Inversion-recovery-prepared SSFP for cardiac-phase-resolved delayed-enhancement MRI.

Delayed-enhancement magnetic resonance imaging (DE-MRI) can be used to visualize myocardial infarction (MI). DE-MRI is conventionally acquired with an inversion-recovery gradient-echo (IR-GRE) pulse sequence that yields a single bright-blood image. IR-GRE imaging requires an accurate estimate of the inversion time (TI) to null the signal from the myocardium, and a separate cine acquisition is required to visualize myocardial wall motion. Simulations were performed to examine the effects of a steady-state free precession (SSFP) readout after an inversion pulse in the setting of DE-MRI. Using these simulations, a segmented IR-SSFP sequence was optimized for infarct visualization. This sequence yields both viability and wall motion images over the cardiac cycle in a single breath-hold. Viability images at multiple effective TIs are produced, providing a range of image contrasts. In a study of 11 patients, IR-SSFP yielded infarct sizes and left ventricular ejection fractions (LVEFs) similar to those obtained by IR-GRE and standard SSFP, respectively. IR-SSFP images yielded improved visualization of the infarct-blood border because of the simultaneous nulling of healthy myocardium and blood. T(1) (*) recovery curves were extracted from IR-SSFP images and showed excellent qualitative agreement with theoretical simulations.

Intracranial arteriovenous malformations: real-time auto-triggered elliptic centric-ordered 3D gadolinium-enhanced MR angiography--initial assessment.

Auto-triggered elliptic centric-ordered three-dimensional (3D) gadolinium-enhanced magnetic resonance (MR) angiography was compared with 3D multiple overlapping thin-slab acquisition time-of-flight (TOF) MR angiography in the evaluation of intracranial arteriovenous malformations (AVMs) in 10 patients. Intraarterial digital subtraction angiography (DSA) was the reference standard. Gadolinium-enhanced MR angiograms were found to be equivalent to DSA images in AVM component depiction in 70%--90% of cases and were consistently superior to TOF MR angiograms.

Monitoring blood oxygen state in muscle microcirculation with transverse relaxation.

Oxygen uptake from the microcirculation is a direct measure of tissue function. Magnetic resonance is capable of detecting differences between oxygenated and deoxygenated blood due to the paramagnetic properties of deoxyhemoglobin. At the level of the microcirculation, however, imaging methods cannot directly visualize the vessels. Instead, bulk MR parameters are investigated for their ability to monitor blood oxygen saturation (%O(2)) changes in the microcirculation of tissue, specifically skeletal muscle. Experiments in an in vitro model verified the feasibility of detecting changes in exponential decay signals, and also verified the prediction of only two distinct decay components. Experiments in a rabbit model demonstrate that T(2)' and monoexponential T(2) decay are not sensitive to blood oxygen changes, but that the long-T(2) component in a biexponential fit is correlated to the blood oxygen state. Assuming a two-pool model for water protons in muscle, and with knowledge of the T(2)-%O(2) relation, estimates of the microcirculation blood oxygen state can be made with some reasonable assumptions. Magn Reson Med 45:662-672, 2001.

Partial discrete Fourier transform (PDFT) multiband encoding.

For conventional multiband encoding techniques such as Hadamard encoding, scan time scales linearly with the number of slices encoded simultaneously. In this work, a new multiband encoding technique called partial discrete Fourier transform (PDFT) encoding is introduced, which overcomes this restriction. This technique incorporates the principle of partial Fourier imaging, allowing the tradeoff of SNR and imaging time without changing the number of slices. The theory behind PDFT encoding and its inherent sensitivity to phase errors are outlined. The theory was validated through simulations, showing that phase errors result in degraded slice localization. The feasibility of PDFT encoding of 12 slices was tested with experimental excitation profile measurements and heart images of a human subject using commercial MRI equipment. Imaging time was reduced to 66% with SNR reduced to 82%. Magn Reson Med 45:118-127, 2001.

Coronary venous oximetry using MRI.

Based on the Fick law, coronary venous blood oxygen measurements have value for assessing functional parameters such as the coronary flow reserve. At present, the application of this measure is restricted by its invasive nature. This report describes the design and testing of a noninvasive coronary venous blood oxygen measurement using MRI, with a preliminary focus on the coronary sinus. After design optimization including a four-coil phased array and an optimal set of data acquisition parameters, quality tests indicate measurement precision on the order of the gold standard optical measurement (3%O(2)). Comparative studies using catheter sampling suggest reasonable accuracy (3 subjects), with variability dominated by sampling location uncertainty ( approximately 7%O(2)). Intravenous dipyridamole (5 subjects) induces significant changes in sinus blood oxygenation (22 +/- 9% O(2)), corresponding to flow reserves of 1.8 +/- 0.4, suggesting the potential for clinical utility. Underestimation of flow reserve is dominated by right atrial mixing and the systemic effects of dipyridamole. Magn Reson Med 42:837-848, 1999.

Tracking oxygen effects on MR signal in blood and skeletal muscle during hyperoxia exposure.

Blood and muscle T1 and T2 relaxivity was examined under normoxic (air; 20.8% O2) and hyperoxic (100% O2) conditions to determine whether the oxygenation state of blood in the large vessels and in the microcirculation can be monitored in vivo. The femoral artery/vein and the soleus and gastrocnemius muscles were examined in healthy human male volunteers. Arterial blood T1 decreased with hyperoxia, while venous blood T2 increased, due to increased dissolved O2 and decreased deoxyhemoglobin, respectively. A biexponential T2 model of muscle is proposed, where the short T2 component reflects primarily the intracellular and interstitial compartments (in fast exchange), and the long T2 reflects blood. In this model, the long T2 component increased with hyperoxia exposure. This was more evident in slow twitch (soleus) than in fast twitch (gastrocnemius) muscle. It is concluded that changes in the long T2 component reflect change in the microcirculation oxygenation state.

Partial volume effects on vascular T2 measurements.

Quantitative, vascular T2 measurements are of interest for applications such as MR oximetry. In the situation of a vessel with long T2 relaxation times embedded in tissue with relatively short T2 values, contamination of the blood signal from the surrounding tissue can bias T2 measurements. Limited data sampling and vessels running obliquely through the imaging slice can cause significant signal contamination. Using a model of these effects to predict the behavior of T2 measurements under a range of conditions, a set of parameters that provides the best combination of measurement accuracy with a signal-to-noise ratio as high as possible is proposed.

T2 accuracy on a whole-body imager.

MR oximetry requires a T2 measurement that is accurate within 5% in vivo. Simple methods are susceptible to signal loss and tend to underestimate T2. Current methods utilize RF pulses or RF cycling patterns that prevent signal loss at each data acquisition. However, using these methods with imperfect pulses, T2 tends to be overestimated due to temporary storage of the magnetization along the longitudinal axis where it decays more slowly with a time constant T1 > T2. To reduce the T1 dependence while preventing signal loss, we utilize simple 90x180y90x composite pulses and good RF cycling patterns. These trains are critical for T2 accuracy over typical ranges of RF and static field inhomogeneities and refocusing intervals. T1 signal decay during each 90x180y90x pulse must be accounted for to yield accuracy within 5% when the pulse-width is 10% or more of the refocusing interval. A simple correction scheme compensates for this T1-related error effectively.