CN101297214A - A highly constrained rear projection reconstruction process in cardiac gated MRI - Google Patents

Abstract

A procedure for cardiac-gated acquisition of MR data during breath-holding uses a hybrid PR pulse sequence to acquire projection views from which the image frame. At each heartbeat phase, a composite image is reconstructed using interleaved projection views acquired during all heartbeats. During the same heartbeat phase, the aforementioned synthetic images are used to reconstruct highly undersampled image frames by using a highly constrained rear projection method.

Description

A highly constrained rear projection reconstruction process in cardiac gated MRI

Cross References to Applications

This application is based on two U.S. Provisional Patent Applications: 60/719,445, filed September 22, 2005, entitled "HIGHLY CONSTRAINED IMAGE RECONSTRUCTION METHOD"; and GATED MAGNETIC RESONANCE IMAGING" application 60/738,444.

Background of the invention

The field of the invention is magnetic resonance imaging methods and systems. More specifically, the present invention relates to the process of reconstructing images from cardiac gated MRI acquisitions.

When a substance such as human tissue is subjected to a uniform magnetic field (polarization field B ₀ ), the magnetic moments of the individual spins in the tissue try to align with the polarization field, but at their characteristic Larmor frequency at Precess around it in any order. If the material or tissue is subjected to a magnetic field in the xy plane near the Larmor frequency (exciting field B ₁ ), the net alignment magnetic moment _Mz will rotate or "tilt" into the xy plane, resulting in a net transverse Magnetic moment M _t . These excited spins emit a signal, and after termination of the excitation signal B ₁ , this signal can be received and processed to form an image.

When using these signals to generate images, magnetic field gradients ( _Gx , _Gy , and _Gz ) can be used. Typically, the region to be imaged is scanned through a series of measurement cycles during which these gradients vary according to the particular localization method used. Each measurement is called a "view" in the art, and the number of views determines the resolution of the image. The resulting set of received NMR signals, or views, or k-space samples is digitized and processed to reconstruct an image using one of many well-known reconstruction techniques. The total scan time is determined in part by the number of measurement cycles or views acquired for an image, so by reducing the number of views acquired, scans can be reduced at the expense of image resolution time.

One of the most common methods for acquiring NMR data sets from which images can be reconstructed is known as the "Fourier transform" imaging technique or the "spin-warping" technique. This technique is discussed in the article entitled "Spin-Warp NMR Imaging and Applications to Human Whole-Body Imaging" by WA Edelstein et al. (see Physics in Medicine and Biology, Vol. 25, pp. 751-756, 1980 for details) . It uses a variable-amplitude phase-encoding magnetic field gradient pulse to phase-encode the spatial information in the direction of the gradient before acquiring NMR signals. In a two-dimensional implementation (2DFT), for example, by applying a phase-encoding gradient (G _y ) along one direction, the spatial information is encoded in that direction, and then there is a readout in a direction orthogonal to the phase-encoding direction. The signal is acquired under the condition of a high magnetic field gradient (G _x ). The readout gradients present during spin-echo acquisition encode spatial information in orthogonal directions. In a typical 2DFT pulse sequence, the amplitude of the phase-encoding gradient pulse G _y (G _y ) is increased in the sequence of views acquired during the scan described above. In a three-dimensional implementation (3DFT), a third gradient _Gz is added to the phase encoding along the third axis before each signal readout. The amplitude of the second phase-encoding gradient pulse _Gz also traverses a number of values during the scan. These 2DFT and 3DFT methods sample k-space in the form of a straight line as shown in Fig. 2, and the k-space samples are located on Cartesian grid coordinates.

Magnetic resonance angiography (MRA) uses the phenomenon of nuclear magnetic resonance to produce images of the body's vasculature and heart. To enhance the diagnostic capabilities of MRA, a contrast agent such as gadolinium can be injected into the patient prior to the MRA scan. The trick to using this method of contrast-enhanced (CE) MRA, as described in US Patent 5,417,213, is to acquire a central k-space view at the moment when contrast agent is flowing through the vasculature of interest. The key to successful CEMRA detection is to collect the centerline of k-space during peak arterial enhancement. If the centerline of k-space is acquired before the contrast agent arrives, severe image artifacts may limit the diagnostic information in that image. Alternatively, enhancement of veins sometimes obscures arterial images acquired after peak arterial contrast agent passage. In many anatomical regions (such as the carotid or renal arteries), the separation between arterial and venous enhancement can be as short as 6 seconds.

The MRA data acquisition process is timed such that the central region of k-space is acquired when the contrast agent reaches the artery of interest. The ability to time the arrival of contrast agents varies considerably, and in many applications it is useful to acquire a series of MRA images in dynamic studies depicting isolated arterial and venous enhancement. A series of time-ordered images can also be used to observe delayed vessel filling patterns caused by disease. This requirement is partially addressed by acquiring a series of time-resolved images using a three-dimensional "Fourier" acquisition method. This three-dimensional "Fourier" Acquisition methods are described.

More recently, projection reconstruction methods have been used to acquire time-resolved MRA data, as described in US Patent 6,487,435. The projection reconstruction method, sometimes called "radial" acquisition, has been known since the advent of magnetic resonance imaging. Instead of sampling k-space in a linear scan like Fourier imaging (i.e., as shown in Figure 2), projection reconstruction methods like Figure 3 acquire a series of views that extend from the center of k-space outwards. Radial lines are sampled. The number of views required to sample k-space determines the length of the scan, and if the number of views collected is insufficient, streak artifacts will be generated in the reconstructed image. The technique described in patent 6,487,435 reduces this streaking by acquiring successive undersampled images with interlaced views and sharing peripheral k-space data between successive images.

