hide
Free keywords:
-
Abstract:
Purpose/Introduction: The evaluation of the regional blood flow in pulmonary vessels and in lung parenchyma along the cardiac cycle can be of importance for the clinical diagnosis of lung diseases. Pulsed arterial spin labeling (ASL) techniques have been used for evaluation of the blood bolus delivery curve in the lung by a variation of the delay time between blood preparation and image recording.1 However, acquiring only few number of label/control scans in a single breat hold, the sensitivity of this technique was not high enough to clearly reveal homogenous signal changes of the lung parenchyma. The goal of this work was to assess the potential of an ECG-triggered pseudo-continuous ASL (PCASL) technique with imaging by True- FISP sequence2 to measure the temporal and spatial characteristics of pulmonary blood flow at 1.5T. Subjects and Methods: Ten healthy volunteers were examined on a 1.5T MR scanner using a PCASL sequence3 with a True-FISP imaging module. Eight measurements with different post labeling delays (PLDs) 100-1500 ms were performed. The labeling pulse was placed nearly perpendicular to the pulmonary trunk (Fig. 1a) and applied during the systolic period by ECG-triggering. True-FISP sequence parameters for coronal imaging (Fig. 1b) were adapted to achieve short TE/TR. PCASL images were acquired by employing a timed breathing protocol. The overall scan time for eight measurements was 17:25 min. All images were registered before further evaluation in order to reduce motion effects using open-source toolbox elastix4 in a self-written MATLAB script. To provide an accurate perfusion signal evaluation of the lung parenchyma, the lung was segmented using a Gaussian mixture model clustering three components (large vessels, small vessels, and parenchyma)5 based on data with shortest PLD. Results: Temporal perfusion-related signal development in large and small pulmonary arteries and in lung parenchyma (was computed using corresponding tissue masks (Fig. 2b, c). In perfusion images of a healthy volunteer (Fig. 2a), large and small pulmonary arteries are visible at short PLDs, whereas the signal from lung parenchyma gets mainly increased for longer PLD[900 ms. Perfusion images of lung parenchyma of three healthy volunteers at different PLDs are shown in Fig. 3A. Temporal development of perfusion-related signal in lung parenchyma (Fig. 3b) is clearly delayed compared to the earlier signal changes in larger vessels in all cases. Discussion/Conclusion: We could demonstrate that ECG-triggered PCASL-True-FISP imaging at 1.5T has the potential to assess the course of labeled blood through pulmonary arteries and lung parenchyma. The sequence provides perfusion images of the lung with high spatial resolution and with high signal intensity without application of contrast media.
