U.S. patent application number 16/394786 was filed with the patent office on 2020-10-29 for mri system and method for detection and correction of patient motion.
The applicant listed for this patent is GE PRECISION HEALTHCARE LLC. Invention is credited to Sangtae Ahn, Rafael Shmuel Brada, Christopher Judson Hardy, Isabelle Heukensfeldt Jansen, Itzik Malkiel, Michael Rotman, Ron Wein.
Application Number | 20200337592 16/394786 |
Document ID | / |
Family ID | 1000005147391 |
Filed Date | 2020-10-29 |
View All Diagrams
United States Patent
Application |
20200337592 |
Kind Code |
A1 |
Brada; Rafael Shmuel ; et
al. |
October 29, 2020 |
MRI SYSTEM AND METHOD FOR DETECTION AND CORRECTION OF PATIENT
MOTION
Abstract
A system and method for detecting, timing, and adapting to
patient motion during an MR scan includes using the inconsistencies
between calculated images from different coil-array elements to
detect the presence of patient motion and, together with the
k-space scan-order information, determine the timing of the motion
during the scan. Once the timing is known, various actions may be
taken, including restarting the scan, reacquiring those portions of
k-space acquired before the movement, or correcting for the motion
using the existing data and reconstructing a motion-corrected image
from the data.
Inventors: |
Brada; Rafael Shmuel;
(Hod-Hasharon, IL) ; Hardy; Christopher Judson;
(Schenectady, NY) ; Ahn; Sangtae; (Guilderland,
NY) ; Heukensfeldt Jansen; Isabelle; (Schenectady,
NY) ; Malkiel; Itzik; (Givatayim, IL) ;
Rotman; Michael; (Petach-Tikva, IL) ; Wein; Ron;
(Ramat Hasharon, IL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
GE PRECISION HEALTHCARE LLC |
Wauwatosa |
WI |
US |
|
|
Family ID: |
1000005147391 |
Appl. No.: |
16/394786 |
Filed: |
April 25, 2019 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01R 33/4818 20130101;
A61B 5/0555 20130101; G01R 33/56509 20130101; G16H 30/40 20180101;
G01R 33/5608 20130101 |
International
Class: |
A61B 5/055 20060101
A61B005/055; G01R 33/565 20060101 G01R033/565; G01R 33/48 20060101
G01R033/48; G01R 33/56 20060101 G01R033/56 |
Claims
1. A magnetic resonance imaging method comprising: generating
intensity-corrected single-coil images from raw magnetic resonance
(MR) data of an imaged subject, wherein the raw MR data comprise
data collected by each coil of a plurality of coils in a receiving
coil array of an MR system, and wherein the raw MR data are
associated with a scan order used during acquisition thereof, the
scan order having a plurality of time steps in which k-space is
filled in a predetermined manner; identifying inconsistencies among
the intensity-corrected single-coil images; calculating, for at
least one time step of the scan order, a motion score using the
scan order and the inconsistencies to identify timing associated
with motion occurring during the acquisition; and performing a
corrective action based at least on the timing associated with the
motion to ameliorate the effects of the motion on an MR image
produced using at least a portion of the raw MR data.
2. The method of claim 1, wherein identifying inconsistencies among
the intensity-corrected single-coil images comprises: applying a
Fourier transform to the intensity-corrected single-coil images to
transform image data of the intensity-corrected single-coil images
into k-space data; and determining differences between the k-space
data of pairs of intensity-corrected single-coil images.
3. The method of claim 2, wherein determining differences between
the k-space data of pairs of intensity-corrected single-coil images
comprises taking an absolute difference between the Fourier
transform of two intensity-corrected single-coil images.
4. The method of claim 2, wherein determining differences between
the k-space data of pairs of intensity-corrected single-coil images
comprises generating a k-space difference map representing a
difference between the k-space data for a pair of the
intensity-corrected single-coil images.
5. The method of claim 4, wherein identifying inconsistencies
between the intensity-corrected single-coil images comprises
identifying locations of relatively bright columns or rows in the
k-space difference map.
6. The method of claim 4, wherein identifying inconsistencies among
the intensity-corrected single-coil images comprises projecting the
k-space difference map along the readout direction onto an axis to
produce a plot of integrated k-space difference as a function of
phase-encoding index, and identifying peaks that are relatively
larger than others, and wherein a size of a peak is proportional to
a degree of inconsistency.
7. The method of claim 6, wherein calculating, for the at least one
time step of the scan order, the motion score comprises integrating
over all previous time steps the peaks of the plot of integrated
k-space difference.
8. The method of claim 1 wherein k-space is undersampled to
accelerate imaging, by use of parallel imaging, compressed sensing,
or some other method.
9. The method of claim 1, comprising, for the at least one time
step of the scan order, zero-filling regions of k-space in the raw
MR data that have not yet been sampled such that the raw MR data
comprise regions of k-space that are zero-filled.
10. The method of claim 9, comprising determining differences
between k-space data of pairs of intensity-corrected single-coil
images by generating a k-space difference map by taking an absolute
difference between the Fourier transform of two intensity-corrected
single-coil images using only the zero-filled k-space data
collected up to a particular time step.
11. The method of claim 10, wherein identifying inconsistencies
between the intensity-corrected single-coil images comprises
identifying peaks in the calculated k-space difference by
calculating a 1D projection of the absolute value along rows or
columns of k-space, and detecting peaks.
12. The method of claim 11, wherein detecting peaks comprises
calculating the negative of the second derivative of the signal at
the time step.
13. The method of claim 12, wherein calculating the motion score,
for the at least one time step of the scan order, comprises:
removing peaks that are due to borders between sampled and
unsampled locations of k-space according to scan order information;
and calculating the motion score by taking a sum of the remaining
peaks.
14. The method of claim 13, wherein the motion score is calculated
using the amplitude of the removed peaks to normalize the remaining
peaks.
