coil are separate but electromagnetically coupled components.

[0032] The RF transmitter coil unit 108 is disposed, for example, to enclose the imaging space 124, and produces RF magnetic field pulses orthogonal to the main magnetic field Bo produced by the static magnetic field magnet 102 within the imaging space 124 to excite the nuclei. In the static magnetic field space or imaging space 124 where a static magnetic field Bo is formed by the static magnetic field magnet 102, the RF transmitter coil unit 108 transmits, based on a control signal from the controller 120, an RF pulse that is an electromagnetic wave to the subject 122 and thereby generates a high-frequency magnetic field Bi. This excites proton spins (also referred to as “magnetization”) in the slice to be imaged of the subject 122. The RF receiver coil unit 106 receives, as a magnetic resonance signal, the electromagnetic wave generated when the proton spins thus excited in the slice to be imaged of the subject 122 returns into alignment with the initial magnetization vector. The RF receiver coil unit 106 is disposed, for example, to enclose the region to be imaged of the subject 122. In some examples, the RF receiver coil unit 106 may be referred to as the surface coil or the receive coil. In some embodiments, the RF receiver coil unit 106 may transmit the RF pulse and receive the MR signal. In other embodiments, the RF receiver coil unit 106 may only be used for receiving the MR signals, but not transmitting the RF pulse.

[0033] In contrast to the RF receiver coil unit 106, which may be disconnected from the MRI system 100 and replaced with another RF coil unit, the RF transmitter coil unit 108 is fixedly attached and connected to the MRI system 100. Furthermore, whereas local coils such as the RF receiver coil unit 106 can transmit to or receive signals from only a localized region of the subject 122, the RF transmitter coil unit 108 generally has a larger coverage area. The RF transmitter coil unit 108 may be used to transmit or receive signals to the whole body of the subject 122, for example.

[0034] Thus, the MRI system 100 transmits electromagnetic pulse signals to the subject 122 placed in the imaging space 124 with the static magnetic field formed therein to perform a scan for obtaining magnetic resonance signals from the subject 122. One or more images of the subject 122 can be reconstructed based on the magnetic resonance signals thus obtained by the scan. [0035] The T/R switch 112 can selectively electrically connect the RF transmitter coil unit 108 to the data acquisition unit 118 when operating in receive mode, and to the RF driver 114 when operating in transmit mode. Similarly, the T/R switch 112 can selectively electrically connect the RF receiver coil unit 106 to the data acquisition unit 118 when the RF receiver coil unit 106 operates in receive mode, and to the RF driver 114 when operating in transmit mode. When the RF receiver coil unit 106 and the RF transmitter coil unit 108 are both used in a single scan, for example if the RF receiver coil unit 106 is configured to receive MR signals and the RF transmitter coil unit 108 is configured to transmit RF signals, then the T/R switch 112 may direct control signals from the RF driver 114 to the RF transmitter coil unit 108 while directing received MR signals from the RF receiver coil unit 106 to the data acquisition unit 118.

[0036] The RF driver 114 is used to drive the RF coils (e.g., RF transmitter coil unit 108) and form a high-frequency magnetic field in the imaging space 124. The RF driver 114 modulates, based on a control signal from the controller 120 and using a gate modulator, the RF signal received from an RF oscillator into a signal of predetermined timing having a predetermined envelope. The RF signal modulated by the gate modulator is amplified by an RF power amplifier and then output to the RF transmitter coil unit 108.

[0037] The gradient coil driver 116 drives the gradient magnetic field coil 104 based on a control signal from the controller 120 and thereby generates a gradient magnetic field in the imaging space 124. The gradient coil driver 116 includes three systems of driver circuits (not shown) corresponding to the three gradient coil systems included in the gradient magnetic field coil 104.

[0038] The data acquisition unit 118 includes circuitry (e.g., a pre-amplifier, a phase-sensitive detector, an anal og/digi tai converter) used to acquire the magnetic resonance signals received by the RF receiver coil unit 106. In the data acquisition unit 118, the phase-sensitive detector detects, using the output from the RF oscillator of the RF driver 114 as a reference signal, the magnetic resonance signals received from the RF receiver coil unit 106 and amplified by the pre-amplifier, and outputs the detected phase-sensitive analog magnetic resonance signals to the analog/digital converter for conversion into digital signals. The digital signals thus obtained are output to the computing system 130. [0039] The MRI system 100 includes a table 110 for placing the subject 122 thereon. The subject 122 may be moved inside and outside the imaging space 124 by moving the table 110 based on control signals from the controller 120.

[0040] The controller 120 includes a processor configured to execute machine readable instructions stored in a non-transitory memory. The memory may comprise, for example, a semiconductor memory device, such as a random-access memory (RAM) and a flash memory, a hard disk, an optical disk, a ROM, flexible disk, magneto-optical disk, CD-ROM, or non-volatile memory card. The controller 120 is connected to the computing system 130 and processes the operation signals input to the computing system 130 and furthermore outputs control signals to controls the table 110, RF driver 114, gradient coil driver 116, and data acquisition unit 118. The controller 120 also controls, to obtain a desired image, the computing system 130.

[0041] The computing system 130 includes a user input device 138, such as a touchscreen, keyboard, and/or a mouse. The input device 138 is used by an operator, for example, to input such data as an imaging protocol and to set a region where an imaging sequence is to be executed. The data about the imaging protocol and the imaging sequence execution region are output to the controller 120.

[0042] The computing system 130 includes a processor 132 configured to execute machine readable instructions stored in a non-transitory memory 134. The computing system 130 is connected to the controller 120 and performs data processing based on control signals received from the controller 120. The computing system 130 is also connected to the data acquisition unit 118 and generates spectrum data by applying various image processing operations to the magnetic resonance signals output from the data acquisition unit 118.

[0043] The computing system 130 a display device 136 that displays an image on the display screen of the display device based on control signals received from the controller 120. The display 136 displays, for example, an image regarding an input item about which the operator inputs operation data from the input device 138. The display 136 also displays a two-dimensional (2D) slice image or three-dimensional (3D) image of the subject 122 generated by the computing system 130. For example, the processor 132 may execute instructions stored in memory 134 to perform one or more image reconstruction techniques on the data received from the data acquisition unit 118 in order to form the images, process the images (e.g., remove image artifacts from the images), store the images in memory, display the images via display 136, and/or send the images to a remote image storage device.

[0044] FIG. 2 schematically shows first example pulse sequences that may be applied by an MRI system (e.g., the MRI system 100 of FIG. 1) to perform PME MRI on a patient. The example shown in FIG. 2 is based on a sequential multi-slice acquisition protocol, a technique that excites and images slices through an object in a sequential manner: after exciting and imaging a specific slice with a pulse sequence, an adjacent slice is excited and imaged with the next iteration (“repetition”) of the pulse sequence. In a patterned multi-slice excitation, the target slice n (shown by a first fill in FIG. 2) in the acquisition protocol is excited as well as one additional slice (n+1, shown by a second fill in FIG. 2) to be acquired in the subsequent pulse sequence (also called “pulse interval" or “pulse sequence repetition”). Acquisition in each interval n is preceded by a crusher which suppresses gradient echo (GE) signals, and leads to the generation of just a spin echo signal from pulses in intervals n-1 and n, indicated as SE [n-l,n]. The slice location is shifted by one slice width on subsequent repetitions analogous to a sequential scanning protocol in conventional multi-slice MRI. Slice select gradients (not shown in FIG. 2) accompany the RF excitations.

[0045] Pulse sequences for three pulse intervals/repetitions (TR -1, TR 0, and TR 1) are shown in FIG. 2, including a first pulse sequence 210 for repetition TR -1, a second pulse sequence 220 for TR 0, and a third pulse sequence 230 for TR 1. For the first pulse sequence 210, a first RF pulse (RF -1) is applied (in combination with a slice selection gradient) to excite two adjacent slices, slice -1 and slice 0. The first RF pulse may be a multiplex of the individual RF pulses needed to stimulate each of the two slices, which may have the same or different flip angles. For example, the first RF pulse can be designed to use a 180°-90° combination to maximize the spin echo (SE) signal (where the 180° pulse is directed to slice -1 and the 90° pulse is directed to slice 0). For slice 0, which will be imaged in the next pulse sequence (the second pulse sequence 220), the first RF pulse coverts longitudinal magnetization into transverse magnetization. For slice -1, which is imaged in the first pulse sequence 210, the first RF pulse refocuses magnetization excited in a pulse sequence immediately prior to the first pulse sequence 210. After the first RF pulse, a crusher gradient (Cr -1) is applied to select a slice for imaging, which in the first pulse sequence is slice - 1. Thus, the first RF pulse excites both slice -1 and slice 0 but the crusher gradient selects signal from only slice -1 . [0046] The crusher gradient may be applied in the slice selection direction (e.g., along the slice-select axis), although crusher gradients applied in the phase- and/or frequency-encoding directions are also possible. The crusher gradient may be a single pulse having an amplitude selected to both eliminate signal originating from outside the target slice (e.g., slice 0) and refocus the SE signal (e.g., for slice -1). Thus, for the Gs gradient (e.g., the slice selection gradient), a slice selection pulse may be applied in synchronization with the first RF pulse and a crusher pulse may be applied immediately following the slice selection pulse (e.g., without any intervening Gs gradient pulses). As mentioned above, the crusher gradient suppresses GE signals and leads to the generation of only a SE signal during the acquisition period.

[0047] Following the first crusher gradient, a first acquisition (Acq -1) is performed. The first acquisition may include application of one or more readout gradient pulses (e.g., along the readout/frequency encoding direction, Gr). The one or more readout gradient pulses may have positive and/or negative polarities of suitable amplitude. During the first acquisition, MR signals are received by one or more receive RF coils (e.g., the RF receiver coil unit 106 of FIG. 1), which are then stored and used to reconstruct one or more images of the patient.

[0048] Once the first acquisition has been performed (e.g., upon conclusion of the readout pulses), the second pulse sequence 220 is carried out. The second pulse sequence 220 is similar to the first pulse sequence 210 but is shifted by one slice. Thus, for the second pulse sequence 220, a second RF pulse (RF 0) is applied to excite two adjacent slices, slice 0 and slice 1. For slice 0, the second RF pulse refocuses the transverse magnetization induced by the first RF pulse. The second RF pulse may be a multiplex of the individual RF pulses needed to stimulate each of the two slices. To generate a SE for slice 0, the first RF pulse, from the first pulse sequence 210, may include a pulse of 90° directed to slice 0 while the second RF pulse, from the second pulse sequence 220, may include a pulse of 180° directed to slice 0, to generate the SE [-1,0], Concurrently with the second RF pulse, a slice selection gradient is applied to select the slice 0 for imaging and slice 1 for preparation. After the second RF pulse, a crusher gradient (Cr 0) is applied to refocus signal from a single slice for imaging, which in the second pulse sequence is slice 0.

[0049] Thus, the second RF pulse excites both slice 0 and slice 1 but the crusher gradient refocuses signal from only slice 0 for imaging, while simultaneously spoiling (dephasing) signal from slice 1 so that it will not be observed by Acq 0. [0050] The second crusher gradient (Cr 0) may be similar to the first crusher gradient. Following the second crusher gradient, a second acquisition (Acq 0) is performed to acquire MR signals of slice 0, which may be performed similarly to the first acquisition.

[0051] Each subsequent pulse sequence/repetition may be performed similarly as described above, with each repetition shifted by one respective slice. For example, for the third pulse sequence 230, a third RF pulse (RF +1) is applied to slice 1 and slice 2, followed by a third crusher gradient (Cr +1) and then a third acquisition to acquire MR signals of slice 1. In this way, each repetition may include an RF pulse that excites two slices followed by a crusher gradient and then an acquisition. Following the acquisition, a next RF pulse is applied for the next interval/repetition. Applied to the measurement of SE signals, PME allows substantial shortening of scan time compared to conventional SE MRI, which requires two RF pulse intervals to acquire data for single slice, with one RF pulse converting longitudinal magnetization into transverse magnetization, and the subsequent pulse serving to refocus this magnetization. With PME, each RF pulse performs both these functions for two slices adjacent in the slice acquisition protocol. Thus, what otherwise would require multiple pulse intervals for each slice, here both contrast generation and image data acquisition is accomplished efficiently by collapsing multiple intervals into one and having each TR produce image data from a specific slice with SE-type contrast. The gradient crusher following the RF excitation allows for this approach, which serves to both eliminate signal originating from outside the target slice and refocus the SE signal.

