On the cause of spatial displacement of long T₁ species in segmented inversion recovery prepared imaging

Abstract

Using inversion recovery steady-state free precession segmented k-space imaging for the detection of myocardial infarction, we noticed that some structures appeared in the wrong locations of the image. In this work, the spatial displacement is demonstrated and explained from both theoretical and experimental points of view. The effect is due to a change in phase from segment to segment of the detected magnetization from species with long T 1 's such as cysts, fluid collections, and cerebrospinal fluid. Depending on the number of k-space segments and view ordering, structures can be replicated throughout the image or displaced by half of the phase-encoding field of view. Attempts to reduce acquisition times with short repetition time (TR) techniques have always been limited by signalto-noise constraints. Hence, the strong signal available with short TRs accounts for the increased use of fully refocused gradient-echo techniques (1) (called steady-state free precession (SSFP), fast imaging with steady-state freeprecession, fast imaging employing steady-state excitation, or balanced turbo fast-field-echo). Gradient refocusing increases signal through the reuse of magnetization from repetition time to repetition time and provides images with contrast dependent on the ratio of the T 2 relaxation time to the T 1 relaxation time.With the use of T 1 shortening contrast agents, although T 2 * weighted sequences can be used, one typically chooses a T 1 weighted pulse sequence. To introduce T 1 weighting into an SSFP pulse sequence, a preparatory 90°radiofrequency pulse (saturation) or 180°radiofrequency pulse (inversion) is used prior to the refocused gradient-echo pulse sequence. Applications to date include MR angiography (2), arterial spin-labeling (3), first-pass myocardial perfusion (4), and delayed-hyperenhanced imaging (5).To overcome artifacts associated with physiologic motion or improve the efficiency of preparatory pulses such as saturation or inversion recovery (IR) and chemical fat saturation, a technique termed k-space segmentation is used (6). k-space segmentation introduces a repeated MR pulsed experiment. If full magnetization recovery does not occur between segments as data collection is repeated, the observed signal will be dependent on the magnetization state at the end of the previous segment. When an inversion pulse is used, this incomplete recovery may cause alternating longitudinal magnetization available at the beginning of each segment. A change in phase from segment to segment of the detected magnetization of species with long T 1 's such as cysts, fluid collections, and cerebrospinal fluid will result. Species with shorter T 1 's, such as blood and myocardium, recover longitudinal magnetization more quickly and the detected magnetization remains in phase. This change in phase from segment to segment causes a predictable spatial displacement when an interleaved acquisition is used. In this work, this phenomenon and its effect on the final image are demonstrated and explained from both theoretical...

Cardiac imaging techniques for physicians: Late enhancement

“…Fluid regions such as pericardial effusion, cysts, or cerebral spinal fluid (CSF) with long T1 relaxation time may lead to ghosting in segmented protocols, and will appear bright on magnitude reconstructed images which might be confused with fat. The PSIR method is shown to correct the polarity of pericardial effusion (69, 70), and the use of cardiac surface coil arrays with adaptive coil combining has been shown to reduce CSF ghosting (71) which in some instances might be confused with focal enhancement. Single shot methods avoid the problem of ghosting.…”

Magnetic resonance separation imaging using a divided inversion recovery technique (DIRT)

“…Initially, it was reported that this artifact was caused by oscillations in the transient approach to steady state for regions with long T 1 relaxation times (6). It was later shown through computer simulations, phantom experiments, and human studies that this artifact is due to a change in phase from k-space segment to k-space segment of the detected magnetization from species with long T 1 relaxation times such as cysts, fluid collections, and cerebrospinal fluid (7). Depending on the number of k-space segments and view ordering, structures can be replicated throughout the image or displaced by half of the phase-encoding field of view.…”

“…If phase encoding is performed such that interleaving of kspace segments (odd and even k-space lines) is done prior to image reconstruction, species with long T 1 relaxation times will have an alternating phase in the raw data reconstruction matrix. This is equivalent to multiplication by À1 on alternating rows of the raw data matrix (7). This k-space data multiplication is commonly used in MR image reconstruction (often called fft_shift, swapping, or chopping) and is equivalent to shifting the reconstructed image by half of the field of view.…”

“…The cause of the displacement of long T 1 species in a segmented IR-bSSFP pulse sequence follows directly from the evolution of the magnetization during excitation and recovery (7). Figure 1 shows a schematic of the pulse sequence developed in this paper.…”

“…Recently, an artifact was identified during late gadolinium-enhanced myocardial infarct imaging using an inversion recovery balanced steady-state free precession (IR-bSSFP) technique (6,7). This artifact is a spatial displacement of tissues and was shown to be due to a magnetization phase change resulting from incomplete magnetization recovery between subsequent inversions.…”