15. The method of claim 13, comprising applying a threshold to the
calculated motion score to determine whether the calculated motion
score is indicative of motion.
16. The method of claim 1, wherein performing the corrective action
comprises reacquiring portions of k-space acquired before
movement.
17. The method of claim 1, wherein performing the corrective action
comprises correcting for the motion using the existing raw MR data
using a deep-learning neural network or iterative optimization.
18. A magnetic resonance imaging (MRI) method comprising:
generating intensity-corrected single-coil images from raw magnetic
resonance (MR) data of an imaged subject, wherein the raw MR data
comprise data collected by each coil of a plurality of coils in a
receiving coil array of an MRI system; identifying inconsistencies
among the intensity-corrected single-coil images; identifying
whether motion occurred during acquisition of the raw MR data based
at least on the inconsistencies; and performing further operations
of the MRI system in response to determining that motion occurred
during the acquisition.
19. The method of claim 18, wherein performing further operations
of the MRI system in response to determining that motion occurred
during the acquisition comprises restarting an MR scan to
re-acquire raw MR data of the imaged subject.
20. The method of claim 18, wherein identifying inconsistencies
among the intensity-corrected single-coil images comprises:
applying a Fourier transform to the intensity-corrected single-coil
images to transform image data of the intensity-corrected
single-coil images into k-space data; determining differences
between the k-space data of pairs of intensity-corrected
single-coil images; calculating motion scores for at least some
time steps of a scan order associated with acquisition of the raw
MR data based at least on the determined differences, the scan
order defining phase encode as a function of time step; and
identifying presence and timing of motion based on the calculated
motion scores.
21. The method of claim 20, wherein performing further operations
of the MRI system in response to determining that motion occurred
during the acquisition comprises reacquiring portions of k-space
acquired before movement.
22. A magnetic resonance imaging (MRI) method comprising: obtaining
raw magnetic resonance (MR) data of a subject, wherein the raw MR
data comprises data collected by each coil of a plurality of coils
in a receiving coil array of an MRI system, and wherein the raw MR
data is associated with a scan order used during acquisition
thereof, the scan order having a plurality of time steps in which
k-space is filled in a predetermined manner; using inconsistencies
among calculated intensity-corrected single-coil images produced
from the raw MR data to detect motion of the subject and, together
with the scan order, determine the timing of the motion during the
acquisition; and performing image reconstruction using the timing
of the motion to generate a single motion-corrected image.
23. The method of claim 22 comprising: dividing the data collected
by each coil of the plurality of coils coil data into pre-motion
and post-motion datasets based on the timing; reconstructing images
for each coil from zero-filled pre-motion k-space dataset and
zero-filled post-motion k-space dataset; and providing the two sets
of images as inputs into a deep-learning neural network to
reconstruct the single motion-corrected image.
Description
BACKGROUND
[0001] In general, magnetic resonance imaging (MRI) examinations
are based on the interactions among a primary magnetic field, a
radiofrequency (RF) magnetic field and time varying magnetic
gradient fields with gyromagnetic material having nuclear spins
within a subject of interest, such as a patient. Certain
gyromagnetic materials, such as hydrogen nuclei in water molecules,
have characteristic behaviors in response to external magnetic
fields. The precession of spins of these nuclei can be influenced
by manipulation of the fields to produce RF signals that can be
detected, processed, and used to reconstruct a useful image.
[0002] Patient motion is one of the biggest sources of inefficiency
in clinical MRI, often requiring re-scans or even second visits by
the patient. In particular, patient motion can cause blurriness,
artifacts, and other inconsistencies in MR images. Certain
approaches to correct motion require either some sort of hardware
for monitoring the motion (adding to cost and patient setup time),
or navigator sequences (which take time away from the imaging
sequence). Accordingly, a need exists for improved methods for data
acquisition and reconstruction in magnetic resonance imaging
techniques that are sensitive to patient motion.
BRIEF DESCRIPTION
[0003] In one embodiment, a magnetic resonance imaging method
includes generating intensity-corrected single-coil images from raw
magnetic resonance (MR) data of an imaged subject. The raw MR data
include data collected by each coil of a plurality of coils in a
receiving coil array of an MR system, and the raw MR data are
associated with a scan order used during acquisition thereof, the
scan order having a plurality of time steps in which k-space is
filled in a predetermined manner. The method also includes
identifying inconsistencies among the intensity-corrected
single-coil images, and calculating, for at least one time step of
the scan order, a motion score using the scan order and the
inconsistencies to identify timing associated with motion occurring
during the acquisition. The method further includes performing a
corrective action based at least on the timing associated with the
motion to ameliorate the effects of the motion on an MR image
produced using at least a portion of the raw MR data.
[0004] In another embodiment, a magnetic resonance imaging (MRI)
method includes generating intensity-corrected single-coil images
from raw magnetic resonance (MR) data of an imaged subject. The raw
MR data include data collected by each coil of a plurality of coils
in a receiving coil array of an MRI system. The method also
includes identifying inconsistencies among the intensity-corrected
single-coil images, identifying whether motion occurred during
acquisition of the raw MR data based at least on the
inconsistencies, and performing further operations of the MRI
system in response to determining that motion occurred during the
acquisition.
[0005] In a further embodiment, a magnetic resonance imaging (MRI)
method includes obtaining raw magnetic resonance (MR) data of a
subject, wherein the raw MR data include data collected by each
coil of a plurality of coils in a receiving coil array of an MRI
system. The raw MR data are associated with a scan order used
during acquisition thereof, the scan order having a plurality of
time steps in which k-space is filled in a predetermined manner.