[0052] FIG. 3 shows an example pulse sequence diagram 300 for one repetition/pulse interval of the PME MRI carried out according to FIG. 2, e.g., PME to generate SE-type contrast. Pulse sequence diagram 300 includes a first plot 310 showing acquisition data (and labeled Acq), specifically ADC signal data (e g., indicative of MR signal collection from RF receive coils by the data acquisition unit 118) over time, a second plot 320 showing RF amplitude as transmitted by RF transmitter coil(s) (e.g., the RF transmitter coil unit 108) over time (labeled RF), a third plot 330 showing the amplitude of the readout/frequency encoding gradient (e.g., along the read-out axis, Gr) over time (labeled Gr), a fourth plot 340 showing the amplitude of the phase encoding gradient (e.g., along the phase-encode axis, Gp) over time (labeled Gp), and a fifth plot 350 showing the amplitude of the slice selection (Gss) and crusher (Ger) gradients (e.g., along the sliceselect axis, Gs) over time (labeled Gs). Ger need not be exclusively (or at all) on the same gradient axis as Gss but can also be executed on one or more of the other axes. Timepoints of interest are shown by the dashed lines.

[0053] Between time TO and time Tl, an RF pulse is generated, as shown by second plot 320. The RF pulse may be a multiplexed pulse comprising a first individual RF pulse configured to excite a first slice (such as slice 0 of FIG. 2) and a second individual RF pulse configured to excite a second, adjacent slice (such as slice 1 of FIG. 2), with the first and second individual RF pulses multiplexed to form one overall RF pulse. A slice selection gradient is also applied between time TO and Tl, as shown by the fifth plot 350 and marked as Gss in FIG. 3, which acts to select the first slice (e.g., slice 0) and prepare the second slice.

[0054] A crusher gradient is applied between time Tl and time T2. The crusher gradient includes a negative polarity pulse in the slice selection direction, shown in fifth plot 350 and marked Ger. The crusher gradient pulse Ger has a larger amplitude than the slice selection pulse Gss, such as 2-5 times the amplitude. The crusher gradient pulse may have a sufficient amplitude and duration to fully disperse spins on the scale of the imaging resolution along that axis, here slice thickness. As shown, the crusher gradient also includes a positive polarity pulse in the phaseencoding gradient (shown by fourth plot 340).

[0055] At time T2, the crusher gradient ends and the acquisition period commences. During acquisition, the readout gradient is played out while the phase encoding gradient is pulsed. The readout gradient may include alternating positive and negative polarity pulses of equal amplitude (as shown by third plot 330). The phase-encoding gradient may include pulses with the same frequency as the readout gradient pulses (as shown by fourth plot 340). During the acquisition period, the ADC digitizes the MR signals obtained by the receive RF coils, as shown by first plot 310. The acquisition period ends at time T3, after which another pulse sequence may be executed for the next two slices (e.g., slice 1 and slice 2).

[0056] While FIG. 2 shows pulse sequences with a single RF pulse for each repetition, example PME pulse sequences that include more than one RF pulse per repetition are possible without departing from the scope of this disclosure, as shown in FIG. 4 and described in more detail below.

[0057] FIG. 4 shows a second example of a set of two-slice PME pulse sequences 400 that may be carried out by an MRI system (e.g., the MRI system 100 of FIG. 1) to perform PME MRI on a patient, where separate RF pulses are used to excite the individual slices. The example shown in FIG. 4 allows added flexibility in the design of the patterned excitation performed with each repetition, as well as shifting the spin echo refocusing time point along the acquisition interval. The second set of pulse sequences 400 is similar to the first set of pulse sequences 200 shown in FIG. 2, such that the target slice n (shown by the first fill in FIG. 4) is excited as well as one additional slice (n+1, shown by the second fill in FIG. 4) to be acquired in the subsequent pulse interval. Acquisition in each interval n is preceded by a crusher which suppresses GE signals, and leads to the generation of just a spin echo signal from pulses in intervals n-1 and n. The slice location is shifted by one slice width on subsequent repetitions. Slice select gradients (not shown in FIG. 4) accompany the RF excitations.

[0058] Pulse sequences for three pulse intervals/repetitions (TR -1, TR 0, and TR 1) are shown in FIG. 4, including a first pulse sequence 410 for repetition TR -1, a second pulse sequence 420 for TR 0, and a third pulse sequence 430 for TR 1. For the first pulse sequence 410, a first set of RF pulses including two RF pulses are applied in succession (RF -l_a and RF -lb) to excite two adjacent slices, slice -1 and slice 0. In some examples, the RF pulse RF -l_a may be directed to the first slice (slice -1) and the RF pulse RF -lb may be directed to the second slice (slice 0), while in other examples, RF -l_a may be directed to the second slice (slice 0) and RF - lb may be directed to the first slice (slice -1). Concurrently with each of the first set of RF pulses, slice selection gradients are applied. This slice select gradient may be of different amplitude for the two RF pulses a and b. After the first set of RF pulses (RF -l_a and RF -lb), a first crusher gradient (Cr -1) is applied to select a single slice for imaging, which in the first pulse sequence interval is slice -1. This first crusher gradient may be similar to the first crusher gradient Cr -1 of FIG. 2. Following the first crusher gradient, a first acquisition (Acq -1) is performed, which may be similar to the first acquisition Acq -1 of FIG. 2.

[0059] Thus, the first set of RF pulses excites both slice -1 and slice 0 but the crusher gradient selects signal from only slice -1.

[0060] Once the first acquisition has been performed (e.g., upon conclusion of the readout pulses), the second pulse sequence 420 is carried out. The second pulse sequence 420 is similar to the first pulse sequence 410 but is shifted by one slice. Thus, for the second pulse sequence 420, a second set of RF pulses (RF 0_a and RF 0b) is applied to excite two adjacent slices, slice 0 and slice 1. To generate a SE for slice 0, one RF pulse of the first set of RF pulses (e.g., RF -lb) may include a pulse of 90° directed to slice 0 while one RF pulse of the second set of RF pulses (e.g., RF 0_a) may include a pulse of 180° directed to slice 0, to generate the SE [-1,0], Concurrently with the second set of RF pulses, slice selection gradients are applied to select the slices for preparation and/or imaging, which for imaging in the second pulse sequence is slice 0. After this second set of RF pulses (RF 0_a and RF -Ob), a second crusher gradient (Cr 0) is applied which may be similar to the first crusher gradient. As with the first crusher gradient, the second crusher gradient refocuses signal from a single slice for imaging, which in the second pulse sequence is slice 0, while dephasing signal from other slice(s). Thus, the second set of RF pulses excites both slice 0 and slice 1 but the slice selection gradient refocuses signal from only slice 0.

[0061] Following the second crusher gradient, a second acquisition (Acq 0) is performed to acquire MR signals of slice 0, which may be performed similarly to the first acquisition.

[0062] Each subsequent pulse sequence/repetition may be performed similarly as described above, with each repetition shifted by one respective slice. For example, for the third pulse sequence 430, a third set ofRF pulses (RF +l_a andRF +lb) is applied to slice 1 and slice 2, followed by a third crusher gradient (Cr +1) and then a third acquisition to acquire MR signals of slice 1. In this way, each repetition may include two RF pulses that excite two slices followed by a crusher gradient and then an acquisition. Following the acquisition, a next set of RF pulses is applied for the next interval/repetition.

[0063] The application of two successive RF pulses per repetition allows shifting the spin echo refocusing point away from under the RF pulses. For example, FIG. 4 shows the SE generated by the RF pulses applied during TR -1 and TR 0 (referred to as SE [-1,0]) has an echo top that has shifted to the right relative to the echo top of the pulse sequence of FIG. 2. This two-pulse approach is also applicable to some three-slice implementations, which will be explained in more detail below. Slice selection gradient waveforms and timings can be made compatible with this approach, which may include refocusing lobes (brief periods of inverted or non-inverted slice selecting gradient) before, in between, and after each set of RF pulses. Thus, two RF pulses may be performed sequentially in time to excite at least two slices to allow shift of a refocusing point of a spin echo or a stimulated echo signal generated by the two RF pulses. In some examples, each slice may have an arbitrary and/or independent thickness and/or flip angle, depending on the slice acquisition protocol (which may also apply to the other PME pulse sequences disclosed herein). In this way, PME may be performed with any suitable flip angle, on any suitable slice thickness, which may be constant across slices or may change across slices. [0064] In some examples, the sequential excitation shown in FIG. 4 may be used to introduce Ti contrast in gradient echo type signals. For example, the first RF pulse of the two RF pulses (per pulse sequence) RF n_a may be an inversion (180°) or saturation (90°) pulse, followed by an excitation pulse (e.g., 90°) for the second RF pulse RF nb (with RF n_a and RF nb exciting different slices). In that case, the Acquisition (Acq) may occur before the crusher (Cr). Unwanted signal from the first RF pulse may be crushed by the slice select gradient of the second RF pulse, or an additional crusher may be inserted between the two RF pulses RF n_a and RF ni,. Accordingly, some PME pulse sequences may include a first RF pulse to excite a first slice, an optional first crusher gradient pulse, a second RF pulse to excite a second slice, acquisition of the first slice, and a second crusher gradient pulse.

[0065] Other example pulse sequences are possible, for example the two excited slices in the example presented with respect to FIG. 2 or FIG. 4 can be spaced to prepare signal for acquisition more repetitions in advance, thus creating a longer SE refocusing time (called echo time or “TE”). Furthermore, with each RF excitation, more than two slices can be excited simultaneously, resulting in spins within the same slice experiencing more than two RF pulses during execution of the multi -slice acquisition protocol.

[0066] FIGS. 5 and 6 show a first example and a second example, respectively, of a set of three-slice PME pulse sequences that may be carried out by an MRI system (e.g., the MRI system 100 of FIG. 1) to perform PME MRI on a patient. Each set of three-slice pulse sequences shown in FIGS. 5 and 6 is similar to the first set of pulse sequences 200 shown in FIG. 2, except that the target slice n is excited as well as two additional slices (n+1 and n+2, shown in FIG. 5, or n+2 and n+3, shown in FIG. 6) to be acquired in subsequent pulse intervals. Acquisition in each interval n is preceded by a crusher which suppresses GE signals, and leads to the generation of a stimulated echo signal from pulses in intervals n and the two prior pulsed intervals (e.g., n-1 and n-2 or n-2 and n-3). The slice location is shifted by one slice width on subsequent repetitions. Slice select gradients (not shown in FIG. 5 or FIG. 6) accompany the RF excitations.

[0067] FIG. 5 shows a first example of a set of three-slice pulse sequences 500. Pulse sequences for three pulse interval s/repetitions (TR -2, TR -1, and TR 0) are shown in FIG. 5, including a first pulse sequence 510 for repetition TR -2, a second pulse sequence 520 for TR -1, and a third pulse sequence 530 for TR 0. The three RF pulses of the pulse intervals shown in FIG. 5 result in the excitation of a target slice 0 that is excited by each of the three RF pulses. [0068] For the first pulse sequence 510, a first RF pulse (RF -2) is applied to excite three adjacent slices, slice -2, slice -1, and slice 0. The first RF pulse may have a suitable flip angle, such as between 90° and 120°. Concurrently with the first RF pulse, a slice selection gradient is applied to select and/or prepare the slices for imaging. A first crusher gradient (Cr -2) is applied following the first RF pulse. Thus, the first RF pulse excites all of slice -2, slice -1, and slice 0, but the crusher gradient refocuses signal from only slice -2.

[0069] The first crusher gradient (Cr -2) may be a single pulse having an amplitude selected to both eliminate signal originating from outside the target slice (e.g., slice -2) and refocus the STE signal from a slice from an earlier repetition. Thus, for slice selection, a gradient pulse (Gss in FIG. 7), may be applied in synchronization with the first RF pulse and a crusher pulse (Ger in FIG. 7) may be applied immediately following the slice selection pulse (e.g., without any intervening gradient pulses). As will be explained in more detail below, the amplitude of the crusher gradient pulse may vary across repetitions to eliminate SE signals.