In US Patent 6,710,686 two methods are described for reconstructing an image from an acquired set of k-space projection views. The most common approach is to reframe k-space samples from their positions on radial sampling trajectories into Cartesian grid coordinates. Then, the image is reconstructed by performing 2D or 3D Fourier transform on the re-framed k-space samples. The second method used to reconstruct the image is to transform the radial k-space projected views into Radon space by Fourier transforming each projected view. An image is reconstructed from these signal projections by filtering them and back-projecting them into the field of view (FOV). As is well known in the art, if the number of acquired signal projections is insufficient to satisfy the Nyquist sampling rule, streaking artifacts will be produced in the reconstructed image.

Figure 4 shows a standard rear projection method. Each acquired signal projection profile 10 is back-projected onto the field of view 12 by projecting each signal sample 14 in the profile 10 through the FOV 12 along the projection path indicated by arrow 16. In backprojecting each signal sample 14 into FOV 12, we do not have any prior information about the subject and assume that the NMR signals in FOV 12 are homogeneous and signal samples 14 should be equally distributed into each pixel traversed by the projection path. For example, FIG. 4 shows the projection path 8 of a single signal sample 14 in a signal projection profile 10 as it passes through N pixels in the FOV 12 . The signal value (P) of the signal sample 14 is equally divided between these N pixels:

S _n =(P×1)/N (1)

where: S _n is the NMR signal value assigned to the nth pixel in a projection path with N pixels.

Clearly, the assumption that the NMR signal is homogeneous in FOV 12 is incorrect. However, as is known in the art, the error caused by this false assumption is minimized if some filter correction is applied to each signal distribution 10 and a sufficient number of filtered distributions are acquired at a corresponding number of projection angles. And image artifacts are suppressed. In a typical filtered back-projection method for image reconstruction, 256×256 pixel 2D images require 400 projections, while 256×256×256 voxel 3D images require 203,000 projections. The number of projected views required for these same images may be reduced to 100 (two-dimensional) and 2000 (three-dimensional) if the method described in the aforementioned US Patent 6,487,435 is used.

The motion of the beating heart becomes a problem when imaging certain arteries, such as the coronary arteries. To reduce motion artifacts in MRI or MRA images, it is common practice to cardiac-gate the view acquisition process with ECG signals representing cardiac phases. As mentioned above, for example, in US Pat. No. 5,329,925, a set or segment of views is acquired during one or more cardiac phases within each cardiac cycle. For example, 8 different views can be acquired during a particular heartbeat phase, and after 16 heartbeats a total of 8 x 16 = 128 different views can be acquired from which an image can be constructed. Because a single breath hold is typically 16 to 20 heartbeats, it is highly desirable to acquire all data during the breath hold to avoid artifacts due to breathing motion.

Although reasonably good monolithic 2D images can be acquired during one or more heartbeat phases during a single breath-hold using projection reconstruction methods and view sharing, previous methods are not fast enough to acquire 3D images during every heartbeat phase or multiple 2D slices. Such images are necessary when the subject under examination does not lie in a single two-dimensional plane (eg coronary arteries), either when multi-slice or three-dimensional image acquisition is required.

Contents of the invention

The present invention is a novel method for generating cardiac gated MR images, and in particular a method for improving the quality of highly undersampled images acquired during specific cardiac phases. During selected heartbeat phases during consecutive heartbeats, a series of undersampled image frames are acquired. Views acquired during successive heartbeats sample interleaved trajectories in k-space, and these samples are combined to reconstruct a composite image depicting the subject at selected heartbeat phases. The composite image is used in a highly constrained rear projection process for each projected view by weighting the distribution of the rear projection signal samples.

It is the discovery of the present invention that high quality frame images can be produced with very few acquired views if a priori information about the NMR signal profile in the FOV 12 is used in the rear projection image reconstruction process rather than assuming a uniform signal profile . Referring to FIG. 5, for example, the signal profile in FOV 12 may include structures like blood vessels 18 and 20. In this case, a more accurate distribution of the signal samples 14 in each pixel is achieved by weighting the distribution according to the known NMR signal profile at that pixel location as the rear projection path 8 traverses these structures. As a result, the majority of signal samples 14 will be distributed at those pixels that intersect structures 18 and 20 . For a rear projection path 8 with N pixels this can be expressed as:

${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Where: P = NMR signal sample value; and

C _n = the signal value at the nth pixel of the composite image along the rear projection path.

The numerator in equation (2) weights each pixel with the corresponding NMR signal value in the composite image, and the denominator normalizes this value such that all rear projection signal samples reflect the projected sum of the image frame and do not Multiplied by the sum of the composite image. It should be noted that although the normalization described above is performed for each pixel individually after rear projection is performed, in many clinical applications it is easier to normalize the projection P prior to rear projection. In this case, the above-mentioned projections P are normalized by dividing by the corresponding value _Pc during projections through the composite image at the same viewing angle. The normalized projection P/ _Pc is back-projected and the resulting image is then multiplied by the composite image.