The method also includes using inconsistencies between calculated
intensity-corrected single-coil images produced from the raw MR
data to detect the presence of motion of the subject and, together
with the scan order, determine the timing of the motion during the
acquisition. The method further includes performing image
reconstruction using the timing of the motion to generate a single
motion-corrected image.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] These and other features, aspects, and advantages of the
present invention will become better understood when the following
detailed description is read with reference to the accompanying
drawings in which like characters represent like parts throughout
the drawings, wherein:
[0007] FIG. 1 is a diagrammatic illustration of an embodiment of a
magnetic resonance imaging system configured to perform the data
acquisition, motion detection and scoring, and image reconstruction
described herein;
[0008] FIG. 2 is a process flow diagram of an embodiment of a
method 80 for identifying the occurrence/timing of motion after an
MR scan has completed;
[0009] FIG. 3 is a set of example intensity-corrected single-coil
images produced during the method of FIG. 2;
[0010] FIG. 4 is a map of the absolute value of the difference
between a Fourier transform of two intensity-corrected single-coil
images;
[0011] FIG. 5 is a projection of the difference map of FIG. 4 onto
an axis;
[0012] FIG. 6 is an example MRI scan order for a 256.times.256 FSE
image with echo-train length of 8;
[0013] FIG. 7 is an example plot of a motion score calculated for
each time step of an image;
[0014] FIG. 8 is a plot of calculated motion score in each image
for a 22-slice series of FSE images;
[0015] FIG. 9 is an image corresponding to the odd-numbered slices
from the 22-slice series of FSE images;
[0016] FIG. 10 is an image corresponding to the even-numbered
slices from the 22-slice series of FSE images;
[0017] FIG. 11 is a process flow diagram of an embodiment of a
method for identifying the occurrence/timing of motion while an MR
scan is in progress;
[0018] FIG. 12 is a schematic representation of the manner in which
k-space is zero-filled according to the method of FIG. 11;
[0019] FIG. 13 is a plot of the negative of the second derivative
of the signal corresponding to a calculated k-space difference;
[0020] FIG. 14 is an example plot of unexpected peaks produced
during the method of FIG. 11 when no motion has occurred;
[0021] FIG. 15 is an example plot of calculated motion score as a
function of time step;
[0022] FIG. 16 is a schematic of partially-filled k-space having
zero-filled sections, sections filled with data acquired before
motion occurred, and sections filled with data acquired after
motion occurred;
[0023] FIG. 17 is a plot of the negative of the second derivative
of the signal corresponding to the partially filled k-space;
[0024] FIG. 18 is a plot of unexpected peaks corresponding to
motion and after removal of peaks corresponding to known motion
states;
[0025] FIG. 19 is an example plot of calculated motion score as a
function of time step;
[0026] FIG. 20 is a plot, for a simulated data set of MR images,
comparing true simulated time of motion versus calculated time of
motion determined according to the method of FIG. 11;
[0027] FIG. 21 is an embodiment of an algorithm for performing a
scan, monitoring for motion during the scan, and aggregating motion
states when motion is detected;
[0028] FIG. 22 is an embodiment of an algorithm for performing a
scan, monitoring for motion during the scan, and aggregating a
final motion state when motion is detected;
[0029] FIG. 23 is an embodiment of an algorithm for performing a
scan, monitoring for motion during the scan, and adapting to the
motion during the scan when motion is detected;
[0030] FIG. 24 is an embodiment of a method of reconstructing a
motion-free image from motion-corrupted datasets; and
[0031] FIG. 25 is another embodiment of a method of reconstructing
a motion-free image from motion-corrupted datasets.
DETAILED DESCRIPTION
[0032] One or more specific embodiments will be described below. In
an effort to provide a concise description of these embodiments,
all features of an actual implementation may not be described in
the specification. It should be appreciated that in the development
of any such actual implementation, as in any engineering or design
project, numerous implementation-specific decisions must be made to
achieve the developers' specific goals, such as compliance with
system-related and business-related constraints, which may vary
from one implementation to another. Moreover, it should be
appreciated that such a development effort might be complex and
time consuming, but would nevertheless be a routine undertaking of
design, fabrication, and manufacture for those of ordinary skill
having the benefit of this disclosure.
[0033] When introducing elements of various embodiments of the
present invention, the articles "a," "an," "the," and "said" are
intended to mean that there are one or more of the elements. The
terms "comprising," "including," and "having" are intended to be
inclusive and mean that there may be additional elements other than
the listed elements. Furthermore, any numerical examples in the
following discussion are intended to be non-limiting, and thus
additional numerical values, ranges, and percentages are within the
scope of the disclosed embodiments.
[0034] As set forth above, patient motion is one of the biggest
sources of inefficiency in clinical MRI, often requiring re-scans
or even second visits by the patient. Research has shown that
patient motion can lead to repeated acquisition sequences in as
much as 20% of MRI exams. This results in significant annual losses
for every scanner as throughput is reduced.
[0035] Disclosed embodiments include a system and method for
detecting, timing, and adapting to patient motion during or after
an MR scan, without the need for external tracking hardware. This
uses the inconsistencies between calculated images from different
coil-array elements to detect the presence of patient motion and,
together with the k-space scan-order information, determines the
timing of the motion during the scan. Once the timing is known,
various actions can be taken, including restarting the scan,
reacquiring those portions of k-space acquired before the movement,
or correcting for the motion using the existing data. This
correction is done either using a deep-learning neural network or
an iterative optimization approach.
[0036] Disclosed embodiments also include an adaptive system for
detecting patient motion in real time during an MR scan without the
need for external monitoring devices or navigation, with the option
of adjusting scan parameters to compensate for inconsistent data.
Once motion is detected, the system can track multiple separate
sub-images to be combined into a motion-free image or can adjust
the scan to re-acquire sections of k-space taken before the motion
occurred.
[0037] An example system for performing the techniques described
herein is discussed with respect to FIG. 1. The embodiments
described herein may be performed by a magnetic resonance imaging
(MRI) system, wherein specific imaging routines (e.g., accelerated
imaging routines for MRI sequences) are initiated by a user (e.g.,
a radiologist). Further, the MRI system may perform data
acquisition, data correction, and image reconstruction.
Accordingly, referring to FIG. 1, a magnetic resonance imaging
system 10 is illustrated schematically as including a scanner 12, a
scanner control circuit 14, and a system control circuitry 16.