[0070] Following the first crusher gradient, a first acquisition (Acq -2) is performed collecting image information for slice -2. This may include application of one or more readout gradient pulses and one or more phase encoding pulses. The one or more readout gradient pulses may have positive and/or negative polarities of suitable amplitude. During the first acquisition, MR signals are received by one or more receive RF coils (e.g., the RF receiver coil unit 106 of FIG. 1), which are then stored and used to reconstruct one or more images of the patient.

[0071] Once the first acquisition has been performed (e.g., upon conclusion of the readout pulses), the second pulse sequence 520 is carried out. The second pulse sequence 520 is similar to the first pulse sequence 510 but is shifted by one slice. Thus, for the second pulse sequence 520, a second RF pulse (RF -1) is applied to excite three adjacent slices, slice -1, slice 0, and slice 1. Concurrently with the second RF pulse, a slice selection gradient is applied. A second crusher gradient (Cr -1) is applied following the second RF pulse. Thus, the second RF pulse excites slice -1, slice 0, and slice 1, but the crusher gradient refocuses signal from only slice -1.

[0072] The second crusher gradient (Cr -1) may be similar to the first crusher gradient or may have a different amplitude, duration, or direction than the first crusher gradient. Following the second crusher gradient, a second acquisition (Acq -1) is performed to acquire MR signals of slice -1, which may be performed similarly to the first acquisition. [0073] Each subsequent pulse sequence/repetition may be performed similarly as described above, with each repetition shifted by one respective slice. Thus, for the third pulse sequence 530, a third RF pulses (RF 0) is applied to excite slice 0, slice 1 , and slice 2, followed by a third crusher gradient (Cr 0) and then a third acquisition (Acq 0) to acquire MR signals of slice 0. The three RF pulses (RF -2, RF -1, and RF 0) each excite slice 0, which along with the crusher gradients results in an STE (STE [-2,- 1,0]) that is measured/acquired during the third acquisition Acq 0.

[0074] In this way, each repetition may include an RF pulse that excites three slices followed by a crusher gradient and then an acquisition. Following the acquisition, a next RF pulse is applied for the next interval/repetition. The amplitude and/or duration of the crusher gradients may be modulated from repetition to repetition to eliminate SE signals, generation tissue displacement sensitivity, and/or generate diffusion contrast. Further, the crusher gradient’s effective direction may be varied to sensitize different water diffusion and tissue displacement directions. Furthermore, specificity to the STE signal may be achieved by reducing contributions of other signals by using flip angles (FAs) close to 90° and by varying gradient crusher moment and/or direction on subsequent pulse intervals.

[0075] FIG. 6 shows a second example of a set of three-slice pulse sequences 600. Pulse sequences for four pulse intervals/repetitions (TR -3, TR -2, TR -1, and TR 0) are shown in FIG. 6, including a first pulse sequence 610 for repetition TR -3, a second pulse sequence 620 for TR - 2, and a third pulse sequence 630 for TR -1, and a fourth pulse sequence 640 for TR 0. Three RF pulses of the four pulse intervals shown in FIG. 6 result in the excitation of a target slice 0 that is excited by each of the three RF pulses RF -3, RF -2 and RF 0, but not RF -1.

[0076] For the first pulse sequence 610, a first RF pulse (RF -3) is applied to excite three slices, slice -3, slice -1, and slice 0. Slice -2, which is intermediate slice -3 and slice -1, is not excited by the first RF pulse. Concurrently with the first RF pulse, a slice selection gradient is applied to select the slice for imaging. A first crusher gradient (Cr -3) is applied following the first RF pulse. Thus, the first RF pulse excites slice -3, slice -1, and slice 0, but the crusher gradient refocuses signal from only slice -3.

[0077] The first crusher gradient (Cr -3) may be a single pulse having an amplitude selected to both eliminate signal originating from outside the target slice (e.g., slice -3) and refocus the STE signal. Thus, for the slice selection gradient (Gss in FIG. 7), a slice selection pulse may be applied in synchronization with the first RF pulse and a crusher pulse may be applied immediately following the slice selection pulse (e.g., without any intervening gradient pulses). As will be explained in more detail below, the amplitude of the crusher gradient pulse may vary across repetitions to eliminate SE signals.

[0078] Following the first crusher gradient, a first acquisition (Acq -3) is performed, which may include application of one or more readout gradient pulses and one or more phase encoding pulses. The one or more readout gradient pulses may have positive and/or negative polarities of suitable amplitude. During the first acquisition, MR signals are received by one or more receive RF coils (e.g., the RF receiver coil unit 106 of FIG. 1), which are then stored and used to reconstruct one or more images of the patient.

[0079] Once the first acquisition has been performed (e.g., upon conclusion of the readout pulses), the second pulse sequence 620 is carried out. The second pulse sequence 620 is similar to the first pulse sequence 610 but is shifted by one slice. Thus, for the second pulse sequence 620, a second RF pulse (RF -2) is applied to excite three slices, slice -2, slice 0, and slice 1. Slice -1, which is intermediate slice -2 and slice 0, is not excited by the second RF pulse. Concurrently with the second RF pulse, a slice selection gradient is applied. A second crusher gradient (Cr -2) is applied following the second RF pulse. Thus, the second RF pulse excites slice -2, slice 0, and slice 1, but the crusher gradient refocuses signal from only slice -2.

[0080] The second crusher gradient (Cr -2) may be similar to the first crusher gradient or may have a different amplitude, duration, or direction than the first crusher gradient. Following the second crusher gradient, a second acquisition (Acq -2) is performed to acquire MR signals of slice -2, which may be performed similarly to the first acquisition.

[0081] Each subsequent pulse sequence/repetition may be performed similarly as described above, with each repetition shifted by one respective slice. Thus, for the third pulse sequence 630, a third RF pulses (RF -1) is applied to excite slice -1, slice 1, and slice 2, followed by a third crusher gradient (Cr -1) and then a third acquisition (Acq -1) to acquire MR signals of slice -1. Slice 0, which is intermediate slice -1 and slice 1, is not excited by the third RF pulse. For the fourth pulse sequence 640, a fourth RF pulse (RF 0) is applied to excite slice 0, slice 2, and slice 3, followed by a fourth crusher gradient (Cr 0) and then a fourth acquisition (Acq 0) to acquire MR signals of slice 0. Slice 1, which is intermediate slice 0 and slice 2, is not excited by the fourth RF pulse. Thus, three RF pulses (RF -3, RF -1, and RF 0) each excite slice 0, which along with the crusher gradients results in an STE (STE [-3, -1,0]) that is measured/acquired during the fourth acquisition Acq 0.

[0082] Thus, FIGS. 5 and 6 show examples of pulse sequences for PME MRI that may be applied to excite three slices per repetition. Three slice variants can, for example, be used to generate a stimulated echo (STE). Measurement of the STE signal while reducing contributions of other signals is possible by varying gradient crusher moment with specific alternation schemes and judicious choice of the RF flip angle. For example, a stimulated echo with a mixing time (TM) of two TR periods may be selected by alternating the amplitude of the crusher on successive repetitions (referred to as an “ab” crusher scheme) and using 90° flip angle (FA) for all slices that the RF pulse affects. The resulting [90°, 0°, 90°, 90°] spatial excitation pattern (whereby one RF pulse with a 90° FA excites three slices with an intermediate non-excited slice) greatly suppresses the SE signals while generating strong STE signals. Longer TE or TM and periods can be generated by having one or more of the pulses affecting slices longer in advance of their time of acquisition. This may be achieved by inserting gaps in the pattern of excited slices (as shown in FIG. 6) and may be accompanied by a modified gradient crusher modulation pattern. Three-slice pulse sequences can also be used to generate an SE signal with inversion-recovery or saturation-recovery weighting, for example by generating a [180°, 90°, 180°] or [90°, 90°, 180°] spatial excitation pattern. Further, more than three slices may be excited in a single pulse sequence (e.g., a four-slice excitation pulse sequence), which may increase Ti contrast in STE techniques by performing inversion or saturation preparations.

[0083] As mentioned above, different crusher gradient schemes may be applied depending on the slice acquisition protocol, and target contrast of the imaging session, number of slices being excited per repetition, etc. Table 1 shows example crusher schemes that may be applied during PME MRI for three-slice excitations (e g., where three slices are excited with each repetition), where the amplitude modulation of the crusher gradients over TR intervals allows selection of the STE signal. In Table 1, the letters a, b, and c indicate different crusher amplitudes. The letters x, y, and z in the excitation pattern indicate non-zero RF flip angle, while the number 0 indicates a zero degree flip angle, i.e. a gap in the excitation pattern for the slice of interest (i.e., that RF pulse does not act on the spins in the slice of interest). The crusher schemes allow suppression of specific echo signals; additional suppression may be possible by judicious choice of RF phase and amplitude. For example, the SE [-3, -1] contribution in the“xyz” scheme is suppressed by choosing x, y, and z all equal to 90°. For proper suppression, the amplitude difference between crushers should be sufficient to generate an intra-voxel dephasing of at least 2n radians over the voxel dimension along the crusher gradient axis for the SE signal. It will be appreciated that the order in which the slices experience the RF pulses is dependent on the slice acquisition protocol. In the sequential protocol of the example shown above, the order of experiencing RF flip angles is counter to the order of the excitation pattern, for example a [180°, 90°] spatial pattern is experienced as a 90°-180° temporal RF pulse sequence.

Table 1

[0084] FIG. 7 shows an example pulse sequence diagram 700 for one repetition/pulse interval of a three-slice excitation PME MRI carried out according to FIG. 5 or FIG. 6, e.g., PME to generate STE-type contrast. Pulse sequence diagram 700 includes a first plot 710 showing ADC signal data (e.g., indicative of MR signal collection from RF receive coils by the data acquisition unit 118) over time (labeled Acq), a second plot 720 showing RF amplitude as transmitted by RF transmitter coil(s) (e.g., the RF transmitter coil unit 108) over time (labeled RF), a third plot 730 showing the amplitude of the readout/frequency encoding gradient (e.g., along the read-out axis, Gr) over time (labeled Gr), a fourth plot 740 showing the amplitude of the phase encoding gradient (e.g., along the phase-encode axis, Gp) over time (labeled Gp), and a fifth plot 750 showing the amplitude of the slice selection (Gss) and crusher (Ger) gradients (e.g., along the slice-select axis, Gs) over time (labeled Gs). Ger need not be exclusively (or at all) on the same gradient axis as Gss but can also be executed on one or more of the other axes. Timepoints of interest are shown by the dashed lines. [0085] Between time TO and time Tl, an RF pulse is generated, as shown by second plot 720. The RF pulse may be a multiplexed pulse comprising a first individual RF pulse configured to excite a first slice (such as slice 0 of FIG. 5), a second individual RF pulse configured to excite a second, adjacent slice (such as slice 1 of FIG. 5), and a third individual RF pulse configured to excite a third, adjacent slice (such as slice 2 of FIG. 5), with the first, second, and third individual RF pulses multiplexed to form one overall RF pulse. A slice selection gradient is also applied between time TO and Tl, as shown by the fifth plot 750 and marked as Gss in FIG. 7. The desired 3 -slice PME pattern results from the combination of this slice selection (Gss) gradient with the dedicated RF pulse.

[0086] A crusher gradient is applied between time Tl and time T2. The crusher gradient includes a negative polarity pulse in the slice selection direction, shown in fifth plot 750 and marked Ger. The crusher gradient pulse Ger has a larger amplitude than the slice selection pulse Gss, such as 2-5 times the amplitude. The amplitude of the crusher gradient pulse may be selected based on the number and thickness of slices excited, the desired diffusion weighting or tissue displacement sensitivity, and how many slices are skipped in the excitation (such as according to Table 1). The frequency and phase encoding gradients may have relatively low amplitudes due to the long duration of the crusher gradient.

[0087] At time T2, the crusher gradient ends and the image acquisition period commences. During acquisition, the readout gradient is played out while the phase encoding gradient is pulsed. The readout gradient may include alternating positive and negative polarity pulses of equal amplitude (as shown by third plot 730). The phase-encoding gradient may include pulses with the same frequency as the readout gradient pulses (as shown by fourth plot 740). During the acquisition period, the ADC digitizes the MR signals obtained by the receive RF coils, as shown by first plot 710. The acquisition period ends at time T3, after which another pulse sequence may be executed for the next three slices (e.g., slice 1, slice 2, and slice 3).