Figure 6 shows a three-dimensional embodiment of the invention, corresponding to a single three-dimensional projected view characterized by viewing angles Θ and φ. The projected view is Fourier transformed to form a signal profile and it is rear projected along the axis 16 and extended into the Radon plane 21 at a distance r along the rear projection axis 16 . As an alternative to filtered rear projection, where the projection signal profile is filtered and uniformly distributed into successive Radon planes, along axis 16 the information in the composite image is used to distribute the projection signal profile values into Radon planes 21 . The composite image in FIG. 6 contains blood vessels 18 and 20 . Weighted signal profile values are stored at image positions x, y, z in the Radon plane 21 based on the intensity at the corresponding positions x, y, z in the composite image. This is a simple multiplication process of the signal distribution value with the corresponding composite image voxel value. This product is then normalized by dividing it by the distribution values in the corresponding image spatial distribution formed from the composite image. The formula used for 3D reconstruction is:

I(x,y,z)=∑(P(r,θ,φ)*C(x,y,z) _(r,θ,φ) /P _c (r,θ,φ)(2a)

where the summation (∑) is performed over all projections in the time frame, and the x, y, z values in a particular Radon plane are the distribution values P(r, θ , φ) to calculate. P _c (r, θ, φ) is the corresponding distribution value from the composite image, and C(x, y, z _{)r, θ, φ} is the composite image value at (r, θ, φ).

Another finding of the present invention is that this method of image reconstruction can be advantageously used during cardiac gated acquisition, where a series of undersampled frame images are acquired during the same cardiac phase. By interleaving the images during the acquisition of successive image frames, views from successive image frames can be combined and used to reconstruct a higher quality composite image. This composite image is then used in the rear projection reconstruction process described above for each image frame.

Another aspect of the invention is the reconstruction process of image frames acquired with a three-dimensional hybrid projection reconstruction pulse sequence during a cardiac gated scan. Projection views are acquired to sample k-space with radial trajectories in two-dimensional slices, and phase encoding is used to acquire multiple slices along the axial direction. For each of the plurality of slice positions, a composite image is reconstructed, and these composite images are used in the rear projection reconstruction process of the two-dimensional slices in each image frame.

Description of drawings

Figure 1 is a block diagram of an MRI system using the present invention;

Figure 2 is an illustration of k-space sampling using Fourier transform techniques;

Figure 3 is an illustration of k-space sampling using projection reconstruction techniques;

Figure 4 is an illustration of a conventional rear projection reconstruction method;

FIG. 5 is an illustration of a rear projection method for 2DPR image reconstruction according to the present invention;

Figure 6 is an illustration of a rear projection method for 3DPR image reconstruction;

Figure 7 is an illustration of a hybrid PR pulse sequence performed by the MRI system of Figure 1 when practicing the preferred embodiment of the present invention;

Figure 8 is an illustration of k-space sampling using the hybrid pulse sequence of Figure 5;

Fig. 9 is a flowchart of many steps in the preferred embodiment of the present invention;

Figure 10 is an illustration of the process of cardiac gating acquisition of data during a heartbeat;

Figure 11 is an illustration of interleaved sampling of k-space with radial sampling trajectories;

FIG. 12 is a flowchart of steps for reconstructing a two-dimensional image frame in accordance with the present invention;

Figure 13 is an illustration of an image produced using the method of Figure 10;

Figure 14 is a flowchart of the steps in the second embodiment of the invention for contrast enhancement;

Figure 15 is an illustration of an image produced using the method of Figure 14;

Figure 16 is a flowchart of steps in another preferred embodiment of the present invention; and

FIG. 17 is an illustration of a cardiac gating acquisition process for data during a heartbeat when practicing the method of FIG. 16 .

Detailed ways

With particular reference to Figure 1, a preferred embodiment of the present invention is used in an MRI system. The MRI system includes a workstation 110 having a display 112 and a keyboard 114 . Workstation 110 includes processor 116, which is a commercial programmable machine that can run a commercial operating system. Workstation 110 provides an operator interface that enables scanning of instructions to be entered into the MRI system.

Workstation 110 is coupled to four servers: pulse sequence server 118 ; data acquisition server 120 ; data processing server 122 ; and data storage server 23 . In a preferred embodiment, data storage server 23 is implemented by workstation processor 116 and associated disk drive interface circuitry. The remaining three servers 118, 120 and 122 are implemented by different processors housed in a single chassis and interconnected by a 64-bit backplane bus. The pulse train server 118 employs a commercial microprocessor and a commercial quad communication controller. Both the data collection server 120 and the data processing server 122 use the same commercial microprocessor, and the data processing server 122 also includes one or more array processors based on commercial parallel vector processors.

Workstation 110 and each processor for servers 118, 120 and 122 are connected to a serial communication network. The serial network transports data downloaded from workstation 110 to servers 118, 120, and 122, and it also transports label data passed between servers and between workstations and servers. In addition, a high-speed data link is provided between the data processing server 122 and the workstation 110 for transferring image data to the data storage server 23 .

Pulse sequence server 118 operates in response to program elements downloaded from workstation 110 to operate gradient system 24 and RF system 26 . The gradient waveforms necessary to perform a given scan are generated and applied to gradient system 24, which excites gradient coils in assembly 28, thereby producing magnetic field gradients _Gx , Gy _{, and} G _Z . The gradient coil assembly 28 forms part of a magnet assembly 30 which also includes a polarizing magnet 32 and an integral RF coil 34 .

The RF excitation waveform is applied to the RF coil 34 by the RF system 26 to perform a prescribed magnetic resonance pulse sequence. RF system 26 receives the responsive NMR signals detected by RF coil 34 , amplifies, demodulates, filters, and digitizes these signals under the direction of commands generated by pulse sequence server 118 . RF system 26 includes an RF transmitter that generates various RF pulses used in the MR pulse sequence. The RF transmitter responds to scan instructions and commands from the pulse sequence server 118 to generate RF pulses having a desired frequency, phase, and pulse amplitude shape. The generated RF pulses may be applied to the integral RF coil 34 or to one or more local coils or coil arrays.