According to the embodiments described herein, the MRI system 10 is
generally configured to perform MR imaging, such as imaging
sequences with adaptive motion correction, various weighting
techniques, fluid attenuation techniques, perfusion techniques,
tensor imaging, and so on. System 10 additionally includes remote
access and storage systems or devices such as picture archiving and
communication systems (PACS) 18, or other devices such as
teleradiology equipment so that data acquired by the system 10 may
be accessed on- or off-site. In this way, acquired data may be
acquired, followed by on- or off-site processing and evaluation.
While the MRI system 10 may include any suitable scanner or
detector, in the illustrated embodiment, the system 10 includes a
full body scanner 12 having a housing 20 through which a bore 22 is
formed. A table 24 is moveable into the bore 22 to permit a patient
26 to be positioned therein for imaging selected anatomy within the
patient.
[0038] Scanner 12 includes a series of associated coils for
producing controlled magnetic fields for exciting the gyromagnetic
material within the anatomy of the subject being imaged.
Specifically, a primary magnet coil 28 is provided for generating a
primary magnetic field generally aligned with the bore 22. A series
of gradient coils 30, 32, and 34 permit controlled magnetic
gradient fields to be generated for positional encoding of certain
of the gyromagnetic nuclei within the patient 26 during examination
sequences. A radio frequency (RF) coil 36 is provided, and is
configured to generate radio frequency pulses for exciting the
certain gyromagnetic nuclei within the patient. In addition to the
coils that may be local to the scanner 12, the system 10 also
includes a set of receiving coils 38 (e.g., a phased array of
coils) configured for placement proximal (e.g., against) the
patient 26. The receiving coils 38 may have any geometry, including
both enclosed and single-sided geometries. As an example, the
receiving coils 38 can include cervical/thoracic/lumbar (CTL)
coils, head coils, single-sided spine coils, and so forth.
Generally, the receiving coils 38 are placed close to or on top of
the patient 26 so as to receive the weak RF signals (weak relative
to the transmitted pulses generated by the scanner coils) that are
generated by certain of the gyromagnetic nuclei within the patient
26 as they return to their relaxed state. The receiving coils 38
may be switched off so as not to receive or resonate with the
transmit pulses generated by the scanner coils, and may be switched
on so as to receive or resonate with the RF signals generated by
the relaxing gyromagnetic nuclei.
[0039] The various coils of system 10 are controlled by external
circuitry to generate the desired field and pulses, and to read
emissions from the gyromagnetic material in a controlled manner. In
the illustrated embodiment, a main power supply 40 provides power
to the primary field coil 28. A driver circuit 42 is provided for
pulsing the gradient field coils 30, 32, and 34. Such a circuit may
include amplification and control circuitry for supplying current
to the coils as defined by digitized pulse sequences output by the
scanner control circuit 14. Another control circuit 44 is provided
for regulating operation of the RF coil 36. Circuit 44 includes a
switching device for alternating between the active and inactive
modes of operation, wherein the RF coil 36 transmits and does not
transmit signals, respectively. Circuit 44 also includes
amplification circuitry for generating the RF pulses. Similarly,
the receiving coils 38 are connected to switch 46 that is capable
of switching the receiving coils 38 between receiving and
non-receiving modes such that the receiving coils 38 resonate with
the RF signals produced by relaxing gyromagnetic nuclei from within
the patient 26 while in the receiving state, and they do not
resonate with RF energy from the transmitting coils (i.e., coil 36)
so as to prevent undesirable operation while in the non-receiving
state. Additionally, a receiving circuit 48 is provided for
receiving the data detected by the receiving coils 38, and may
include one or more multiplexing and/or amplification circuits.
[0040] It should be noted that while the scanner 12 and the
control/amplification circuitry described above are illustrated as
being coupled by a single line, that many such lines may occur in
an actual instantiation. For example, separate lines may be used
for control, data communication, and so on. Further, suitable
hardware may be disposed along each type of line for the proper
handling of the data. Indeed, various filters, digitizers, and
processors may be disposed between the scanner and either or both
of the scanner and system control circuitry 14, 16. By way of
non-limiting example, certain of the control and analysis circuitry
described in detail below, although illustrated as a single unit,
includes additional hardware such as image reconstruction hardware
configured to perform the motion correction and image
reconstruction techniques described herein. Further, in certain
embodiments, the control and analysis circuitry described herein
may be associated with an algorithm used for motion detection
and/or another algorithm used for image reconstruction. Indeed,
where an algorithm is described in the present disclosure, it
should be noted that it may be associated with (e.g., a part of or
connected to) the MRI system 10. The algorithm may, for example, be
implemented as specific hardware components (e.g., specialized
processors), or may be implemented as software (e.g., instructions
and/or sets of instructions) on a computing platform.
[0041] As illustrated, scanner control circuit 14 includes an
interface circuit 50 which outputs signals for driving the gradient
field coils and the RF coil and for receiving the data
representative of the magnetic resonance signals produced in
examination sequences. The interface circuit 50 is coupled to a
control and analysis circuit 52. The control and analysis circuit
52 executes the commands for driving the circuit 42 and circuit 44
based on defined protocols selected via system control circuit 16.
Control and analysis circuit 52 also serves to receive the magnetic
resonance signals and performs subsequent processing before
transmitting the data to system control circuit 16. Scanner control
circuit 14 also includes one or more memory circuits 54, which
store configuration parameters, pulse sequence descriptions,
examination results, and so forth, during operation. Interface
circuit 56 is coupled to the control and analysis circuit 52 for
exchanging data between scanner control circuit 14 and system
control circuit 16. Such data will typically include selection of
specific examination sequences to be performed, configuration
parameters of these sequences, and acquired data, which may be
transmitted in raw or processed form from scanner control circuit
14 for subsequent processing, storage, transmission and display.