[0088] FIG. 8 is a flow chart illustrating a method for performing two-slice PME for MRI, according to an embodiment of the disclosure. Method 800 may be carried out according to instructions stored in memory of a computing device and executed by one or more processors of the computing device, such as controller 120 and/or computing system 130 of FIG. 1, where the computing device is operably coupled to or included as part of an MRI system (e.g., MRI system 100 of FIG. 1). Method 800 may be executed in response to a request (e.g., received via user input to a user input device) to execute a scan protocol that includes a two-slice PME slice acquisition protocol for imaging a patient. The PME performed as part of method 800 may be the PME shown in FIGS. 2, 3, and/or 4 that generated SE-type contrast. The slice acquisition protocol may dictate the contrast type (e.g., SE or STE) and thus the number of slices to be excited per repetition, the total number of slices to be acquired, the slice thickness, the crusher amplitude, and other parameters of the scan of the patient.

[0089] At 802, method 800 includes exciting a dummy slice (slice 0) and a first slice (slice 1) with an RF pulse, where the first slice is a slice of a set of slices (N) to be imaged by the slice acquisition protocol. The dummy slice may be a slice positioned adjacent the first slice but that is not included in the slice acquisition protocol (and thus will not be imaged) and is positioned on the opposite side of the first slice from the second slice in the slice acquisition protocol. The RF pulse that is applied may be a single RF pulse that comprises a multiplexed signal configured to affect both the dummy slice and the first slice. In other examples, two successive RF pulses may be applied, each configured to excite a respective one of the dummy slice and the first slice. Further, a slice selection gradient may be applied concurrently with the RF pulse(s). However, in other examples, the dummy slice may not exist and the RF pulse be a single pulse configured to affect only the first slice.

[0090] At 804, a crusher gradient is applied following application of the RF pulse. The crusher gradient parameters (amplitude and direction) are chosen based on a pre-determined scheme and the crusher gradient serves to eliminate signal originating from outside the target slice and refocus the SE signal. The crusher gradient furthermore may serve to introduce a desired image contrast. [0091] At 806, the first slice (slice 1) and a second slice (slice 2) are excited with an RF pulse. The second slice may be adjacent the first slice in the slice acquisition protocol. During application of the RF pulse, a slice selection gradient may be applied to select and/or prepare the slices for imaging.

[0092] At 808, a crusher gradient is applied. The crusher gradient, as explained above, has a duration and amplitude selected to eliminate signal originating from outside the target slice (slice 1) and refocus the SE signal. The crusher gradient may be applied following the RF pulse, without other gradient pulses or RF pulses being applied between the end of the RF pulse and slice selection gradient and the start of the crusher gradient. At 810, acquisition of slice 1 is performed, following the end of the crusher gradient. The acquisition of slice 1 may include applying pulse(s) in the frequency encoding and phase encoding gradients and sampling the MR signals from the receive coils.

[0093] The above described pulse sequence may be repeated for each subsequent slice in the slice acquisition protocol, from the second slice (n=2) until the next to last slice of the slice acquisition protocol (n=N-l), as indicated at 812. Repeating the pulse sequence may include, as indicated at 814, exciting the current slice (slice n) and a subsequent slice (slice n+1) with an RF pulse. For example, after acquisition of slice 1, another RF pulse may be applied that excites slice 2 and slice 3, accompanied by a slice selection gradient. Repeating the pulse sequence may further include applying a crusher gradient after the RF pulse, as indicated at 816, and performing an acquisition of the current slice n (e.g., slice 2 in the example presented above) after the crusher pulse is applied and complete, as indicated at 818. The pulse sequence may be repeated until slice N-l is acquired, where the slice acquisition protocol dictates that N slices be acquired.

[0094] At 820, slice N is excited along with slice 1 with an RF pulse (and slice selection gradient), which is followed by a crusher gradient at 822. At 824, an acquisition of slice N is performed.

[0095] At 826, method 800 determines if the slice acquisition protocol is to be repeated. For example, the scan protocol may dictate that the N slices acquired according to the 2-slice PME slice acquisition protocol be repeated one or more times, in order to capture MR signals across a range of tissue motion, capture MR signals with different diffusion weighting, etc. If the slice acquisition protocol is to be repeated, method 800 returns to 806 to again excite slice 1 and slice 2 with an RF pulse, followed by application of a crusher gradient at 808 and acquisition of slice 1 at 810, and the pulse sequence is repeated for the remaining slices. During a repeat iteration of the slice acquisition protocol, one or more parameters of the pulse sequences may be adjusted, such as crusher gradient parameters (e g., amplitude, duration, direction).

[0096] Responsive to determining that the slice acquisition protocol is not to be repeated (e.g., all dictated slice acquisitions have been performed), method 800 proceeds to 828 to reconstruct one or more images from the data/MR signals acquired during the acquisition periods of the pulse sequences described above. At 830, the one or more images are saved in memory (of the computing device carrying out method 800 and/or in an image archive, such as a picture archive and communication system) and/or output for display on a display device. Method 800 then ends. [0097] Thus, method 800 provides for MR imaging of a patient using PME to simultaneously excite a target slice and prepare contrast for a subsequent slice. The slice acquisition protocol may include exciting a dummy slice when initially exciting the first slice, as two excitations are needed to properly excite each slice. However, once the slices have all been acquired during a first iteration of the slice acquisition protocol, if the slice acquisition protocol is repeated, the first slice may be excited in a cyclical manner during the second excitation of the final slice of the slice acquisition protocol. For example, for an acquisition protocol that dictates four slices be acquired each iteration of the acquisition protocol, the excited slices (where slice 0 is the dummy slice) may be 0,1; 1,2; 2,3; 3,4; 4,1; 1,2; 2,3; 3,4; 4,1; 1,2 etc.

[0098] Method 800 specifically described a PME pulse sequence for exciting two slices with a single RF pulse for each repetition, but could be modified to excite two slices with two RF pulses for each repetition. Further, while method 800 excited two adjacent slices each repetition, method 800 could be modified to instead excite two non-adjacent slices per repetition. Further still, method 800 could be modified to excite three, or more, slices per repetition, whether the three or more slices are all adjacent (as explained above with respect to FIG. 5), or include one or more intermediate, non-excited slices (as explained above with respect to FIG. 6). The slices may be of equal width, or have varying widths.

[0099] More generally, a desired temporal sequence of RF excitations experienced by each slice can be designed by generating a specific PME pattern; the slice acquisition protocol then in effect converts this pattern into a temporal sequence. The slice acquisition protocol need not include only excitation of consecutive sequential slices; for example, a slice-interleaved scanning protocol (e.g., sequentially scanning first odd-numbered, then even-numbered slices) may be implemented. A slice-interleaved protocol may include adjusting the PME pattern by inserting “null” excitations: for example, the STE acquisition mentioned above would instead of a [90°, 0°, 90°, 90°] pattern for sequential slice acquisition utilize a [90°, 0°, 0°, 0°, 90°, 0°, 90°] pattern for interleaved acquisition using the same timing characteristics (TM and TE duration).

[0100] The PME pulse sequences described herein may be applicable for various MRI scanning protocols, to generate desired contrast, such as spin echo, stimulated echo, gradient echo with Ti weighting, spin echo with Ti weighting, or stimulated echo with Ti weighting. For example, the PME pulse sequences described herein may be used to introduce diffusion contrast by using a relatively large crusher amplitude (e.g., 50 mT/m for several, such as 2-5, milliseconds). Unlike diffusion-weighted (DW) SSFP, the proposed PME approach applied to STE acquisition (e.g., as shown in FIGS. 5 and 6) allows long TM STEs to be selectively generated, increasing diffusion contrast. Since a specifically selected signal pathway is selected in PME, the accurate diffusion weighting factor (“b-value”) can be computed. Similarly, using a large crusher in the PME SE protocol (e.g., as shown in FIGS. 2 and 4) introduces diffusion contrast and allows diffusion-weighted MRI with high time-efficiency since a slice is scanned for each pulse-interval. In phase images, large crushers lead to sensitivity of the SEs and STEs to subtle tissue displacement. This has been exploited with techniques like DENSE and SSFP for the measurement of brain pulsations and elastography. This tissue motion imaging may be more efficiently done with the PME concept using an SE approach.

[0101] Thus, a class of pulse sequences, PME MRI, is disclosed herein that allows timeefficient generation of various contrasts in multi-slice MRI. This approach may be valuable in multiple different implementations, such as the measurement of brain tissue pulsations and water diffusion (examples of which are provided below). In these implementations, collapsing of multiple pulse sequence segments into one allows a gain in time efficiency over conventional approaches. The extent of this gain is dependent on the precise implementation. Larger gains are possible for implementations with longer preparation intervals, for example with relatively large Ti- , T2- , diffusion-, or displacement-weighting. Generally, PME implementations of STE-based MRI allow the largest gains, as they allow collapsing three pulse intervals into one, as compared to two pulse intervals into one for SE-based MRI.

[0102] Applied to the measurement of brain tissue displacement with the cardiac cycle, PME offers advantages over conventional multi-slice MRI as it eliminates the need for a separate contrast preparation segment in the pulse sequence. Depending on the required level of displacement sensitivity, this allows up to 40% improvement in time-efficiency when using SE- based techniques. This translates into an improved temporal resolution for capturing the temporal dynamics of brain tissue motion. Up to now, displacement measurement techniques have employed STE-based DENSE methodology, which may have advantages over SE when extremely high displacement sensitivity is required or high performance gradients are not available. Incorporation of PME into multi-slice STE allows improvements in time efficiency in excess of 50%, as 3 pulse sequence segments can be collapsed into one. [0103] For diffusion-weighting (DW), as with displacement encoding, contrast preparation increases scan time. The most common technique used for DW MRI is multi-slice SE where diffusion preparation is accomplished with a pair of strong gradient pulses around the refocusing pulse. These gradients typically take up a sizable portion of the RF pulse interval. One of these pulses dephases the MRI signal, while the other rephases, and in total twice as many pulses as acquired slices are needed. With PME, each pulse simultaneously serves to dephase one slice and rephase another, halving the number of pulses needed. Compared to conventional SE acquisition, this reduced time required per slice by about 45% for a 2 mm resolution scan (explained in the Examples below). This time savings may vary depending on various parameters, including available gradient strength, requested b-value, and spatial resolution.

[0104] Under particular conditions (e.g., very high b-values (several ms/pm²), or the lack of high performance gradient hardware), DW MRI may require long TE values and lead to excessive signal loss due to T2 decay. STE-based multi-slice acquisition may ameliorate this problem as it allows increasing b by increasing TM rather than TE. As a result, signal loss may be reduced because of the much slower Ti decay that occurs during TM. This comes at the price of an up to 50% signal loss inherent to the STE generation mechanism. In addition, conventional STE requires long magnetization preparation periods that reduce time-efficiency (although this may be alleviated by “bunching” preparations and acquisitions of sets of slices). Alternatively, 3D SSFP allows rapid STE-sensitized DW MRI with moderately high b-values but is sensitive to motion due its multi-shot nature. Furthermore, the various signal pathways that contribute to the SSFP signal make it difficult to control and quantify the effective b-value. PME-based DW MRI overcomes both these issues and allows time-efficient, long TM STE MRI with full brain coverage. Furthermore, it provides control over the contributions of the multitude of echo pathways that typically contribute to the SSFP signal, thereby improving interpretability and quantifiability.

[0105] Like SSFP, the PME approach does not readily allow measurement at the echo top, which in some examples coincides with the RF excitation pulses. This introduces some T2* weighting and associated signal loss. With certain PME implementations, it is possible to overcome this. For example, the T2* weighting and signal loss may be overcome by shifting the STE and SE signals out from under the RF pulse, as shown in FIG. 4. A drawback of this approach is a minor loss in efficiency due to the reduced time available for image acquisition. [0106] Although PME does not dictate the use of specific flip angles (like 90° and 180°), the flip angles discussed herein (and for the Examples presented below) may exceed those typically used in 3D SSFP. Therefore, RF power requirements and tissue deposition may be increased. Similarly, RF power in PME MRI may exceed that used in conventional multi-slice MRI. For the applications discussed above, power was typically below selective absorption rate (SAR) safety limits. However, at high field (7 T and above), SAR limits may limit the range of possible PME applications. Under specific conditions (when adjacent slices are excited with the same flip angle), RF excitation may be simplified and power deposition reduced, removing some of this limitation. [0107] The single-shot echo-planar imaging (EPI) acquisitions shown in the Examples presented below may be too long to fit in a TR that would provide optimal contrast to noise ratio (CNR). However, for applications where large crushers are not required (e.g., for Ti and T2 weighting), motion sensitivity is reduced, and multi-shot acquisition may be practical.