RF system 26 also includes one or more RF receiver channels, which may be connected to a corresponding number of local coils or to a corresponding number of coil elements in a coil array. Each RF receiver channel includes: an RF amplifier for amplifying the NMR signal received by the coil connected to it; and a quadrature detector for detecting the I and Q quadrature components of the received NMR signal and making the They are digitized. The magnitude of the received NMR signal can then be determined at any sampling point by taking the square root of the sum of the squares of the I and Q components:

$\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\sqrt{{\mathrm{<span\; class="notranslate">\; I</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; Q</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; ,</span>$

And the phase of the received NMR signal can also be determined:

φ＝tan ^-1 Q/I.

The pulse sequence server 118 also optionally receives patient data from the physiological acquisition controller 36 . Controller 36 receives signals from a number of different sensors connected to the patient, for example, ECG signals from electrodes or breathing signals from the lungs. The pulse sequence server 118 typically uses such signals to synchronize or "gate" the performance of the scan to the subject's breathing or heartbeat.

The pulse sequence server 118 is also connected to the scan room interface circuit 38, which receives signals from various sensors and magnet systems related to the patient's condition. It is also through the scan room interface circuit 38 that the patient positioning system 40 receives commands to move the patient to the desired position during the scan.

Notably, the pulse sequence server 118 performs real-time control of the MRI system components during a scan. As a result, its hardware elements must be operated with program instructions executed in a timely manner by running the program. The descriptive components for the scan instructions are downloaded from the workstation 110 in the form of objects. Pulse sequence server 118 includes programs that receive these objects and convert them into objects used by the run-time programs.

The digitized NMR signal samples generated by RF system 26 are received by data acquisition server 120 . The data acquisition server 120 operates in response to the descriptive components downloaded from the workstation 110 to receive real-time NMR data and to provide buffer storage so that no data is lost due to data overruns. During certain scans, the data collection server 120 simply passes the collected NMR data to the data processing server 122 . However, data acquisition server 120 is programmed to generate and transmit such information to pulse sequence server 118 during those scans in which it is necessary to obtain information from the acquired NMR data in order to control other aspects of the scan. For example, during a pre-scan, NMR data is acquired and used to calibrate the pulse sequence executed by the pulse sequence server 118 . Also, during a scan, navigation signals can be acquired and used to adjust RF or gradient system operating parameters or to control the order of views for sampling k-space. Additionally, the data acquisition server 120 may be used to process NMR signals that are used to detect the arrival of contrast agents in MRA scans. In all of these examples, the data acquisition server 120 acquires the NMR data and processes it in real time to generate information used to control the scan.

The data processing server 122 receives the NMR data from the data acquisition server 120 and processes it according to the descriptive components downloaded from the workstation 110 . Such processing may include: Fourier transforming raw k-space NMR data to generate two- or three-dimensional images; applying filtering to reconstructed images; performing rear-projection image reconstruction on acquired NMR data; computing functional MR images; Calculate motion or flow images, etc.

The image reconstructed by the data processing server 122 is transmitted back to the workstation 110 and stored. The real-time images are stored in a database memory cache (not shown) and output from the cache to the operator display 112 or display 42, which is placed near the magnet assembly 30 for physician use. Batch mode images or selected real-time images are stored in a master database on disk storage device 44 . The data processing server 122 notifies the data storage server 23 on the workstation 110 when such images have been reconstructed and transferred to the storage device. Workstation 110 may be used by an operator to archive images, produce film, or send images over a network to other devices.

To practice the preferred embodiment of the invention, NMR data is acquired using projection reconstruction or a radial pulse sequence such as that shown in FIG. 7 . This is a fast gradient-recall echo pulse sequence in which a selective, asymmetrically truncated sinc RF excitation pulse 200 is generated in the presence of a slice-selective gradient 202 . The flip angle of the RF pulse 200 is set close to the Ernst angle for _Ti shortened blood, typically 30°-40°.

As explained in detail below, this pulse sequence can be used to acquire a single two-dimensional slice by sampling in a single circular plane of k-space, or it can be used to sample multiple circular planes of k-space (as in Figure 8). 204, 206 and 208) for sampling. When multiple 2D slices are acquired, the axial gradient 202 is a long byte selection gradient followed by a phase encoding gradient lobe 210 and a rewinder gradient lobe 212 of opposite polarity. This axial phase encoding gradient 210 traverses multiple values during the scan to sample from the two-dimensional k-space planes 204 , 206 and 208 .

The two in-plane readout gradients 214 and 216 are exhausted during the acquisition of the NMR echo signal 218 in order to sample k-space along radial trajectories in the two-dimensional plane 204 , 206 or 208 . These in-plane gradients 214 and 216 are perpendicular to the axial gradients, and they are perpendicular to each other. During a scan, they iterate through a series of values to rotate the viewing angle of the radially sampled trace, as described in more detail below. Each in-plane readout gradient is preceded by prephasing gradient lobes 220 and 222 and followed by rewinding gradient lobes 224 and 226 .

It should be apparent to those skilled in the art that sampling trajectories other than the preferred straight line trajectories described above, which start at a point on the peripheral edge of k-space and pass through the center of k-space, can be used Afterwards, it reaches the opposite point on the outer edge of k-space. A variant is to acquire a portion of the NMR echo signal 218 that is sampled along a trajectory that does not span the entire extent of the sampled k-space volume. Another variation equivalent to rectilinear projection of the reconstructed pulse sequence is to sample along a curved path instead of a straight line. Such pulse sequences are described in: "FastThree Dimensional Sodium Imaging" by F.E. Boada et al., MRM, 37:706-715, 1997; "Rapid 3D PC-MRA Using Spiral Projection Imaging" by K.V. Koladia et al., Proc.Intl.Soc.Magn.Reson.Med.13(2005); and "Spiral Projection Imaging: a new fast 3D trajectory" by J.G.Pipe and Koladia et al., Proc.Intl.Soc.Mag.Reson.Med.13( 2005). It should also be apparent that the invention can also be used with 3D and 2D versions of these sampling methods, and that the term "pixel" as used hereinafter is intended to mean a position in a 2D or 3D image.