Therefore, in certain embodiments, the control and analysis circuit
52, while illustrated as a single unit, may include one or more
hardware devices.
[0042] System control circuit 16 includes an interface circuit 58,
which receives data from the scanner control circuit 14 and
transmits data and commands back to the scanner control circuit 14.
The interface circuit 58 is coupled to a control and analysis
circuit 60 which may include a CPU in a multi-purpose or
application specific computer or workstation. Control and analysis
circuit 60 is coupled to a memory circuit 62 to store programming
code for operation of the MRI system 10 and to store the processed
image data for later reconstruction, display and transmission. The
programming code may execute one or more algorithms capable of
performing, by way of example, non-Cartesian imaging sequences and
processing sampled image data (e.g., blades of data, undersampled
data, fluid attenuated data), which will be discussed in detail
below. An additional interface circuit 64 may be provided for
exchanging image data, configuration parameters, and so forth with
external system components such as remote access and storage
devices 18. Finally, the system control and analysis circuit 60 may
include various peripheral devices for facilitating operator
interface and for producing hard copies of the reconstructed
images. In the illustrated embodiment, these peripherals include a
printer 60, a monitor 62, and user interface 64 including devices
such as a keyboard or a mouse.
[0043] Scanner 12 and the control and analysis circuit 52
associated therewith produce magnetic fields and radio frequency
pulses in a controlled manner to excite and encode specific
gyromagnetic material within the patient 26. The scanner 12 and
control and analysis circuit 52 also sense the signals emanating
from such material and create an image of the material being
scanned. In certain embodiments, the scan may include fast spin
echo (FSE) scan, gradient echo (GRE) scan sequences, and the like.
It should be noted that the MRI system described is merely intended
to be exemplary only, and other system types, such as so-called
"open" MRI systems may also be used. Similarly, such systems may be
rated by the strength of their primary magnet, and any suitably
rated system capable of carrying out the data acquisition and
processing described below may be employed.
[0044] Specifically, aspects of the present disclosure include
methods for acquiring magnetic resonance data and processing of
such data to construct one or more motion-corrected images. At
least a portion of the disclosed methods may be performed by the
system 10 described above with respect to FIG. 1. That is, the MRI
system 10 may perform the acquisition techniques described herein,
and, in some embodiments, the data processing techniques described
herein. It should be noted that subsequent to the acquisitions
described herein, the system 10 may simply store the acquired data
for later access locally and/or remotely, for example in a memory
circuit (e.g., memory 62). Thus, when accessed locally and/or
remotely, the acquired data may be manipulated by one or more
processors contained within an application-specific or general
purpose computer. The one or more processors may access the
acquired data and execute routines stored on one or more
non-transitory, machine readable media collectively storing
instructions for performing methods including the motion detection,
image processing, and reconstruction methods described herein.
[0045] To facilitate presentation of certain of the embodiments
described herein, example acquisition and reconstruction sequences
are described below. However, the present disclosure is not limited
to such acquisitions and sequences, unless explicitly stated
otherwise.
[0046] In certain embodiments, 2D MR images are generated from
Cartesian k-space, using either gradient-echo (GRE) or fast spin
echo (FSE) pulse sequences, and acquired with RF receiver coil
arrays of 8 or more coils. Each of the coils has a corresponding
sensitivity to RF signals generated during acquisition, and the
sensitivity of each coil may be mapped to generate sensitivity maps
for the coil array. Image reconstruction may involve the generation
of a partial image corresponding to each coil by 2D Fourier
transformation of the data obtained by a particular coil (referred
to as "coil data"), and multiplication by the conjugate of the
coil's sensitivity map. To generate a full image, these partial
images are summed and the result divided by the sum of squares of
the coil sensitivity maps to give the final image.
[0047] When a patient moves during the scan, the coil data contain
a mixture of Fourier components from two or more motion states. The
resulting image is corrupted and contains motion-related artifacts.
One aspect of the present disclosure involves detecting the
presence of motion and identifying the time during the scan at
which it occurred. In accordance with certain embodiments, this
motion detection may be performed after the scan has been
completed, or may be performed during the scan.
[0048] FIG. 2 is a process flow diagram of an embodiment of a
method 80 for identifying the occurrence/timing of motion, for
example after the scan has completed or in other embodiments during
a scan. As the method 80 may be performed after the scan has been
completed, the method 80 may be performed by the system 10 of FIG.
1, or a remote system having access to the data acquired by system
10. In either case, the method 80 is performed by a general-purpose
or application-specific computing system having appropriately
configured hardware and software for carrying out the steps
described herein. For example, the method 80 may be considered to
be performed by a MR motion detection and correction system having
processing circuitry and memory circuitry, where the memory
circuitry stores instructions (e.g., one or more sets of
instructions associated with a software package) that, when
performed by the processing circuitry, cause the acts associated
with method 80 to be performed. Further, the system that performs
method 80 may also include programming to access and/or store the
coil data obtained by the system 10 for further processing and
analysis. Indeed, the method 80 assumes that a scan has either been
performed and completed, or that the scan is underway. In this
regard, the method 80, in certain embodiments, may include acts
that are not explicitly shown in FIG. 2, such as providing raw
magnetic resonance (MR) data. The acts of "providing" such data may
include, by way of example, accessing MR data that has already been
acquired, performing an acquisition sequence to acquire the MR data
using the MRI system 10 of FIG. 1, and so on.
[0049] As illustrated in FIG. 2, the method 80 includes calculating
(block 82) intensity-corrected single-coil images. This may be
done, for example, by taking the Inverse Fourier transform of the
raw data from each coil in the array, multiplying by the conjugate
of the coil sensitivity map, and dividing by the absolute value
squared of the coil sensitivity maps. FIG. 3 is a set of example
intensity-corrected single-coil images produced by the acts
represented by block 82. As shown, each of the images is associated
with a particular number corresponding to the coil in the coil
array. In this example, the coils are labeled 0-7. The images
resulting from the acts of block 82 are complex--including both
magnitude and phase, and include motion-related artifacts.
