Image analysis – Applications – Biomedical applications
Reexamination Certificate
2001-04-13
2004-04-20
Johnson, Timothy M. (Department: 2625)
Image analysis
Applications
Biomedical applications
C382S274000, C382S264000, C382S132000, C382S218000, C600S410000
Reexamination Certificate
active
06724923
ABSTRACT:
BACKGROUND OF INVENTION
The present invention relates generally to MR data acquisition and, more particularly, to a method and system for imaging with adaptive automated coil selection of a multi-coil imaging receiver assembly in a medical imaging device.
Diagnostic imaging devices, such as magnetic resonance (MR) scanners, can acquire imaging data using a series of receivers. Further, improved imaging devices can employ a coil assembly of phased array coils to acquire the imaging data over a desired imaging field-of-view (FOV). Phased array coils are often used because they yield a higher signal-to-noise ratio (SNR) and increased spatial coverage over the desired FOV. In these known imaging systems, imaging data acquired from each phased array coil is combined to form a final diagnostic image. Despite the implementation of phased array coils, reconstructing an image from the combination of the imaging data from all, or too many, of the phased array coils can produce ghosting in the final image from unwanted noise and artifacts.
Typically, these unwanted artifacts result from the acquisition of data from a phased array coil assembly that dimensionally exceeds the desired FOV of a subject, i.e., a medical patient. For example, in a known MR system that utilizes a coil assembly with six phased array coils, it is customary to utilize four of the six coils during a spinal imaging examination. However, in some spinal exams, the imaging FOV can be covered with only two or three coils to acquire sufficient data to produce a complete image. As a result of using the additional and unnecessary coils, noise and artifacts insensitive to the selected FOV are often included in the final image and result in undesirable ghosting in the final image. To reduce the noise and unwanted artifacts, it is desirable to use only those coils that are sensitive to the desired FOV.
Known systems seek to maintain sensitivity of the FOV by permitting manual selection of coils based upon patient positioning and other positioning tools prior the imaging session. Essentially, in these known diagnostic systems an MR operator or technician manually deselects coils so that these deselected coils do not acquire data during the imaging scan or, alternatively, specifically excludes data from arbitrary coils that are acquired during scanning from final image reconstruction on a trial and error basis. To properly deselect the appropriate coils, the operator must know exactly which coils to deactivate prior to the diagnostic scan, which is a difficult and time consuming task, and often requires guess work. The task of properly selecting the coils is exaggerated by the fact that the patient and patient table can frequently change positions during an imaging session. Further, deselection of the proper coils is a cumbersome task and one prone to human error that could result in requiring a total rescan. To require the operator to repeatedly deselect coils with each new patient and/or table position only increases the difficulty of deselecting the appropriate coils. Further, with multi-slice imaging techniques it would be necessary for the operator to deselect different coils during the imaging session—a daunting task in operator-based selection.
Attempts to automate coil selection have, in fact, not been made completely automatic. For example, in one approach the coils are selected only after requiring a system operator to furnish the scanner with 2-3 parameters that are based on the positioning of a landmark, or reference mark, placed on the subject at some known distance from isocenter. Such a system requires accurate placement of the landmark, or requires a measurement of the distance between the landmark and isocenter, and that measurement must be input into the system together with a dimension of the field-of-view between a pair of boundary limits, relative to the isocenter. Further, after the operator inputs the necessary parameters, some such known systems utilize conventional logic circuits together with a lookup table to select or deselect certain coils. Other known systems use a pilot scan, or prescan, to acquire parameters of a selected slice or imaging sequence and compare these properties to properties of the coil array that are stored in a lookup table. Not only do these systems rely on static data, they are subject to human error and/or require additional hardware configurations, and in the case of using markers, the images have to either be manually gridded or additional time and resources are expended on automatic gridding if the marker is machine identifiable. Examples of two coil selection systems include U.S. Pat. Nos. 5,138,260 and 6,134,465. Although such systems have functioned adequately, it would be desirable and advancement in the art, to design a fully automated system that does not rely on manual intervention, lookup table, and/or landmarks.
It would therefore be desirable to design a method and system to fully automate coil selection of an imaging device, or data from specific coils, to increase FOV sensitivity and clarity for multi-channel phased array imaging. It would similarly be desirable to design a method and system to determine phased array coil positions adaptively using an on-the-fly index gauge determination that is not dependent on a status lookup table, thereby facilitating selection of coils for image reconstruction with reduced artifacts.
SUMMARY OF INVENTION
This invention includes a method and system that is an adaptive and full-automated coil selection technique that overcomes the aforementioned drawbacks. The present invention facilitates image reconstruction with an increased signal-to-noise ratio and reduced artifact presence without manual selection of specific coils, without the use of reference marks and landmarks, and without using look-up tables. Through automatic selection and/or deselection of coils, the present invention discriminates between coils in an imaging field-of-view (FOV) based on an index gauge. Using imaging data from those coils with the best sensitivity within the FOV enables reconstruction of a final image with reduced noise and reduced artifacts from the image. The present invention implements automatic selection or deselection of coils based on an adaptive and dynamic analysis rather than reliance on manual selection, operator inputs, and/or landmark placement processes often prone to human error and requiring additional hardware.
In one embodiment of the present invention, preliminary scanning data is acquired during an initial fast scan. With a fast scan, a limited number of k-space data lines are acquired and sampled. By employing a fast scan, the present invention allows for a preliminary analysis to be completed to determine those phased array coils most sensitive to a FOV before acquiring a full set of imaging data. Typically, a fast scan or preliminary scan requires less time to complete than a full reconstruction imaging scan. After the preliminary imaging data is acquired during the fast scan, the imaging data is analyzed and evaluated to determine which coils to activate during a full reconstruction imaging session. Analyzing and evaluating the preliminary scan to determine which coils to activate or deactivate during a full imaging session may be carried by adaptively determining an index gauge for at least one k-space data line acquired for each coil, comparing the index gauge on-the-fly, and then automatically selecting or deselecting a number of MR coils to activate for image data acquisition based on the comparison of the dynamically determined index gauge.
In accordance with one aspect of the present invention, a method to automatically select/deselect coils, for image reconstruction, using prescan image data of an MRI device having a number of coils is provided. The method includes the step of initializing a fast scan to acquire k-space data sets via a number of phased array coils. The number of k-space data sets includes at least one k-space data line for each of a plurality of MR coils. In evaluating the k-space data, an index g
Ma Jingfei
Tan Guosheng
Choobin Barry
Della Penna Michael A.
GE Medical Systems Global Technology Co. LLC
Horton Carl B.
Johnson Timothy M.
LandOfFree
Automatic coil selection of multi-receiver MR data using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic coil selection of multi-receiver MR data using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic coil selection of multi-receiver MR data using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3266873