The MRI system of FIG. 1 uses the pulse sequence described above to acquire a series of cardiac gated frame images. In a first preferred embodiment, only a single 2D slice is acquired in each image frame. Referring to FIGS. 1 and 9 , after the subject is in the chamber of the MRI system and the ECG signal is coupled to the physiological acquisition controller 36 , the system waits for an ECG trigger signal, as shown in decision block 300 . When a trigger signal is received, as indicated at 302, a set of image frames is acquired. This is illustrated in Figure 10, where the cardiac cycle is initiated at 304 by an ECG trigger signal, and a set of six image frames 306-311 are acquired at predetermined times or "heartbeat phases" during the subsequent RR interval. In a preferred embodiment, each image frame acquisition process at process block 312 includes 10 projection views set at those viewing angles that sample the two-dimensional k-space as uniformly as possible. This is a highly under-sampled data set and without the present invention would produce a very poor quality image when performing typical image reconstruction.

When the last image frame is acquired during the RR interval as determined by decision block 314 , the system loops back and waits for the next ECG trigger at decision block 300 . A similar set of image frames is acquired during the next RR interval, except that the 10 views acquired for each image frame 306-311 during the subsequent RR interval are interleaved with the previously acquired views. This is illustrated in Figure 11, where the projection view shown by dotted line 230 was taken during one RR interval, the projection view shown by dotted line 232 was taken during another RR interval, and the projection view shown by dotted line 232 was taken during another RR interval. Projection view shown by solid line 234 . During a typical breath hold, 16-20 such interleaved projection views may be acquired.

As described above, the views in each acquired image frame are arranged to sample k-space as uniformly as possible and to satisfy the Nyquist sampling rule out to radius r, as shown in Figure 11. The combined interleaved projections 230 , 232 and 234 also sample k-space as uniformly as possible, they sample k-space more densely, and satisfy the Nyquist sampling rule out to larger radii R. As a result, when data have been collected for all heartbeats within a breath hold as detected at decision block 316 in FIG. 9, as shown at 318 in FIG. The stage acquires quite a few views that are staggered and evenly spaced.

Still referring to FIG. 9 , as indicated by process block 320 , the composite data set 318 is used to reconstruct a composite image for each heartbeat phase. During a typical breath-hold with 16 to 20 heartbeats, the composite image set will contain 16 to 20 times the data contained in a single image frame and can be reconstructed using conventional filtered rear projection techniques Moderately artifact-free image. As indicated by process block 321, the resulting composite image may also be edited or filtered to remove unwanted structures. This can be done manually by displaying the composite image and deleting unwanted structures, or automatically by filtering out detectable structures or tissues.

As indicated by process block 324, a series of image frames can now be reconstructed for each heartbeat phase. The reconstruction process of an image frame will now be described, and an important aspect of the present invention is that a synthetic image of a heartbeat phase can be used to reconstruct the image frame of the heartbeat phase.

With particular reference to FIG. 12, the first step is to transform the image frame k-space projections (ten in the preferred embodiment) into radon space by performing a Fourier transform, as indicated by process block 330. The result is a set of signal profiles 10 as shown in FIG. 5 . As indicated by process block 332, each such signal profile is then rear projected into the VOI as indicated by path 8 in FIG. This rear projection is weighted with the composite image as described above in connection with equation (2). That is, the rear projection value (P) at any pixel (n) in the composite image is weighted by the magnitude (C _n ) of that pixel (n).

Next, the rear projection signal value (S _n ) is added to the image frame being reconstructed, as indicated by process block 334 . The system signal is then returned to decision block 336 to rear project the next signal distribution 10 as shown in process blocks 338 and 332 . Thus, all the signal values (S _n ) in the rear projection signal distribution 10 are added to the aforementioned image frame with a weight determined by the corresponding pixel values in the higher quality composite image. The composite image is of higher quality because it is reconstructed from more projected views and this produces fewer artifacts. The composite image is also of higher quality because the projected views from which it was reconstructed were acquired over a long time span. In general, the SNR of an image frame is proportional to the square root of its acquisition duration. It is the finding of the present invention that through this unique highly constrained reconstruction process, a higher quality composite image is delivered to the above image frames.

Returning to the flowchart in FIG. 9, the image frames for one heartbeat phase are reconstructed from their corresponding composite images, and then the image frames for the next heartbeat phase are reconstructed, as indicated by process block 340. show that. When the image frames for all heartbeat phases have been reconstructed as determined by decision block 342 , they may be displayed in a number of ways as shown by process block 344 .

The scans described above produce a series of image frames during each phase of the heartbeat. This is illustrated in Figure 13, where each reconstructed image frame 345 is associated with a particular heartbeat phase and a particular heartbeat. These image frames 345 can be displayed in many different ways. First, at any selected time point (ie, heartbeat) during breath-holding, an image depicting the subject in successive heartbeat phases may be displayed. If the subject were a human heart, these successive images of the heartbeat phases would show the structure of the heart as it changes during a heartbeat.

During a train of heartbeats, an image 345 of any particular heartbeat phase may also be viewed. In this case, heart motion is frozen, and one can see changes in heart structure over time. This display mode is particularly useful when contrast media is used, and the series of images 345 described above depicts the flow of contrast media into the field of view. Embodiments of this contrast enhancement of the present invention will now be described.