[0050] Returning to FIG. 2, once the intensity-corrected
single-coil images are generated, motion may be identified based on
inconsistencies among the images. Blocks 84-90 detail example
processes that are performed to identify the inconsistencies among
the images to score the motion and identify timing associated with
the motion (e.g., which time steps of the scan order are associated
with motion).
[0051] According to the method 80, a Fourier transform is applied
(block 84) to the intensity-corrected single-coil images to
transform the data back into k-space. K-space difference maps are
then determined (block 86) on a pair-wise basis. For example, the
acts represented by block 86 may include taking the absolute
difference between the Fourier transform (the k-space data) of two
intensity-corrected single-coil images. FIG. 4 depicts a map of the
absolute value of the difference between the Fourier transform of
two intensity-corrected single-coil images. FIG. 5 is a plot of the
difference, generated by projection of the difference map along the
readout direction onto the x-axis.
[0052] Once the difference maps are generated, the method 80
includes identifying (block 88) discontinuities in the difference
maps. For example, referring to FIG. 4, discontinuities are visible
as bright columns. In this embodiment, the brighter the column, the
higher the difference (inconsistency). Referring to FIG. 5, the
discontinuities may be visible as large peaks, where the larger the
peak, the greater the difference (inconsistency). That is, a size
of a peak is proportional to a degree of inconsistency between the
k-space data of two intensity-corrected single-coil images.
[0053] Using scan-order information and the difference maps or
difference values, a motion score is calculated (block 90) for each
time step. FIG. 6 is an example MRI scan order for a 256.times.256
FSE image with echo-train length of 8. The horizontal axis
indicates the scan step in time, and the vertical axis indicates
the column, or phase-encode, in Fourier space being scanned at a
given time step. In other embodiments, the vertical axis may
indicate the row, or phase-encode, in Fourier space being scanned.
Note that in this scan order there is a period of 8 before coming
back to scanning the adjacent column in k-space. However, the
present disclosure is not limited to this example scan order. In
general, the elapsed time is not constant between steps. In certain
embodiments, the motion score is calculated in accordance with
block 90 by taking the integral of the difference, for example by
integrating the peaks of the plot of the difference shown in FIG.
5. FIG. 7 is an example plot of a motion score calculated for each
time step of an image (a single-coil image). The y-axis is the
value of the motion score and the x-axis is the time step.
[0054] The most significant motion score is then selected (block
92), for example by selecting the highest motion score. The
selected motion score is then used to determine (block 94) whether
there is significant motion, and if so, the time at which the
motion occurred. By way of example, the calculated motion scores
may be compared against a threshold. If the motion score is at or
above the threshold, then significant motion may be considered to
have occurred. In the plot of FIG. 7, significant motion occurred
at time step 144, as shown by line 96.
[0055] FIG. 8 is a plot of calculated motion score in each image
for a 22-slice series of FSE images, where the subject was
instructed to move his head partway through the scan. Here the odd
numbered slices were first acquired in interleaved fashion, then
followed by the even numbered slices. More specifically, FIG. 8 is
a plot of maximum calculated motion score as a function of image
number. FIG. 9 is an image corresponding to the odd-numbered
slices, while FIG. 10 is an image corresponds to the even-numbered
slices. In comparing FIGS. 9 and 10, the odd slices appear to have
motion-related artifacts, while the even slices do not appear to
have motion-related artifacts. The difference between the motion
scores in the odd and even slices is clearly visible in the plot of
FIG. 8.
[0056] As set forth above, the disclosed embodiments allow motion
detection both after the scan of the subject of interest is
completed, as well as during scanning. FIG. 11 is a process flow
diagram of an embodiment of a method 110 for identifying the
occurrence/timing of motion while the scan is in progress. Like the
method 80, the method 110 may be performed by the system 10 of FIG.
1, or a remote system having access to the data acquired by system
10. In either case, the method 110 is performed by a
general-purpose or application-specific computing system having
appropriately configured hardware and software for carrying out the
steps described herein. For example, the method 110 may be
considered to be performed by a MR motion detection and correction
system having processing circuitry and memory circuitry, where the
memory circuitry stores instructions (e.g., one or more sets of
instructions associated with a software package) that, when
performed by the processing circuitry, cause the acts associated
with method 110 to be performed. Further, the system that performs
method 110 may also include programming to access and/or store the
coil data obtained by the system 10 for further processing and
analysis.
[0057] The method 110 of FIG. 11 includes certain steps that are
performed in a similar manner to the method 80 of FIG. 2. However,
the data that are input to the method 110 are different in that
they are incomplete for most of the scan duration. As set forth in
FIG. 11, the method 110 includes, for each time step during the
image acquisition, calculating an intensity-corrected single-coil
image using only zero-filled k-space data that have already been
collected. That is, the acts of block 112 may include taking the
Inverse Fourier transform of the raw data from each coil in the
array (the data collected as of that time step), multiplying by the
conjugate of the coil sensitivity map, and dividing by the absolute
value squared of the coil sensitivity map. For regions of k-space
that have not been sampled during the scan, those regions of
k-space are zero-filled.
[0058] The method 110 then proceeds in a similar manner to the
method 80 of FIG. 2, where the single coil images are Fourier
transformed (block 84) back into k-space, and k-space difference
maps are calculated (block 86) between the data for two coils at a
time. This results in the k-space difference between pairs of
single-coil images using only the zero-filled k-space data
collected up to a particular time step. Referring to FIG. 12, for
instance, which is a schematic representation of the manner in
which k-space is zero-filled, the k-space data include first
sections 114, where k-space is filled with acquired data, and
second sections 116, where the k-space region is zero-filled. As
shown in FIG. 12, there may be boundaries between the first
sections 114 and the second sections 116.
