Analysis of cardic MR relaxation time images with application to quantifying myocardial perfusion reserve indexes
AssigneesGE Medical Systems Global Technology Company, LLC, The John Hopkins University
InventorsSandeep N. Gupta, Garth M. Beache, Daniel A. Herzka
Life of this patent
Estimating the term…
Abstract
The present invention is a method and apparatus for semi-automatically registering low signal to noise ratio relaxation-time images, with particular application to tracking motion in moving organs and quantifying myocardial perfusion reserve. For each image acquired in a time series, a corresponding high signal to noise ratio image is extrapolated from the collected data. The high signal to noise ratio images are registered to track motion, and the resulting registration data is copied onto the corresponding relaxation time image.
Claims22
A method for deriving registered time series data from low signal to noise ratio relaxation time images, and particularly of images affected by motion during long acquisition times, the method comprising the steps of:
- acquiring a time series of relaxation time MR images;
- extrapolating a high signal to noise ratio image from each of the time series of relaxation time MR images;
- registering each of the successive high signal to noise ratio images to obtain registration data;
- applying the registration data from each of the high signal to noise ratio images to the corresponding relaxation time MR images.
A method for performing blood oxygen level dependent magnetic resonance imaging to analyze oxygenation changes in a region of interest, the method comprising the following steps:
- applying an MRI scan protocol to obtain a transverse relaxation time image of a body of interest at timed intervals to obtain a time series of images, the scan protocol including a plurality of echo pulses for acquiring a series of images for each scan,
- applying a curve fitting technique to the series of images acquired from each scan protocol to construct a transverse relaxation time image for the scan, and to extrapolate a proton density image of the body of interest at an echo time substantially equal to zero;
- registering successive proton density images in the time series;
- correlating the registration data to the corresponding T2* images;
- identifying a region of interest on each of the T2* images;
- and
- analyzing a signal intensity of the images to determine oxygenation changes in the blood.
A method for analyzing regional myocardial perfusion reserve using blood oxygenation level dependent magnetic resonance imaging, the method comprising the following steps:
- acquiring a first time series of relaxation time images indicative of blood flow-state in the heart when a patient is at rest;
- acquiring a first plurality of high signal to noise ratio images indicative of the internal organs, tissue, and skeletal structure of the region being imaged, wherein each of the plurality of high signal to noise ratio images corresponds to one of the first plurality of relaxation time images;
- registering successive images of the first plurality of high signal to noise ratio images;
- mapping the registration data determined from the first plurality of high signal to noise ratio images to the corresponding first time series of MR images;
- identifying a region of interest on each of the first time series of relaxation time images;
- acquiring a second time series of relaxation time images indicative of blood flow-state in the heart when a patient is at stress;
- acquiring a second plurality of high signal to noise ratio images indicative of the location of heart tissue, wherein each of the second plurality of magnitude images corresponds to one of the second time series of relaxation time images;
- registering successive of the second plurality of high signal to noise ratio images;
- mapping the registration data determined from the second set of high signal to noise ratio images to the corresponding image of the second time series of MR images;
- identifying a region of interest on each of the second time series of MR images;
- comparing a signal intensity from the first time series of images and the second time series of images to quantify myocardial perfusion reserve.
Dependencies are inferred from the claim text.