Figure 14 depicts a preferred method of using a contrast agent during an ECG gated scan. This particular embodiment uses two breath holds, and the first step is to guide the patient to establish a first reference breath hold, as shown at process block 350 . This may be accomplished with a monitoring device, such as that described in US Patent 5,363,844 entitled "Breath-hold Monitor For MR Imaging", which provides visual feedback to the patient about breathing motion. As indicated by process block 352, next, during this initial breath hold, a series of cardiac phase image frames are acquired as described above.

Then, contrast agent is delivered as indicated by process block 354 and a second breath-hold is re-established at the reference location as indicated by process block 356 . Using the breath hold monitor described above, the patient inhales and then exhales until the feedback light on the monitor indicates that the reference position has been reached. As indicated by process block 358, another set of cardiac phase image frames is acquired during this second breath hold as the contrast agent flows into the field of view.

As indicated at process block 360, for each heartbeat phase, an anterior contrast mask image is generated. This is accomplished by combining all projections acquired during each heartbeat phase during the first breath hold; and performing a conventional filtered rear-projection image reconstruction process with the combined projections. For example, if 20 heartbeats occur during this first breath hold and 10 interleaved projection views are acquired during each heartbeat phase, a total of 10 x 20 = 200 projection views are used to reconstruct each mask image. Thus, a mask image 362 is generated for each heartbeat phase, as shown in FIG. 15 .

Next, for each heartbeat phase, an unmasked composite image is reconstructed, as indicated by process block 364 . This is achieved by combining all the interlaced projections acquired during each heartbeat phase during the second post-contrast breath hold; and performing a conventional Filtered rear projection image reconstruction process. Thus, an unmasked composite image 366 is produced for each heartbeat phase, as shown in FIG. 15 .

Now, as indicated by process block 368, for each heartbeat phase, a final composite image is generated. This is accomplished by subtracting the masked image 362 for each heartbeat phase from its corresponding unmasked composite image 366 . As shown in FIG. 15, thus, for each heartbeat phase, a composite image 370 is generated. These masked composite images 370 indicate those image pixels that undergo intensity changes due to the arrival of contrast agent, which may be arteries in MRA studies or ventricles in cardiac studies. It should be apparent that the same masked composite image 370 can also be produced by subtracting the front-contrast projection view from the corresponding rear-contrast projection view; and reconstructing the masked composite image 370 from the different projection views .

Now, as indicated by process block 372, a set of image frames may be reconstructed for each heartbeat phase. This is accomplished using a masked composite image 370 for each heartbeat phase and the projections acquired for that heartbeat phase during each heartbeat as described above and shown in FIG. 12 . To generate the desired sparse data set for this highly constrained rear-projection reconstruction process, the corresponding masked projection view is subtracted from the projection view used to reconstruct the image frame. As shown in FIG. 15 , thus, for each heartbeat and for each heartbeat phase, an image frame 374 is reconstructed. These can be displayed in many different ways, as discussed above and shown at process block 380 .

Another preferred embodiment of the present invention uses the multi-slice capability of the hybrid PR pulse sequence in FIG. 7 to acquire multiple image frames during each heartbeat and each heartbeat phase. Multiple adjacent slices provide a three-dimensional volume from which a maximum intensity projection (MIP) image can be generated. This is important when the structure being imaged does not lie exactly in one two-dimensional plane. This multi-slice embodiment can be used in the contrast-enhanced acquisition process described above in connection with FIG. 14 , but a multi-slice embodiment without contrast enhancement will now be described in connection with FIG. 16 . This embodiment is similar in many respects to the embodiment described above in connection with FIG. 10 , and in FIG. 16 substantially identical steps are identified with the same reference numerals.

With particular reference to Figures 16 and 17, a cardiac triggered scan is performed in which a series of image frames are acquired during multiple heartbeat phases during each heartbeat in a breath hold. During each heartbeat, a series of image frames 306-311 are acquired at process block 301, similar to above. However, instead of acquiring a single set of 10 projection views for sampling a single k-space slice, two sets of 10 projection views are acquired at each heartbeat phase. Each group of 10 projected views was subjected to a different phase encoding process along the axial gradient (Fig. 7), such that two adjacent 2D slices of k-space were sampled. As will now be described, at each heartbeat phase, an image frame consisting of three two-dimensional slices is finally formed. Referring to Figure 8, these include a central k-space slice 206 (hereinafter referred to as "A") and two peripheral k-space slices 204 and 208 (hereinafter referred to as "B" and "C"). However, in order to reduce the scan time required to acquire each image frame, only two of the above three slices are acquired during any single heartbeat phase.

As best shown in FIG. 17, the central slice A is captured together with one of the peripheral slices B or C during each frame of image acquisition during a particular cardiac phase. The sampling is such that, at any heartbeat phase, the central slice A is acquired, one of the peripheral slices B or C is acquired, and the other peripheral slices can be generated from temporally adjacent acquisitions. For example, in FIG. 17, A and B slices are captured for the first heartbeat phase image frame 380 captured during the nth heartbeat of a breath hold during 20 repetitions of the pulse sequence of FIG. A and C slices are acquired at 382 and 384 during the same heartbeat phase of n-1 and n+1 heartbeats. Thus, when reconstructing the image frame for the first heartbeat phase of the nth heartbeat, by interpolating between temporally adjacent C slices acquired at 382 and 384 during n-1 and n+1 heartbeats , then calculated slice C data.

