Surgery – Diagnostic testing – Detecting nuclear – electromagnetic – or ultrasonic radiation
Reexamination Certificate
1998-06-26
2001-05-29
Lateef, Marvin M. (Department: 3737)
Surgery
Diagnostic testing
Detecting nuclear, electromagnetic, or ultrasonic radiation
C324S307000, C324S309000
Reexamination Certificate
active
06240310
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to magnetic resonance (MR) imaging, and more particularly relates to cardiac MR studies of human patients. In its most immediate sense, the invention relates to acquisition of MR cardiac cine data sets for rapidly assessing the function of a human patient's left ventricle.
To make such an assessment, it is necessary to visualize the left ventricle and to distinguish the myocardium from the left ventricular blood pool during all phases of the cardiac cycle. Conventionally, this is done by acquiring a series of temporally spaced-apart images of the left ventricle (such a series of images is referred to as a “cine” acquisition). However, conventional cine acquisitions have a number of drawbacks.
One such drawback is that conventional cine acquisitions are slow. When using a conventional gradient-echo pulse sequence such as the two-dimensional FLASH pulse sequence that is presently implemented on the MAGNETOM® Vision MR imager manufactured by Siemens AG, a study to assess the function of the entire left ventricle can take 10 to 20 minutes to complete.
Another drawback is that a conventional cine acquisition can produce images of low quality. In a conventional cine acquisition, the images produced are actually composite images formed by combining information acquired during multiple cardiac cycles. Thus, the quality of the images produced by a conventional cine acqusition depends upon the degree to which the patient's cardiac rhythm remains invarient during the course of the acquisition. Additionally, if the acquisition takes place during free breathing, the heart moves together with the diaphragm. This results in blurring and image artifacts.
Although these drawbacks can be reduced by using various measures, such drawbacks have not been overcome. For example, image artifacts caused by respiration can be reduced by having the patient hold his or her breath and performing a cine acquisition for a single slice position within a single breath-hold using rapid image acquisition techniques. However, some patients are unable to hold their breath repeatedly, as they must do to acquire the necessary series of images. And, because assessment of the function of the entire left ventricle requires a series of breath-holds, changes in the patient's breath-hold positions will cause the images of the various slices to be misregistered.
Likewise, rapid two-dimensional gradient-echo techniques can produce a series of images in which each one is acquired fast enough to eliminate the effects of respiratory motion. However, due to the current limitations of gradient hardware the temporal resolution of these techniques is not adequate to assess changes in left ventricular function or wall motion.
Furthermore, known cine acquisition techniques require accurate and consistent ECG gating to produce adequate image quality and such gating is not always feasible. For example, the environment within the magnet of an MR imager is often unsuitable for establishing adequate ECG gating for cardiac patients. And, even when ECG gating is successful, arrhythmias can cause cycle-to-cycle variations in the MR signal, resulting in additional image artifacts.
It would therefore be advantageous to provide a method for conducting an MR study, and particularly a cardiac MR study for evaluating the function of a human patient's left ventricle, which could be completed rapidly, which could produce images of adequate quality, which would not require ECG gating, and which would not require multiple breath-holds.
The invention proceeds from the realization that spatial data can be shared from one image to the next to produce additional images and to therefore reduce the temporal spacing between images. (It is alreadly known to share data in some of the lines of the MR data or k-space matrix. The extra images thereby created have been called “echo-shared images”. In the past, such echo-shared images have been generated only in combination with a segmented k-space acquisition.) The invention proceeds from the further realization that the each image, including the shared images, should include data not used in any other image, if each image is to provide unique information. The center line of the k-space matrix (i.e. the line of MR data acquired at a phase encoding of zero) is the line that provides the most MR signal for an image. Therefore, acquiring an additional center line for each image can efficiently provide unique information for each image, with a small increase in acquisition time. The invention still further proceeds from the realization that if there are large discontinuities in the MR data used to reconstruct an image, the image will contain artifacts. Therefore, it is necessary to make sure that such discontinuities do not arise.
In accordance with the invention, in addition to acquiring sufficient MR data to completely fill the k-space matrix for a particular image, one additional line of MR data is acquired. This line is the one acquired at a phase-encoding gradient of zero. This line is then, together with lines of data acquired for the current image and the next image, used to generate a new “shared” image. If the lines of MR data were to be acquired using a standard linear traversal through k-space, there would be a large discontinuity near the center of k-space due to motion occurring during the acquisition, and this discontinuity would cause artifacts in the shared image. Additionally, the k-space trajectory would be drastically different between the original images and the shared images. To avoid this discontinuity and this difference, the phase-encoding gradient is varied stepwise in a pattern that oscillates between a minimum and a maximum. Each oscillation (from minimum to maximum and from maximum to minimum) includes the acquisition of the zero phase-encoding line.
As will become evident below, such a data acquisition scheme allows many lines of MR data to be reused in reconstruction of multiple images. And, the time penalty caused by acquisition of additional lines of MR data at a zero phase-encoding is very small. In accordance with the preferred embodiment, MR data acquired during an interval of approximately 290 mS can be used to reconstruct 3.5 images, while a conventional image acquisition technique can only acquire enough data to reconstruct 2 images during that period of time. This improvement in temporal resolution provides adequate visualization of the function and the wall motion of the left ventricle.
In accordance with the preferred embodiment, a method in accordance with the invention is used in a gradient-echo MR pulse sequence in combination with a single breath-hold. In appropriate instances, a contrast agent may be administered to the patient and the acqusition of data may be gated to the patient's cardiac cycle.
REFERENCES:
patent: 4710717 (1987-12-01), Pelc et al.
patent: 4939463 (1990-07-01), Sekihara et al.
patent: 5195525 (1993-03-01), Pelc
patent: 5251628 (1993-10-01), Foo
patent: 5474067 (1995-12-01), Laub
patent: 5653233 (1997-08-01), Pelc et al.
patent: 5692508 (1997-12-01), Simonetti et al.
patent: 6009341 (1999-12-01), Edelman
Bundy Jeffrey M.
Simonetti Orlando P.
Jay Mark H.
Lateef Marvin M.
Mantis Mercader Eleni
Siemens Medical Systems Inc.
LandOfFree
Method of acquiring MR data especially for cardiac cine data... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of acquiring MR data especially for cardiac cine data..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of acquiring MR data especially for cardiac cine data... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2533913