[0108] Another potential area of PME application is rapid asymmetric spin-echo (ASE) MRI, which provides a combination of T2 and T2* weighting and may be less susceptible to draining vein artifacts and Bo field inhomogeneities than purely T2-weighted MRI. At the same time, PME based ASE MRI can provide whole brain coverage with better time resolution than is available with conventional ASE MRI. This can find potential use in applications where excellent temporal resolution is desired, such as fMRI or bolus-tracking based on intravenously injected contrast agents.

[0109] Further, PME may be compatible with simultaneous multi-slice (SMS) MRI as long as the slice separation of the latter is larger than the extent of the pattern excited with PME. In this way, PME may be used to excite both a first set of two or more slices and a second set of two or more slices with an RF pulse or a set of RF pulses, and acquire the signals from one slice of each of these sets simultaneously.

[0110] While some of the PME implementations described herein use a sequential or interleaved slice excitation scheme of shifting the excitation pattern by a specific number of slices on successive RF pulses, this is not strictly required. In principle, a slice acquisition with any type of order, or even randomized order, could be used. However, the interleaved scheme (as well as a sequential slice acquisition scheme) allows easily interpretable relationship between PME pattern and the time-sequence of RF pulses each imaged slice experiences (a traversal through the spatial excitation pattern counter to the slice order). This becomes more complicated when more randomized slice excitation schemes are used.

[0111] PME may be advantageously applied to STE- and SE-based MRI, as they require more than one RF excitation pulse and readily allow suppression of signal that is excited but not targeted to be measured in a specific pulse sequence interval. As described herein, this was accomplished with a crusher gradient which also served to generate the desired contrast. If this “outside” signal originates from a sufficiently remote location, it may be possible to omit the crusher and separate the unwanted and desired GRE signals through parallel imaging reconstruction. This may be useful for GRE experiments that require a slice selective magnetization preparation, e.g., when using an inversion pre-pulse to generate Ti contrast.

[0112] Spin-echo EPI (SE EPI) is an efficient technique for applications such as DW MRI and displacement-encoded MRI. The application of SMS techniques to SE EPI may further improve efficiency, allowing high angular resolution DW schemes in clinically feasible scan times. However, some inefficiency still remains due to the time needed to establish the demanded contrast. As mentioned above, PME may be used to perform SMS MRI, wherein both a first set of two or more slices and a second set of two or more slices are excited with an RF pulse or a set of RF pulses, and the signals from one slice of each of these sets is acquired simultaneously. By combining PME and SMS, the efficiency of SE EPI scans may be further increased, as explained in more detail below.

[0113] FIG. 9 shows an example of a set of SMS-PME pulse sequences 900 that may be carried out by an MRI system (e.g., the MRI system 100 of FIG. 1) to perform SMS-PME MRI on a patient. Each set of pulse sequences shown in FIG. 9 is similar to the first set of pulse sequences 200 shown in FIG. 2, except that two target slices are excited as well as two additional slices to be acquired in a subsequent pulse interval. Thus, FIG. 9 shows two-slice PME combined with twofold SMS acceleration. Acquisition in each interval is preceded by a crusher which suppresses GE signals, and leads to the generation of a stimulated echo signal. The slice location is shifted by one slice width on subsequent repetitions. Slice select gradients (not shown in FIG. 9) accompany the RF excitations.

[0114] Pulse sequences for three pulse intervals/repetitions (TR 0, TR 1 and TR 2) are shown in FIG. 9, including a first pulse sequence 910 for repetition TR 0, a second pulse sequence 920 for TR 1 , and a third pulse sequence 930 for TR 2. The two RF pulses of each pulse interval shown in FIG. 9 result in the excitation of two target slices that are excited by a respective one of the two RF pulses.

[0115] For the first pulse sequence 910, a first RF pulse set 902 is applied to excite a first slice, herein slice 1, and a parallel slice that is sufficiently spatially separated from the first slice, herein slice 31. (FIG. 9 shows a 60-slice acquisition and it is to be appreciated that the parallel slice excited in the first pulse sequence 910 may be a different slice depending on how many slices are to be acquired.) The RF pulses of the first RF pulse set 902 may have a suitable flip angle, such as between 90° and 120°, though a flip angle of 90° is shown herein. The RF pulses of the first RF pulse set 902 may be offset in the time domain. For example, the RF pulse that excites slice 31 may be initiated 0.9 ms after the RF pulse that excites slice 1 is initiated. Concurrently with the first RF pulse set 902, a slice selection gradient is applied to select and/or prepare the slices for imaging, which is described in more detail below with respect to FIG. 10B. A crusher gradient is applied following the first RF pulse set. As TR 0 aims to prepare contrast for a subsequent repetition/interval, no acquisition occurs during TR 0.

[0116] Upon conclusion of the crusher gradient, the second pulse sequence 920 is performed for TR 1. The second pulse sequence 920 includes a second RF pulse set 904. The second RF pulse set 904 may include a first multiplexed RF pulse that comprises a first individual RF pulse configured to excite the first slice (such as slice 1 of FIG. 9) with a flip angle of 180° and a second individual RF pulse configured to excite a second, adjacent slice (such as slice 2 of FIG. 9) with a flip angle of 90°, with the first and second individual RF pulses multiplexed to form one overall multiplexed RF pulse. Similarly, the second RF pulse set 904 may include a second multiplexed RF pulse comprising a third individual RF pulse configured to excite the parallel slice (such as slice 31 of FIG. 9) with a flip angle of 180° and a fourth individual RF pulse configured to excite a second, adjacent parallel slice (such as slice 32 of FIG. 9) with a flip angle of 90°, with the third and fourth individual RF pulses multiplexed to form the second multiplexed RF pulse. The multiplexed RF pulses may be offset from each other such that the peak of the second multiplexed RF pulse occurs after the peak of the first multiplexed RF pulse. Thus, the second RF pulse set 904 excites four slices: slice 2 and 32 with a flip angle of 90°, and slice 1 and 31 with a flip angle of 180°. This approach is similar to the pulse sequences shown in FIG. 4, but doubling (for two-fold SMS, also referred to as SMS-2) the number of excited slices for both the 90° and 180° excitations to excite a duplicate set of slices some distance away. During reconstruction, this duplicate set is distinguished from the original set based on the spatial information contained in the coil sensitivity profiles.

[0117] Concurrently with the second RF pulse set 904, a slice selection gradient is applied. A crusher gradient 906 is applied following the second RF pulse set 904. The crusher gradient 906 may be a single pulse having an amplitude selected to both eliminate signal originating from outside the target slices (e.g., slices 1 and 31) and refocus the SE signal from a slice from an earlier repetition. Thus, for slice selection, a gradient pulse may be applied in synchronization with the second RF pulse set 904 and a crusher pulse may be applied immediately following the slice selection pulse (e.g., without any intervening gradient pulses). In some examples, the amplitude of the crusher gradient pulse may vary across repetitions to select STE signals.

[0118] Following the crusher gradient, a first acquisition 908 is performed collecting MR signals for the first slice (slice 1) and the parallel slice (slice 31). The first acquisition may include application of one or more readout gradient pulses and one or more phase encoding pulses. The one or more readout gradient pulses may have positive and/or negative polarities of suitable amplitude. During the first acquisition 908, MR signals are received by receive RF coils (e.g., the RF receiver coil unit 106 of FIG. 1), which are then stored and used to reconstruct one or more images of the patient. MR signals of the first slice (slice 1) may be disentangled from MR signals of the parallel slice (slice 31) based on the information contained in the receive RF coil sensitivity profiles, such that the signals of the various receive RF coils are combined in specific ways to separate the MR signals of the first slice and the parallel slice.

[0119] Once the first acquisition has been performed (e.g., upon conclusion of the readout pulses), the third pulse sequence 930 is carried out. The third pulse sequence 930 is similar to the second pulse sequence 920 but is shifted by one slice. Thus, for the third pulse sequence 930, a third RF pulse set 912 is applied to excite four slices (slice 2, slice 3, slice 32, and slice 33, with slices 2 and 32 excited with a flip angle of 180° and slices 3 and 33 excited with a flip angle of 90°). Concurrently with the third RF pulse set 912, a slice selection gradient is applied. A second crusher gradient 914 is applied following the third RF pulse set 912. Thus, the third RF pulse set 912 excites slice 2, slice 3, slice 32, and slice 33, but the crusher gradient refocuses signal from only slices 2 and 32.

[0120] The second crusher gradient 914 may be similar to the crusher gradient 906 or may have a different amplitude, duration, or direction than the first crusher gradient. Following the second crusher gradient 914, a second acquisition 916 is performed to acquire MR signals of slices 2 and 32, which may be performed similarly to the first acquisition.

[0121] Each subsequent pulse sequence/repetition may be performed similarly as described above, with each repetition shifted by one respective slice. Thus, for a subsequent pulse sequence (e.g., for TR 3), a fourth RF pulse set is applied to excite slice 3, slice 4, slice 33, and slice 34, followed by a third crusher gradient and then a third acquisition to acquire MR signals of slice 3 and slice 33.

[0122] In this way, by combining SMS and PME, four slices may be excited in parallel, so that magnetization of two upcoming slices may be prepared while simultaneously acting on two current slices to be imaged. Crusher gradients are included to generate tissue displacement or diffusion contrast. The RF pulse for exciting four slices may result in peak RF amplitude limitations imposed by the MRI system (e.g., the MRI system of FIG.1), and thus a time shift is incorporated between RF pulses, as explained above. For example, FIG. 10A shows an example RF pulse set 1000 including a first multiplexed RF pulse 1002 and a second multiplexed RF pulse 1004. The first multiplexed RF pulse 1002 may be configured to excite two slices (e.g., slice 1 and slice 2) and the second multiplexed RF pulse 1004 may be configured to excite two additional slices (e.g., slice 31 and slice 32). The peak of the second multiplexed RF pulse 1004 may be timeshifted relative to the peak of the first multiplexed RF pulse 1002, such as occurring 0.91ms later. In doing so, peak RF amplitude may be maintained under the peak RF amplitude limit.

[0123] In some examples, the slice selection gradients accompanying the RF pulse sets may be alternated in polarity over repetitions/shots to suppress the refocusing of the lipid signal, as explained below. Further, the crusher gradient scheme results in full through-slice refocusing of the water signal in all four slices.

[0124] FIG. 10B schematically shows a plot 1050 of alternating slice-selection gradient polarity across two RF pulse sets for an SMS-PME MRI protocol. Specifically, gradient moments for slice 2 and slice 32 are shown across the second RF pulse set 904 and the third RF pulse set 912 of FIG. 9. As explained above, the second RF pulse set 904 excites slice 2 and 32 at a first flip angle (herein 90°, as shown by 7t/2) and excites slice 1 and slice 31 at a second flip angle (herein 180°, as shown by 7i). In the example shown in FIG. 10B, the second RF pulse set 904 includes four time-shifted RF pulses, with each RF pulse exciting one slice. The third RF pulse set 912 excites slice 3 and 33 at the first flip angle (herein 90°, as shown by TT/2) and excites slice 2 and slice 32 at the second flip angle (herein 180°, as shown by K). In the example shown in FIG. 10B, the third RF pulse set 912 includes four time-shifted RF pulses, with each RF pulse exciting one slice. A slice-selection gradient of a first polarity (e.g., positive) is applied during the second RF pulse set 904 and a slice-selection gradient of a second polarity (e.g., negative) is applied during the third RF pulse set 912.