[0059] Returning to FIG. 11, the method 110 further includes
identifying (block 118) discontinuities or peaks in the difference
maps or in the difference information. For example, the acts of
block 118 may include identifying discontinuities/peaks in the
calculated k-space difference by calculating the 1D projection of
the absolute value along rows and/or columns and detecting peaks.
The peak detection may be performed, for instance, by calculating
the negative of the second derivative of the signal at the
particular time step, as plotted in FIG. 13.
[0060] Once the discontinuities or peaks are identified in
accordance with block 118, the method 110 includes zeroing out
(block 120) expected peaks using scan order information. For
instance, peaks unrelated to motion are generated at any borders
between so-far empty (zero-filled regions) and populated regions of
k-space. These peaks are removed in accordance with block 120. In
the situation shown in FIGS. 11 and 12, no motion has occurred and
all peaks result from boundaries between the first sections 114 and
the second sections 116. Thus, as shown in FIG. 14, a plot 122 of
the "unexpected" positive peaks shows no peaks because no patient
motion has occurred.
[0061] The method 110 includes calculating (block 124) a score for
motion. By way of example, calculating the score in accordance with
block 124 may include taking the sum of the remaining "unexpected"
positive peaks. As shown in FIG. 15, this results in a plot 126 of
calculated motion score as a function of time step. Again, because
motion does not occur until a time step (noted by line 128)
occurring after the time step at which data have been collected, no
appreciable peaks show in the plot 126.
[0062] In certain embodiments, the unexpected peaks may be
normalized. For example, the amplitude of the expected peaks may be
used to normalize the unexpected peaks, resulting in the peaks
being dimensionless. A natural threshold of about 0.5 may be used
in such cases for motion detection. However, the present disclosure
is not limited to using 0.5 as a threshold, but it has been found
that normalized unexpected peaks above about 0.5 tend to correlate
well with the presence of motion.
[0063] As there will be non-zero values for the motion score
resulting from irregularities that are not a result of motion, the
method 110 includes applying a threshold (block 130) to the
calculated motion scores to determine whether the calculated motion
score is indicative of motion. Thus, motion may be identified if
the motion score is at or above the threshold, for example. In the
example shown in the motion score plot 126 of FIG. 15, none of the
values would be at or above an appropriately selected
threshold.
[0064] In certain situations, the system performing the method 110
may identify multiple instances of movement. In those situations,
additional actions may be performed as part of the method 110. For
example, once motion-related peaks are identified, they can be
added to the list of expected peaks (block 132). Upon
identification of the motion-related peaks in accordance with block
132, the method 110 may cycle back to the acts associated with
block 120, were the expected peaks are removed (zeroed out). This
may allow for the detection of subsequent motion states by
processing of new unexpected peaks (at a later time step).
[0065] To help illustrate, FIGS. 16-19 depict various data plots
associated with a detected motion event. In particular, FIG. 16
depicts a schematic of partially-filled k-space having the first
sections 114, the second sections 116, and third sections 140. In
this particular example, the first and third sections 114, 140
represent sections of k-space that are filled with acquired data,
but differ in that the first sections 114 are acquired before the
motion event, and the third sections 140 are acquired after the
motion event. The second sections 116, as previously noted,
represent zero-filled sections of k-space.
[0066] FIG. 17 depicts a plot of the negative of the second
derivative of the signal, and includes multiple peaks. Removal of
the expected peaks results in the unexpected peak plot depicted in
FIG. 18. The graph of FIG. 19 shows the calculated motion score up
to the current time step, with the known motion event occurring at
a time step denoted by the line 142.
[0067] The method set forth in FIG. 11 was tested on a simulated
data set. The simulated data set included approximately 6500 MR
images with randomly selected motion timing and shift. The
algorithm ran in an automatic manner on all cases and the true
motion timing was compared to the calculated motion timing. A
relatively simple motion score was used for calculating the motion
timing. The results are shown in FIG. 20. The x values of the
points represent the calculated motion timings while the y values
represent the true motion timings. As shown, the calculated motion
timings very closely matched the simulated true motion timings. The
outliers are related to a specific unhandled end case of the
calculation and do not represent a limitation of the method.
[0068] From the foregoing, it should be appreciated that patient
motion during an MR scan sequence creates a composite k-space
dataset having a section of k-space acquired before the motion and
another section of k-space acquired after the motion (e.g., as
shown by the first and third sections 114 and 140 of FIG. 16). In
cases where the patient moves multiple times, the k-space dataset
may include multiple (e.g., two or more than two) acquired
sections, each corresponding to the span of time between patient
motion. The Fourier transform of such a composite dataset contains
motion artifacts whose severity depends on the timing and size of
the motion. Thus, it may be desirable to not only assess whether
motion has occurred, but also leverage the information already
available from the methods described herein to ascertain the
severity of the motion states. For example, the motion score
magnitude may be used to determine the severity of a given motion
event, vis-a-vis the effect that the motion has on the quality of
the images produced from a given scan (or the effect of the motion
on sub-images).
[0069] Once the motion has been detected and the timing is known,
various actions can be taken, including restarting the scan,
reacquiring those portions of k-space acquired before the movement,
or correcting for the motion to reconstruct an artifact-free MRI
image, using the existing data. The manner in which the effects of
motion can be mitigated depends on, among other things, the time at
which the motion was detected versus the time at which the motion
occurred. For example, in situations where motion is not detected
until after the scan has been completed, the methods available to
ameliorate the effects of the motion may not be the same as those
available when the motion is detected during the scan. FIGS. 36-40
detail various methods that may be performed by the MR system 10 in
different motion situations.
[0070] FIG. 21 depicts a process flow diagram of an embodiment of
an algorithm 260 performed, for example, by control and analysis
circuitry 52, 60 of the MR system 10 in situations where motion is
detected during a scan. The algorithm 260 includes various
operations, including beginning the scan at operation 262. This
begins the process of acquiring new data at operation 264.