Referring again to FIG. 16 , after all image frames for all heartbeats in a breath hold have been acquired as determined at decision block 316, a composite set of images is reconstructed from the acquired data as indicated at process block 323 . First, a one-dimensional fast Fourier transform is performed along the axial phase-encoding axis of the acquired k-space data set. As shown in FIG. 8 , the obtained mixed spatial data sets are all composed of three axial projection view pictures (A, B and C). For each slice A, B, and C, three composite images of each heartbeat phase are reconstructed. More specifically, for each heartbeat phase, slice A projections collected throughout the breath hold period are combined into composite A data set 390 for the second heartbeat phase as shown in FIG. 17 , and slice B projections are combined into composite The B data set 392 , slice C projections are combined into a composite C data set 394 . Those projections that are combined to form the composite data sets 390, 392, and 394 are interleaved as described above with reference to FIG. . Importantly, during this heartbeat phase, composite data sets 390, 392, and 394 sample slices A, B, and C more densely than any captured image frame. As a result, a composite image can be reconstructed from the composite data sets 390, 392, and 394 formed for each heartbeat phase using conventional image reconstruction methods. Each of these three slices is individually reconstructed using conventional 2D image reconstruction methods. This could be a filtered backprojection process or a reframe process of 2D projections in each slice followed by a 2D Fast Fourier Transform.

Following the reconstruction process of the composite image, as indicated by process block 325, the image frames for each heartbeat phase are reconstructed. As in the embodiments described above, this image reconstruction process uses a highly constrained rear projection method and composite images in order to improve SNR and reduce image artifacts in highly undersampled image frames.

The reconstruction process of each image frame can be implemented in many different ways. First, a limited set of A, B, and C phase-encoded projection views for one heartbeat phase can be Fourier transformed along the axial phase-encode gradient axis to form three slices. Each of the three slices is constructed from a finite set of projected views, from which a slice image is reconstructed using the method shown in Figure 12. That is, each projected view is Fourier transformed into Radon space, and then it is backprojected using the composite image of the slice to weight each backprojection value. Three adjacent image slices are thus reconstructed for each heartbeat phase during each heartbeat and these can be displayed as a three-dimensional image. Furthermore, a Maximum Intensity Pixel Projection (MIP) image can be generated from the three-dimensional image.

In the reconstruction method described above, Fourier transform is first performed along the axial phase-encoding gradient axis. This is preferred when the A, B and C phase-encoded projection views in the reconstructed image frame do not interleave each other. If they are interleaved, an alternative approach is preferred, in which each A, B, and C phase-encoded projection views are individually rear-projected. The resulting data set is then reframed to align those samples along the axial phase-encoding axis. A Fourier transform along the axial phase encoding gradient axis is then performed on the resulting reframed mixed data set.

Instead of converting the image reconstruction process into a reconstruction process of three two-dimensional slices, the three-dimensional reconstruction process of the A, B and C phase-encoded data sets can be directly performed in each heartbeat phase. A number of methods are described in co-pending US patent application Ser. No. 11/482,372, filed July 7, 2006, entitled "HIGHLY CONSTRAINED IMAGE RECONSTRUCTION METHOD." These methods are incorporated herein by reference.

Claims (29)

1. the method for an experimenter who is used for producing the field of view (FOV) be positioned at Magnetic resonance imaging (MRI) system image comprises the steps:

A) at each time between heart beat period and of heartbeat repeatedly in each heart phase of a plurality of heart phase, with one group of projection view of MRI system acquisition;

B) projection view that is collected in the corresponding heart phase with repeatedly heartbeat produces a composograph to each heart phase, and each composograph is used to be depicted in the experimenter in the corresponding heart phase;

C) come the experimenter of reconstruct in selected heart phase image by following process:

C) i) will select the one group of projection view that is collected in the heart phase and be backprojected among the FOV, and use the value of corresponding pixel in the composograph that is depicted in the experimenter in the selected heart phase that the value that is backprojected in each image pixel is weighted; With

C) ii) the rear-projection value of each image pixel is sued for peace.

2. the method for claim 1 is characterized in that, the projection view that is collected in each heart phase interlocks.

3. the method for claim 1 is characterized in that, at step c) i) in calculate each image pixel rear-projection value S by following formula _n:

$S_n=(p\&times;C_n/\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}{C}_n$

Wherein, P=is by the projection view value of rear-projection;

C _nCorresponding pixel value in the=composograph;

S _n=along the value of n pixel of backprojection path; And

N=is along the sum of the pixel of backprojection path.

4. the method for claim 1 is characterized in that, step b) comprises editor's composograph so that the probability of therefrom removing an object and this object being appeared in the image that reconstructs reaches minimum basically.

5. the method for claim 1 is characterized in that, step c) is included in carries out earlier this projection view being carried out Fourier transform before the rear-projection to each projection view.

6. the method for claim 1 is characterized in that, in each heart phase of a plurality of heart phase between heart beat period each time of heartbeat repeatedly, and repeating step c) so that produce image.

7. the method for claim 1 is characterized in that, under the guidance that mixes 2D PR pulse train execution in step a), this all gathers the projection views of organizing more in each heart phase, these projection views are used to describe corresponding a plurality of of experimenter.

8. method as claimed in claim 7 is characterized in that, step b) is included in the composograph that produces each sheet in each heart phase.

9. method as claimed in claim 8 is characterized in that step c) comprises: each sheet place in selected heart phase, utilize with selected heart phase and the corresponding composograph of sheet and come reconstruct experimenter's image.

10. method as claimed in claim 9 is characterized in that, the number of the picture of institute's reconstruct is greater than the number of the projection view group of utilizing mixing 2D PR pulse train to collect in each heart phase in the heart phase of selecting in step c).