[0125] Plot 1050 further includes a gradient moment for slice 32, shown by line 1052, and a gradient moment for slice 2, shown by line 1054. As shown by line 1052, the 90° pulse for slice 32 generates a signal that is dephased by the positive gradient applied during the second RF pulse set 904. The signal is then rephased by the negative gradient applied during the third RF pulse set 912. The refocusing pulse for slice 32 (e.g., the RF pulse with the flip angle of 180°) is performed during the third RF pulse set 912 and the negative gradient that continues after the refocusing pulse brings the moment to zero. Similarly, as shown by line 1054, the 90° pulse for slice 2 generates a signal that is dephased by the positive gradient applied during the second RF pulse set 904. The signal is then rephased by the negative gradient applied during the third RF pulse set 912. The refocusing pulse for slice 2 (e.g., the RF pulse with the flip angle of 180°) is performed during the third RF pulse set 912 and the negative gradient that continues after the refocusing pulse brings the moment to zero. The timing of the pulses, with a time shift of two times t between the two 90° pulses, one times t between the 90° and 180° pulses and one times t shift between the two 180° pulses, works to refocus the signals from both slice 2 and slice 32 at the same time at the end of the slice select gradient.

[0126] FIG. 11 is a flow chart illustrating a method for performing SMS-PME for MRI, according to an embodiment of the disclosure. Method 1100 may be carried out according to instructions stored in memory of a computing device and executed by one or more processors of the computing device, such as controller 120 and/or computing system 130 of FIG. 1, where the computing device is operably coupled to or included as part of an MRI system (e.g., MRI system 100 of FIG. 1). Method 1100 may be executed in response to a request (e.g., received via user input to a user input device) to execute a scan protocol that includes a SMS-PME slice acquisition protocol for imaging a patient. The SMS-PME performed as part of method 1100 may be the SMS- PME shown in FIG. 9 that generates SE-type contrast. The slice acquisition protocol may dictate the contrast type (e.g., SE or STE) and thus the number of slices to be excited per repetition, the total number of slices to be acquired, the slice thickness, the crusher amplitude, and other parameters of the scan of the patient.

[0127] At 1102, method 1100 includes exciting a first slice (slice 1) and a parallel slice (N/2+1) with an RF pulse, where the first slice and the parallel slice are slices of a set of slices (N) to be imaged by the slice acquisition protocol. For example, if the slice acquisition protocol dictates that 60 slices be acquired, the parallel slice may be slice 31 (e.g., 60/2+1). In some examples, one or more dummy slices may also be excited. The dummy slice may be a slice positioned adjacent the first slice but that is not included in the slice acquisition protocol (and thus will not be imaged) and is positioned on the opposite side of the first slice from the second slice in the slice acquisition protocol. The RF pulse that is applied may be a single RF pulse that comprises a multiplexed signal configured to affect both the first slice and the parallel slice. In other examples, two successive RF pulses may be applied, each configured to excite a respective one of the first slice and the parallel slice. Further, a slice selection gradient may be applied concurrently with the RF pulse(s).

[0128] At 1104, a crusher gradient is applied following application of the RF pulse. The crusher gradient parameters (amplitude and direction) are chosen based on a pre-determined scheme and the crusher gradient serves to eliminate signal originating from outside the target slices and refocus the SE signal. The crusher gradient furthermore may serve to introduce a desired image contrast.

[0129] At 1106, the first slice (slice 1), the parallel slice (slice N/2+1), a second slice (slice 2), and a second parallel slice (slice N/2+2) are excited with an RF pulse set comprising at least two RF pulses. For example, a first RF pulse may be applied to excite the first slice and the second slice and a second RF pulse may be applied to excite the parallel slice and the second parallel slice. The second slice may be adjacent the first slice in the slice acquisition protocol and the second parallel slice may be adjacent to the parallel slice in the slice acquisition protocol. During application of the RF pulse set, a slice selection gradient may be applied to select and/or prepare the slices for imaging.

[0130] At 1108, a crusher gradient is applied. The crusher gradient, as explained above, has a duration and amplitude selected to eliminate signal originating from outside the target slices (slice 1 and N/2+1) and refocus the SE signal. The crusher gradient may be applied following the RF pulse set, without other gradient pulses or RF pulses being applied between the end of the RF pulse set and slice selection gradient and the start of the crusher gradient. At 11 10, acquisition of slice 1 and slice N/2+1 is performed, following the end of the crusher gradient. The acquisition of slice 1 and slice N/2+1 may include applying pulse(s) in the frequency encoding and phase encoding gradients and sampling the MR signals from the receive coils, wherein MR signals originating from slice 1 and MR signals originating from slice N/2+1 are separated by making appropriate combinations of the signals measured with the receive coils, taking advantage of the differences in the spatial profiles of coil sensitivities.

[0131] The above described pulse sequence may be repeated for each subsequent slice in the slice acquisition protocol, other than the final two slices of the acquisition protocol (slice N and N/2), as indicated at 1112. Repeating the pulse sequence may include, as indicated at 1114, exciting the next four slices with an RF pulse set. For example, after acquisition of slice 1 and slice N/2+1, another RF pulse set may be applied that excites slice 2 and slice 3 and slice N/2+2 and N/2+3, accompanied by a slice selection gradient. Repeating the pulse sequence may further include applying a crusher gradient after the RF pulse set, as indicated at 1116, and performing an acquisition of the two current slices (e.g., slice 2 and slice N/2+2 in the example presented above) after the crusher pulse is applied and complete, as indicated at 1118. The pulse sequence may be repeated until slice N-l and slice N/2-1 are acquired, where the slice acquisition protocol dictates that N slices be acquired.

[0132] At 1120, slice N and slice N/2 are excited along with slice 1 and slice N/2+1 with an RF pulse set (and slice selection gradient), which is followed by a crusher gradient at 1122. At 1124, an acquisition of slice N and slice N/2 is performed.

[0133] At 1126, method 1100 determines if the slice acquisition protocol is to be repeated. For example, the scan protocol may dictate that the N slices acquired according to the SMS-PME slice acquisition protocol be repeated one or more times, in order to capture MR signals across a range of tissue motion, capture MR signals with different diffusion weighting, etc. If the slice acquisition protocol is to be repeated, method 1100 returns to 1106 to again excite slice 1, slice N/2+1, slice 2, and slice N/2+2 with an RF pulse set, followed by application of a crusher gradient at 1108 and acquisition of slice 1 and N/2+1 at 1110, and the pulse sequence is repeated for the remaining slices. During a repeat iteration of the slice acquisition protocol, one or more parameters of the pulse sequences may be adjusted, such as crusher gradient parameters (e.g., amplitude, duration, direction). [0134] Responsive to determining that the slice acquisition protocol is not to be repeated (e.g., all dictated slice acquisitions have been performed), method 1100 proceeds to 1128 to reconstruct one or more images from the data/MR signals acquired during the acquisition periods of the pulse sequences described above. At 1130, the one or more images are saved in memory (of the computing device carrying out method 1100 and/or in an image archive, such as a picture archive and communication system) and/or output for display on a display device. Method 1100 then ends. Alternatively, the reconstruction of the images could commence as soon as sufficient MR data is available (e.g., after the acquisition of the slices at 1118 or after a first iteration of the slice acquisition protocol is complete), and continue as additional MR data is acquired.

[0135] Thus, method 1100 provides for MR imaging of a patient using SMS-PME to simultaneously excite two target slices and prepare contrast for a subsequent slice. The slice acquisition protocol may include exciting a dummy slice when initially exciting the first slice and parallel slice, as two excitations are needed to properly excite each slice. However, once the slices have all been acquired during a first iteration of the slice acquisition protocol, if the slice acquisition protocol is repeated, the first slice and the parallel slice may be excited in a cyclical manner during the second excitation of the final slice of the slice acquisition protocol.

[0136] Method 1100 specifically described an SMS-PME pulse sequence for exciting four slices with one composite RF pulse for each repetition, but could be modified to excite four slices with the composite pulse time-separated into four RF pulses for each repetition. Further, while method 1100 excited two sets of adjacent slices each repetition, method 1100 could be modified to instead excite two sets of non-adjacent slices per repetition. Further still, method 1100 could be modified to excite six, or more, slices per repetition. The slices may be of equal width, or have varying widths.

Examples

Example 1 : Measurement of brain tissue pulsations associated with the cardiac cycle

[0137] PME MRI was implemented and evaluated on a 3 T Siemens Prisma (Siemens Healthineers, Erlangen, Germany) equipped with a 32-channel head coil. Scans were performed on healthy volunteers under an IRB-approved protocol. Anatomical localizer scans were performed with magnetized prepared rapid gradient echo (MPRAGE) at 1 mm isotropic resolution. A pulse-oximeter signal, recorded with a BIOP AC MP150 system (Biopac, Goleta, CA, USA), served as cardiac signal for the analysis of brain pulsations. The PME implementation evaluated here used an echo-planar imaging (EPI) readout after a crusher gradient and used 2-slice excitation pulse sequences analogous to those shown in FIG. 2.

[0138] PME was accomplished by combining a slice selection gradient with a frequency modulated pulse similar to that used in simultaneous multi-slice (SMS) MRI, as explained above with respect to FIG. 10B. To avoid having slices at the start of the slice stack (early in the slice scanning protocol) not experiencing the full excitation sequence, the PME pattern at the end of the stack was adjusted to properly excite slices at the start of the stack (see below), in preparation for repeat acquisition of the slice stack. In other words, the PME pattern was cyclical in the slice direction.

[0139] A 2-slice excitation generating an SE combined with strong crusher gradients of 150 ms-mT-m^-1 was implemented to create sensitivity to tissue motion in three orthogonal directions. This was achieved with a crusher duration of 2.9 ms. For a sequential slice scanning protocol, this would require a [180°, 90°] PME pattern. However, in order to facilitate incorporation of lipid suppression, an interleaved slice scanning protocol was used, combined with alternating the slice frequency offset and slice select gradient polarity on subsequent excitations. This required inserting “null” excitations, leading to a [180°, 0°, 90°] PME spatial pattern. For this purpose, a 4 ms long PME pulse with time-bandwidth product (TBWP) of 4 was used for the 90° pulse, and TBWP 5.2 for the 180° pulse, thus yielding a 180° pulse slice thickness 30% larger than for the 90° pulse.

[0140] Image readout was performed with a custom single shot EPI sequence with 2 mm isotropic resolution using rate 3 SENSE acceleration. Forty-nine 2 mm thick slices were collected using 63 ms TE and a slice TR of 40 ms (1960 ms volume repetition time). Cyclical excitation involved shifting the leading edge of the (90° pulse) excitation pattern for slice 49 (end of slice stack) to the start of the stack (slices 1). Multiple (N=200) repetitions were collected to abundantly sample the full range of possible delays between cardiac systole, and the displacement-sensitive period of each acquired slice. This resulted in a total scan time of 392 s (6m32s) for each of three experiments with orthogonal crusher orientations.

[0141] Data were analyzed with IDL v8.8.3 (Harris Geospatial, Melbourne, FL, USA). EPI images were aligned to AC-PC line based on conventional (non-PME) 3D MPRAGE anatomical localizers. EPI data were analyzed in complex format in order to quantify the phase shifts induced by cardiac-cycle associated tissue pulsations. First, for each slice independently, the 50% of the volumes that show the smallest phase change over repetitions were identified, thus identifying a subset of data minimally affected by cardiac pulsatile effects. Voxel-wise trends (up to 4th order polynomial) in the temporal phase evolution were then computed on these data. After spatial smoothing (Gaussian, 4 voxels) of these fitted trends, they were removed from all time series complex data. This included temporal mean phase, e.g. originating from coil phase and Bo inhomogeneity. Subsequently, images were resorted according to phase in the cardiac cycle based on ECG timing estimated from the pulse-oximeter signal. For this purpose, the latter was time shifted by -250 ms to provide an approximate correction for its delay with ECG. Timing to the nearest subsequent heartbeat was also extracted from the pulse-oximeter measurement, which allowed inclusion of a second copy of the data in this sorted set, using negative latency values. Sorted data were compared to a convolved (Gaussian weighting function, width chosen to achieve averaging over approximately 10 samples) version of the sorted data, based on which outliers were identified as exceeding 2.5 SD in the difference between filtered and unfiltered sorted data. Data were thereafter again convolved with a Gaussian weighting function for averaging purposes, typically with a full width at half maximum of 0.1 s.