[0071] Once data have been acquired, the algorithm 260 performs a
query 266 to determine whether motion has been detected, using any
one or a combination of the methods described herein. If motion has
not been detected in the most recent shot, the k-space data are
aggregated at operation 268 with the previous k-space data, and if
the scan has not finished (query 270), the scan is continued as
normal at operation 272. If the scan is finished, another query 274
is performed to determine if motion has been detected and if the
scan is motion-free, the image is reconstructed according to
conventional techniques.
[0072] If, at query 266, motion is detected, the data previously
collected are saved as one motion state at operation 278, and a new
motion state is started at operation 280. The new motion state
initially includes only the most recent k-space data collected. As
the scan continues, k-space data will be aggregated to this motion
state as long as further motion is not detected. The results of the
operations described to this point result in continuing to either
add to the current motion state or creating new motion states until
the scan is complete.
[0073] At query 274, if there were multiple motion states, each
aggregate (each set of k-space data corresponding to a single
motion state) is separately reconstructed at operation 282. In this
respect, each reconstructed motion state results in a motion-free
sub-image and multiple motion-free sub-images 284 are produced.
[0074] At operation 286, various known techniques can be used to
combine the different sub-images, or to separately reconstruct them
into full images. For example, the sub-images 284 can be registered
and combined to create a motion-free image, through methods known
in the art. Or the k-space data from each motion state can be
reconstructed using parallel imaging, compressed sensing, or a
sparse-reconstruction neural network. The resulting images can then
be registered and combined, through methods known in the art. As
one example, operation 286 may include joint iterative estimation
of motion and image, with timing constraints. The timing
constraints (i.e., motion timing) are obtained based on the neural
network predictions.
[0075] Using a similar sequence, instead of aggregating the
separate motion states, the k-space data may be adaptively
reacquired as shown in FIG. 22. In particular, algorithm 290 of
FIG. 22 includes many of the same operations as the algorithm 260
of FIG. 21, including operations 262, 264, 268, 272, and 276 as
well as queries 266, 270, and 274.
[0076] For the algorithm 290, if motion is detected at query 266,
then the system (e.g., control and analysis circuitry 52, 60)
determines whether enough of k-space has been filled to make
possible parallel imaging/compressed sensing (PICS) or use of a
sparse-image-reconstruction neural network at query 292.
[0077] If not enough of k-space has been filled, then the algorithm
290 continues to acquire data by adding data to a new motion state
at operation 294. If necessary, lines of k-space filled in during
previous motion states are re-acquired. Previous motion state data
may be discarded or used for other purposes.
[0078] Once enough of k-space has been filled to make parallel
imaging or sparse image reconstruction possible, the scan is ended
at operation 296. The final image is reconstructed at operation 298
with just the portion of k-space acquired in the final motion state
using one of the aforementioned reconstruction algorithms (e.g.,
PICS recon or a sparse-image-reconstruction neural network).
[0079] In certain embodiments, the detected motion may be so severe
that the data are essentially unusable. FIG. 23 depicts an
embodiment of an algorithm 300 that involves ending the scan early
if motion is detected. For example, the algorithm 300 may include
many of the operations and queries described previously with
respect to FIGS. 36 and 37, except that once severe motion is
detected at query 266, the scan is ended at operation 302. For
example, the motion score predicted by the CNN 180 may be so high
that the motion may be considered severe and the scan ended.
[0080] Ending the scan in this manner allows an operator to take
adaptive actions at operation 304. For example, the operator may
instruct the subject to remain still, assist the subject if
remaining still is difficult, or a more motion-robust imaging
sequence may be utilized (e.g., automatically chosen by the
system). Once adaptive correction is performed, the scan may be
re-started at operation 306.
[0081] The algorithm 300 may be used in combination with the two
algorithms 260, 290 described above, by taking advantage of the
fact that the motion score reflects not only the presence but also
severity of motion. For instance, the scan can be ended early if
severe motion is detected multiple times, but otherwise one of the
other algorithms can be implemented in response to a smaller motion
score. This algorithm 300 also makes use of the quality score to
allow selection of a particular tolerance for motion scores. For
instance, depending on the intended use for the finished scans,
minor motion artifacts may not affect the diagnosis. The same
neural network with a dynamic threshold allows multiple thresholds
to be customized to specific applications.
[0082] Disclosed embodiments also include methods for image
reconstruction when motion has occurred. For example, FIGS. 39 and
40 both depict embodiments of methods that can be used to
reconstruct a motion artifact-free image. FIG. 24, in particular,
is a method 310 for reconstructing a motion artifact-free image by
first dividing coil data (block 312) into pre-motion and
post-motion datasets. For example, using the scan order and the
timing of when the motion occurred, the coils' k-space data is
broken into two sets. The first set includes the parts of k-space
that were scanned before the subject movement occurred and the
second set the data after the movement occurred.
[0083] After the coil data are divided, for each coil two images
are reconstructed (block 314). The first image is reconstructed
using the zero-filled k-space data collected before the movement
occurred, and the second image is reconstructed using the zero
filled k-space data collected after the movement. In method 310,
the two sets of images for each coil are fed (block 316) into a
deep-learning neural network that reconstructs a single
motion-corrected image.
[0084] Method 320, on the other hand and as depicted in FIG. 25,
includes the acts represented by blocks 312 and 314, but instead
the partial k-space images are each processed using a sparse
reconstruction algorithm (block 322). The images resulting from the
sparse reconstruction algorithm may then be further processed and
combined, or fed to a neural network to generate a final
motion-free image (block 324).
[0085] Technical effects of the invention include automatic
detection and timing of patient movement, and mitigation of the
effects of the patient movement on an overall MR scan. Remedial
actions may include restarting the scan, reacquiring those portions
of k-space acquired before the movement, or correcting for the
motion using the existing data. In this way, the motion detection
and correction techniques described herein may improve the
throughput of MRI machines, improve the patient experience and
reduce burden on MR technicians.
[0086] This written description uses examples to disclose the
invention, including the best mode, and also to enable any person
skilled in the 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 skilled
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.
* * * * *