11. method as claimed in claim 10 is characterized in that, also comprises: in selected heart phase, utilize the projection view that does not collect during this selected heart phase to produce one group of additional projection view.

12. method as claimed in claim 11, it is characterized in that, before the heartbeat of this selected heart phase and between twice heart beat period afterwards,, calculate one group of additional projection view by carrying out interpolation between the many groups projection view that in same heart phase, is collected.

13. the method for an experimenter who is used for producing the field of view (FOV) be positioned at Magnetic resonance imaging (MRI) system image comprises the steps:

A) in each heart phase of a plurality of heart phase between heart beat period each time of heartbeat repeatedly, with a plurality of views of this experimenter of MRI system acquisition;

B) utilize in step a) the view that collects from repeatedly heartbeat and selected heart phase to come the reconstruct composograph, this composograph has a numerical value at each composograph pixel place, and this numerical value is used for being illustrated in the experimenter of inherent this pixel position of FOV of selected heart phase; And

C) come this experimenter's of reconstruct image by following process:

C) i) in the view that is collected in the heart phase of from step a), selecting, produce image data set; With

C) ii) utilize this image data set and highly constrained rear projecting method to produce this experimenter's image, this highly constrained rear projecting method is weighted with the image pixel of corresponding pixel value in the composograph to each rear-projection.

14. method as claimed in claim 13, it is characterized in that, step a) is included in the projection view that each heart phase is gathered a plurality of phase encodings, step b) is included in the composograph that selected heart phase reconstruct is used for each phase encoding, and uses these composographs to come this experimenter's of reconstruct described image in step c).

15. method as claimed in claim 13, it is characterized in that, the view that collects in the heart phase of selecting in step a) is the projection view that collects by staggered projected angle between continuous heart beat period, at step c) i) in the image data set that produced comprise and select one group of described projection view, and in step b) from reconstructing composograph all staggered projection views basically.

16. method as claimed in claim 13, it is characterized in that, in step a), gather view in a plurality of selected heart phase, in step b), therefrom reconstruct corresponding a plurality of composograph, and in step c), reconstruct a plurality of images in corresponding a plurality of heart phase.

17. method as claimed in claim 13 is characterized in that, produces a plurality of images the view that collects in the corresponding heart phase of repeatedly selecting between heart beat period in step c).

18. method as claimed in claim 13 is characterized in that, step c) ii) comprises makes each image pixel normalization.

19. a method that produces the cardiac gated image of contrast enhancing with Magnetic resonance imaging (MRI) system comprises the steps:

A) in following the selected heart phase of gating signal and at each time between heart beat period, of heartbeat repeatedly with one group of view of MRI system acquisition;

B) reconstruct masked images all basically views that in selected heart phase, collect;

C) supply with contrast preparation;

D) in following the selected heart phase of gating signal and at each time between heart beat period of heartbeat repeatedly, contrast view with after one group of the MRI system acquisition;

E) reconstruct unshielded composograph all basically views that in selected heart phase, collect;

F) by from unshielded composograph, deducting masked images, produce composograph; And

G) reconstruct a two field picture the contrast view after in heart phase selected between heart beat period, collect one group, wherein use composograph that each pixel in this two field picture is weighted.

20. method as claimed in claim 19, it is characterized in that, described view is a projection view, and uses a kind of highly constrained rear projecting method to come execution in step g), this highly constrained rear projecting method comprises: the back contrast view that collects in selected heart phase is carried out rear-projection; And the rear-projection value is weighted with corresponding value in the composograph.

21. method as claimed in claim 19 is characterized in that, the view that collects in step d) interlocks.

22. method as claimed in claim 19 is characterized in that, at the repeating step g between heart beat period each time of heartbeat repeatedly) so that reconstruct one two field picture.

23. method as claimed in claim 19 is characterized in that, in another selected heart phase, repeating step d), e), f) with g) so that produce a two field picture.

24. method as claimed in claim 23 is characterized in that, heartbeat repeatedly each time between heart beat period in each heart phase of a plurality of selected heart phase, repeating step g) so that reconstruct one two field picture.

25. a method that produces the cardiac gated image of contrast enhancing with Magnetic resonance imaging (MRI) system comprises the steps:

A) in following the selected heart phase of gating signal and at each time between heart beat period of heartbeat repeatedly, with contrast projection view before one group of the MRI system acquisition;

B) supply with contrast preparation;

C) in following the selected heart phase of gating signal and at each time between heart beat period of heartbeat repeatedly, contrast projection view with after one group of the MRI system acquisition;

D) from corresponding back contrast projection view, deduct preceding contrast projection view;

What e) collect in selected heart phase produces composograph all projection views through deducting basically; And

F) reconstruct a two field picture the one group of projection view through deducting that collects in the heart phase of using a kind of highly constrained rear projecting method to select between a heart beat period, this highly constrained rear projecting method comprises: the projection view through deducting that collects in selected heart phase is carried out rear-projection; And the rear-projection value is weighted with corresponding value in the composograph.

26. method as claimed in claim 25 is characterized in that, the view that collects in step d) interlocks.

27. method as claimed in claim 52 is characterized in that, at the repeating step f between heart beat period each time of heartbeat repeatedly) so that reconstruct one two field picture.

28. method as claimed in claim 25 is characterized in that, in another selected heart phase, repeating step c), d), e) with f) so that produce a two field picture.

29. method as claimed in claim 28 is characterized in that, heartbeat repeatedly each time between heart beat period in each heart phase of a plurality of selected heart phase, repeating step f) so that reconstruct one two field picture.

Images