[0142] Sample results for PME-based measurement of brain tissue pulsations with the cardiac cycle are shown in FIG. 12, which shows tissue displacement with cardiac cycle measured with PME SE with pulse sequences according to FIG. 2. A single slice out of the 49-slice stack is shown. As appreciated, FIG. 12 shows cardiac-timing dependent displacement, in pm per 40 ms, as a function of displacement encoding direction. The total displacement (bottom row) is calculated as the root-mean squares combination of the three encoding directions. The results of FIG. 12 show good qualitative correspondence to previous reports. A centropetal displacement following systole was observed that reached 20 pm for each 40 ms TR and lasted -200 ms, leading to an estimated total displacement of 100 pm. The time axis is in seconds relative to cardiac event after correcting for a 250 ms PPG latency.

Example 2: Measurement of water diffusion with stimulated echo acquisition

[0143] Interleaved two- and 3-slice PME implementations with pulse sequences analogous to FIGS. 2 and 6, respectively, were used to generate SE- and STE-based diffusion contrast. The two- slice SE excitation used the [180°, 0°, 90°] PME pattern described in Example 1 and incorporated the same lipid suppression strategy. Diffusion weighting was varied on successive repetitions of the slice acquisition protocol according to a twelve-direction cycling scheme and two b-values (b=l 158 and b=6 s/mm'²). This cycling was repeated multiple times (n=30 for low-b and n=180 for high-b) to evaluate the required number of averages to derive parametric maps. Crusher durations of 12.44 ms were used with amplitudes of 45.8 mT/m'¹ and 3.2 mT/m’¹ for high- and low-b diffusion weightings, respectively. Other parameters included 84 ms TE (center of the EPI acquisition window), 52 ms slice TR, 2444 ms volume repetition time, isotropic 2 mm resolution, 30.6 ms EPI readout duration, and 47 axial-oblique slices along AC -PC direction. For comparison, conventional (no PME) DW MRI was performed as well using identical EPI readout and TE=73.5 ms, TR=4500 ms (95.7 ms slice TR). Lower spatial resolution acquisitions were performed as well: 1) a 2.5 mm isotropic resolution version with 1133 s/mm'² high- and 3 s/mm'² low-b values, and 71.2 ms TE; 2021 ms TR (PME SE) versus 65.0 ms TE; 3900 ms TR (conventional SE) and 2) a 45-slice 3 mm version with 1137 s/mm'² high- and 2 s/mm'² low-b value, and 66.5 ms TE; 1800 ms TR (PME SE) versus 60.0 ms TE; 3800 ms TR (conventional SE). For the 3-mm resolution scan 6 diffusion directions were used. Other parameters were as listed above.

[0144] A 3 -slice PME excitation was used to evaluate STE-based DW MRI with a TM period of 5 TRs. This utilized a 90°-90°-0-0-0-0-90° order of RF excitations and thus a [90°, 0°, 0°, 0°, 0°, 90°, 90°] PME pattern when using a sequential slice scanning protocol. As in Example 1, lipid suppression was facilitated by using an interleaved slice scanning protocol and inserting null excitations; this led to a [90°, 0° (9x), 90°, 0°, 90°] PME pattern. EPI acquisitions were performed at 2, 2.5 and 3 mm³ resolution as above, with crusher duration and amplitude of 5.4, 6.0 and 6.3 ms and 45.4, 45.6 and 45.7 mT/m respectively. Other parameters: TE=68.3, 55.3, 50.4 ms, TR=2068, 1645, 1376 ms, respectively. Again, cyclical excitation was performed to avoid incomplete excitations at the beginning of the slice stack. A six-direction diffusion weighting scheme was used. In contrast to the SE implementation, cycling through this scheme was performed on successive pulse intervals (not on repeated acquisitions of the entire slice stack in the acquisition protocol). This in effect resulted in the diffusion gradients simultaneously generating diffusion weighting as well as implicitly suppressing unwanted SE signals. On successive repetitions of the slice stack, a different diffusion weighting was used to ensure each slice was sampled at the full set of possible diffusion weightings. This was implicitly accomplished by choosing the number of diffusion weightings and number of slices to be not divisible by one another.

[0145] Data were analyzed with IDL v8.8.3 (Harris Geospatial, Melbourne, FL, USA). As in Example 1, EPI data were aligned to AC-PC line based on MPRAGE anatomical localizers. Diffusion data was analyzed based on EPI magnitude images. Mean diffusivity (MD), fractional anisotropy (FA), and fiber orientation maps were generated using “An introduction to diffusion tensor image analysis” Lauren J O'Donnell, Carl-Fredrik Westin, Neurosurg Clin N Am 2011 :22 pl 85, the entire contents of which are hereby incorporated by reference for all purposes.

[0146] Sample results for PME-based measurement of water diffusion are shown in FIG. 13 for a single slice out of the 47 or 45-slice stack. The top row of images includes 3.0 mm isotropic resolution, the center row 2.5 mm isotropic resolution and the bottom row of images includes 2.0 mm isotropic resolution. At 3 mm resolution, SE PME MRI has sufficient sensitivity to allow acquisition of orientation maps in as little as 13 seconds, combining six b=l 137 maps + one b=2 map, each acquired with 1800 ms volume TR. Conventional SE MRI requires about 27 s for this resolution. High quality maps of MD, FA, and diffusion direction were obtained for both STE and SE approaches that showed good qualitative correspondence to conventional (non-PME) technology. It is to be appreciated that other than the image labeled MP -RAGE, each image of FIG. 13 was generated in color with different colors representing different diffusion directions but is shown in grayscale in FIG. 13.

Example 3 : measurement of SNR and other parameters of scans combining SMS and PME [0147] Scans were performed on four healthy young volunteers (3 female, 1 male) under an IRB-approved protocol. All measurements were performed on a Siemens 3T Prisma scanner (Erlangen, Germany) with a 32-channel receive head-coil. The protocol included five scans, all with PME-SE and a 2D EPI acquisition with SENSE rate 2. Diffusion weighting was applied in 12 directions equally distributed on a sphere, with a low b of 16-20 s/mm² and a high-b of 1012- 1060 s/mm². The RF pulse duration was 5.2 ms (for a single excitation) with a bandwidth-time- product of four. The RF time shift was 0.9 ms. The scans were: 1) SMS-PME as described above with respect to FIGS. 9 and 11 at 1.75 mm isotropic resolution with 62 slices in 1932 ms; 2) same as scan-1, but with PME only and a volume TR (vTR) of 3864 ms; 3) SMS-PME at 2 mm isotropic resolution with 56 slices in 1624 ms; 4) as scan-3, but PME only and a vTR of 3248 ms; and 5) as scan-4 without the RF time shift. Total scan time was 9m36s for 1.75 mm³ resolution, and 5m37s for 2 mm³ resolution. The total number of low-b averages was 43, 22, 29, 15, and 15 for scans 1- 5 respectively. The number of averages for each high-b direction was 21, 10, 14, 7, and 7 for scans 1-5.

[0148] SENSE reconstruction was used for both in-plane and through-plane (the latter for SMS) unaliasing. Voxel-wise mean diffusivity (MD), fractional anisotropy (FA), and fiber orientation maps were generated from the fitted diffusion tensor matrix.

[0149] The signal-to-noise (SNR) ratio calculation was performed with image noise level derived from data acquired in a shot without RF, in combination with the SENSE matrix. Temporal signal stability (tSNR) was calculated over low-b data and each high-b direction independently. Both SNR and tSNR were converted to SNR/unit-time by dividing by j vTR.

[0150] FIGS. 14A and 14B show both the SNR and tSNR for PME only (referred as SMS1- PME in FIG. 14A) and SMS-PME (referred to as SMS2-PME in FIG. 14A) at 1.75 mm³ resolution, showing very similar results. Specifically, SNR and tSNR maps were generated as described above and histograms were generated depicting SNR and tSNR per unit time averaged over all 12 diffusion directions for the four subjects, and FIG. 14A illustrates the histograms for the high b- value and FIG. 14B for the low b-value. FIG. 15 shows similar findings for the 2 mm³ data, comparing tSNR for SMS2-PME and SMS 1 -PME with and without pulse time-shift. Overall, the time-shifted RF pulse did not introduce additional SNR penalty but reduced peak amplitude by 26% and 51% for SMS1-PME and SMS2-PME respectively. FIG. 16 shows mean diffusivity (top row), fractional anisotropy (middle row), and colormap (bottom row) for PME diffusion data acquired with (right three columns) and without SMS (left three columns). The SMS2 (e.g., SMS- PME) had no impact on distortion and image quality of the diffusion measures. It is to be appreciated that the colormap images showing dominant diffusion directions was originally generated in color but is shown in grayscale in FIG. 16.

[0151] Thus, PME and SMS may be combined to accelerate SE-based diffusion MRI. An additional twofold acceleration was achieved with SMS2-PME over PME while maintaining the tSNR per unit time. An V2 improvement in tSNR per unit time was not observed with SMS2-PME expected from the increased number of averages. This may be attributed to increased saturation and a modest (-10%) g-factor penalty that was observed with SMS2-PME. This additional acceleration is beneficial for high angular resolution diffusion imaging and reducing sensitivity to patient motion. In addition, tSNR gain is expected in applications where vTR is longer, which would reduce the penalty from saturation effects.

[0152] This disclosure provides support for a method for imaging a patient according to a patterned multi-slice excitation (PME) magnetic resonance imaging (MRI) slice acquisition protocol, the method comprising: during a first pulse interval, applying a first radiofrequency (RF) pulse together with a first slice selection gradient to excite a first slice and a second slice, the second slice following the first slice in the slice acquisition protocol, and acquiring MR signals of only the first slice; and during a second pulse interval following the first pulse interval, applying a second RF pulse together with a second slice selection gradient to excite the second slice and a third slice, the third slice following the second slice in the slice acquisition protocol, and acquiring MR signals of only the second slice. In a first example of the method, the method further comprises, during the first pulse interval, applying a first crusher gradient following the first RF pulse and before acquiring the MR signals of only the first slice, and during the second pulse interval, applying a second crusher gradient following the second RF pulse and before acquiring the MR signals of only the second slice. In a second example of the method, optionally including the first example, the method further comprises, during the first pulse interval, applying a first crusher gradient following the first RF pulse and after acquiring the MR signals of only the first slice, and during the second pulse interval, applying a second crusher gradient following the second RF pulse and before acquiring the MR signals of only the second slice. In a third example of the method, optionally including one or both of the first and second examples, the first RF pulse also excites a third slice and the second RF pulse also excites a fourth slice, and further comprising during a third pulse interval following the second pulse interval, applying a third RF pulse to excite the third slice and the fourth slice, the fourth slice following the third slice in the slice acquisition protocol, and acquiring MR signals of only the third slice. In a fourth example of the method, optionally including one or more or each of the first through third examples, each RF pulse excites four or more slices and MR signals from only a single slice are acquired each pulse interval. In a fifth example of the method, optionally including one or more or each of the first through fourth examples, each RF pulse excites four or more slices and MR signals from multiple slices are acquired each pulse interval. In a sixth example of the method, optionally including one or more or each of the first through fifth examples, MR signals are acquired between the first RF pulse and the first crusher gradient. In a seventh example of the method, optionally including one or more or each of the first through sixth examples, the method further comprises generating one or more images from the MR signals acquired for each of the excited slices.

[0153] This disclosure also provides support for a method for imaging a patient according to a patterned multi-slice excitation (PME) magnetic resonance imaging (MRI) slice acquisition protocol, the method comprising: during a first repetition of a pulse sequence of the slice acquisition protocol: controlling a radiofrequency (RF) coil to emit an RF pulse to excite a target slice to be imaged and one or more additional slices to be imaged in one or more subsequent repetitions of the pulse sequence; controlling a gradient coil system to generate a slice selection gradient during the RF pulse; controlling the gradient coil system to generate a crusher gradient following the RF pulse; and after the crusher gradient is complete, acquiring MR signals of only the target slice. In a first example of the method, the RF pulse, combined with a prior RF pulse, generates a spin echo at the target slice. In a second example of the method, optionally including the first example, the RF pulse, combined with two prior RF pulses, generates a stimulated echo at the target slice. In a third example of the method, optionally including one or both of the first and second examples, the RF pulse, combined with one or more prior RF pulses, generates a gradient echo with T1 weighting at the target slice. In a fourth example of the method, optionally including one or more or each of the first through third examples, the RF pulse, combined with two prior RF pulses, generates a spin echo with T1 weighting at the target slice. In a fifth example of the method, optionally including one or more or each of the first through fourth examples, the RF pulse, combined with three prior RF pulses, generates a stimulated echo with T1 weighting at the target slice. In a sixth example of the method, optionally including one or more or each of the first through fifth examples, the RF pulse excites the target slice and an adjacent slice in the slice acquisition protocol. In a seventh example of the method, optionally including one or more or each of the first through sixth examples, the RF pulse excites the target slice and a non-adjacent slice in the slice acquisition protocol. In an eighth example of the method, optionally including one or more or each of the first through seventh examples, the RF pulse excites the target slice, a first adjacent slice, and a second adjacent slice, wherein the target slice, the first adjacent slice, and the second adjacent slice are consecutive slices in the slice acquisition protocol. In a ninth example of the method, optionally including one or more or each of the first through eighth examples, the RF pulse excites the target slice and two additional slices, wherein at least one intervening slice among the target slice and the two additional slices in the slice acquisition protocol or between the two additional slices in the slice acquisition protocol is not excited by the RF pulse. In a tenth example of the method, optionally including one or more or each of the first through ninth examples, the RF pulse excites one or more additional slices to generate Tl-weighting. In an eleventh example of the method, optionally including one or more or each of the first through tenth examples, the RF pulse excites multiple slices of unequal width. In a twelfth example of the method, optionally including one or more or each of the first through eleventh examples, a duration and/or an amplitude of the crusher gradient is varied across repetitions of the pulse sequence to vary sensitivity to water diffusion or tissue motion. In a thirteenth example of the method, optionally including one or more or each of the first through twelfth examples, a duration and/or an amplitude of the crusher gradient is varied across repetitions of the slice acquisition protocol to vary sensitivity to water diffusion or tissue motion.

[0154] This disclosure also provides support for a magnetic resonance imaging (MRI) system configured to image a region of interest (ROI) of a patient, the MRI system comprising: a set of gradient coils configured to provide magnetic gradients along respective orthogonal directions; a radiofrequency (RF) system configured to transmit RF pulses and receive MR signals representing the ROI; and a controller configured to control the set of gradient coils and the RF system to perform a plurality of patterned multi-slice excitation (PME) pulse sequences, each PME pulse sequence configured to excite at least two slices of the ROI and acquire MR signals of a target slice of the at least two slices. In a first example of the system, each PME pulse sequence includes an excitation segment where the RF system is controlled to emit at least one RF pulse to excite at least two slices, an acquisition segment where the set of gradient coils is controlled to acquire the MR signals of the target slice, and a crusher segment that is performed intermediate the excitation segment and the acquisition segment or after the excitation segment and the acquisition segment, where the set of gradient coils is controlled to generate a crusher gradient during the crusher segment. In a second example of the system, optionally including the first example, the controller is configured to modulate an amplitude of the crusher gradient across multiple PME pulse sequences. In a third example of the system, optionally one or both of the first and second examples, the at least one RF pulse includes two RF pulses performed sequentially in time to excite at least two slices each with arbitrary and independent thickness and flip angle and to allow shift of a refocusing point of a spin echo or a stimulated echo signal generated by the two RF pulses. In a fourth example of the system, optionally including one or more or each of the first through third examples, the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice and a second slice of the ROI and acquire MR signals of the first slice; and a second PME pulse sequence configured to excite the second slice and a third slice of the ROI and acquire MR signals of the second slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence. In a fifth example of the system, optionally including one or more or each of the first through fourth examples, the first PME pulse sequence also excites the third slice, the second PME pulse sequence also excites a fourth slice, and the plurality of PME pulse sequences further includes a third PME pulse sequence configured to excite the third slice, the fourth slice, and a fifth slice of the ROI and acquire MR signals of the third slice, wherein the third PME pulse sequence is performed immediately after the second PME pulse sequence. In a sixth example of the system, optionally including one or more or each of the first through fifth examples, the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice, a third slice, and a fourth slice of the ROI and acquire MR signals of the first slice; a second PME pulse sequence configured to excite a second slice, the fourth slice, and a fifth slice of the ROI and acquire MR signals of the second slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence; a third PME pulse sequence configured to excite the third slice, the fifth slice, and a sixth slice of the ROI and acquire MR signals of the third slice, wherein the third PME pulse sequence is performed immediately after the second PME pulse sequence; and a fourth PME pulse sequence configured to excite the fourth slice, the sixth slice, and a seventh slice of the ROI and acquire MR signals of the fourth slice, wherein the fourth PME pulse sequence is performed immediately after the third PME pulse sequence. In a seventh example of the system, optionally including one or more or each of the first through sixth examples, the plurality of PME pulse sequences is performed to acquire MR signals of a set of slices of the ROI, wherein each PME pulse sequence of the plurality of PME pulse sequences includes a first crusher gradient, and wherein the controller is further configured to control the set of gradient coils and the RF system to perform a second plurality of PME pulse sequences to acquire additional MR signals of the set of slices of the ROI, each PME pulse sequence of the second plurality of PME pulse sequences including a second crusher gradient having a different amplitude and/or duration than the first crusher gradient. In an eighth example of the system, optionally including one or more or each of the first through seventh examples, the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice, a second slice, a first parallel slice, and a second parallel slice of the ROI and acquire MR signals of the first slice and the first parallel slice, wherein the first slice and the first parallel slice are spatially separated from each other; and a second PME pulse sequence configured to excite the second slice, a third slice, the second parallel slice, and a third parallel slice of the ROI and acquire MR signals of the second slice and the second parallel slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence. In a ninth example of the method, optionally including one or more or each of the first through eighth examples, the first slice and the second slice are spatially adjacent to each other, the first parallel slice and the second parallel slice are spatially adjacent each other, and the third slice is spatially adjacent the second slice and the third parallel slice is spatially adjacent the second parallel slice.

[0155] As used herein, an element or step recited in the singular and preceded with the word “a” or “an” should be understood as not excluding plural of said elements or steps, unless such exclusion is explicitly stated. Furthermore, references to “one embodiment” of the present invention are not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features. Moreover, unless explicitly stated to the contrary, embodiments “comprising,” “including,” or “having” an element or a plurality of elements having a particular property may include additional such elements not having that property. The terms “including” and “in which” are used as the plain-language equivalents of the respective terms “comprising” and “wherein.” Moreover, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements or a particular positional order on their objects.

[0156] This written description uses examples to disclose the invention, including the best mode, and also to enable a person of ordinary skill in the relevant art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to those of ordinary skill in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

Claims

1. A method for imaging a patient according to a patterned multi-slice excitation (PME) magnetic resonance imaging (MRI) slice acquisition protocol, the method comprising: during a first pulse interval, applying a first radiofrequency (RF) pulse together with a first slice selection gradient to excite a first slice and a second slice, the second slice following the first slice in the slice acquisition protocol, and acquiring MR signals of only the first slice; and during a second pulse interval following the first pulse interval, applying a second RF pulse together with a second slice selection gradient to excite the second slice and a third slice, the third slice following the second slice in the slice acquisition protocol, and acquiring MR signals of only the second slice.

2. The method of claim 1, further comprising, during the first pulse interval, applying a first crusher gradient following the first RF pulse and before acquiring the MR signals of only the first slice, and during the second pulse interval, applying a second crusher gradient following the second RF pulse and before acquiring the MR signals of only the second slice.

3. The method of claim 1, further comprising, during the first pulse interval, applying a first crusher gradient following the first RF pulse and after acquiring the MR signals of only the first slice, and during the second pulse interval, applying a second crusher gradient following the second RF pulse and before acquiring the MR signals of only the second slice.

4. The method of claim 1, wherein the first RF pulse also excites the third slice and the second RF pulse also excites a fourth slice, and further comprising during a third pulse interval following the second pulse interval, applying a third RF pulse to excite the third slice and the fourth slice, the fourth slice following the third slice in the slice acquisition protocol, and acquiring MR signals of only the third slice.

5. The method of claim 1, wherein each RF pulse excites four or more slices and MR signals from only a single slice are acquired each pulse interval.

6. The method of claim 1, wherein each RF pulse excites four or more slices and MR signals from multiple slices are acquired each pulse interval.

7. The method of claim 1, wherein MR signals are acquired between the first RF pulse and a first crusher gradient.

8. The method of claim 1, further comprising generating one or more images from the MR signals acquired for each of the excited slices.

9. A magnetic resonance imaging (MRI) system configured to image a region of interest (ROI) of a patient, the MRI system comprising: a set of gradient coils configured to provide magnetic gradients along respective orthogonal directions; a radiofrequency (RF) system configured to transmit RF pulses and receive MR signals representing the ROI; and a controller configured to control the set of gradient coils and the RF system to perform a plurality of patterned multi-slice excitation (PME) pulse sequences, each PME pulse sequence configured to excite at least two slices of the ROI and acquire MR signals of a target slice of the at least two slices.

10. The MRI system of claim 9, wherein each PME pulse sequence includes an excitation segment where the RF system is controlled to emit at least one RF pulse to excite at least two slices, an acquisition segment where the set of gradient coils is controlled to acquire the MR signals of the target slice, and a crusher segment that is performed intermediate the excitation segment and the acquisition segment or after the excitation segment and the acquisition segment, where the set of gradient coils is controlled to generate a crusher gradient during the crusher segment.

11. The MRI system of claim 10, wherein the controller is configured to modulate an amplitude of the crusher gradient across multiple PME pulse sequences.

12. The MRI system of claim 10, wherein the at least one RF pulse includes two RF pulses performed sequentially in time to excite at least two slices each with arbitrary and independent thickness and flip angle and to allow shift of a refocusing point of a spin echo or a stimulated echo signal generated by the two RF pulses.

13. The MRI system of claim 9, wherein the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice and a second slice of the ROI and acquire MR signals of the first slice; and a second PME pulse sequence configured to excite the second slice and a third slice of the ROI and acquire MR signals of the second slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence.

14. The MRI system of claim 13, wherein the first PME pulse sequence also excites the third slice, the second PME pulse sequence also excites a fourth slice, and the plurality of PME pulse sequences further includes a third PME pulse sequence configured to excite the third slice, the fourth slice, and a fifth slice of the ROI and acquire MR signals of the third slice, wherein the third PME pulse sequence is performed immediately after the second PME pulse sequence.

15. The MRI system of claim 9, wherein the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice, a third slice, and a fourth slice of the ROI and acquire MR signals of the first slice; a second PME pulse sequence configured to excite a second slice, the fourth slice, and a fifth slice of the ROI and acquire MR signals of the second slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence; a third PME pulse sequence configured to excite the third slice, the fifth slice, and a sixth slice of the ROI and acquire MR signals of the third slice, wherein the third PME pulse sequence is performed immediately after the second PME pulse sequence; and a fourth PME pulse sequence configured to excite the fourth slice, the sixth slice, and a seventh slice of the ROI and acquire MR signals of the fourth slice, wherein the fourth PME pulse sequence is performed immediately after the third PME pulse sequence.

16. The MRI system of claim 9, wherein the plurality of PME pulse sequences is performed to acquire MR signals of a set of slices of the ROI, wherein each PME pulse sequence of the plurality of PME pulse sequences includes a first crusher gradient, and wherein the controller is further configured to control the set of gradient coils and the RF system to perform a second plurality of PME pulse sequences to acquire additional MR signals of the set of slices of the ROI, each PME pulse sequence of the second plurality of PME pulse sequences including a second crusher gradient having a different amplitude and/or duration than the first crusher gradient.

17. The MRI system of claim 9, wherein the plurality of PME pulse sequences includes: a first PME pulse sequence configured to excite a first slice, a second slice, a first parallel slice, and a second parallel slice of the ROI and acquire MR signals of the first slice and the first parallel slice, wherein the first slice and the first parallel slice are spatially separated from each other; and a second PME pulse sequence configured to excite the second slice, a third slice, the second parallel slice, and a third parallel slice of the ROI and acquire MR signals of the second slice and the second parallel slice, wherein the second PME pulse sequence is performed immediately after the first PME pulse sequence.

18. The MRI system of claim 17, wherein the first slice and the second slice are spatially adjacent to each other, the first parallel slice and the second parallel slice are spatially adjacent each other, and the third slice is spatially adjacent the second slice and the third parallel slice is spatially adjacent the second parallel slice.