PURPOSE

To apply the navigator-based slice tracking method to prospectively compensate the respiratory motion for kidney pseudo-continuous arterial spin labeling (pCASL) using EPI acquisition.

METHODS

Measurements were performed using an 18-channel body and spine receive RF coil on a 1.5T Aera Siemens scanner (Siemens Healthineers AG, Erlangen, Germany). A single gradient-echo slice selection and projection readout at the location of the diaphragm along the inferior-superior (IS) direction is acquired as a navigator. Navigator acquisition and fat suppression were interleave inserted before each transverse imaging slice of the readouts of 2D gradient-echo EPI based pCASL sequence (Fig. 1), with 45 measurements including 4000 training navigators at the beginning of measurements obtained in 5 minutes. Sequence parameters were as follows: TE =12 ms, FOV=400×240 mm³, in-plane iPAT factor=3, matrix size=120×72×8, resolution=3.3×3.3×8 mm³, slice gap = 4 mm, labeling duration = 1650 ms, postlabeling delay (PLD) =2000 ms, TR = 6500 ms, TI for SPAIR (SPectral Attenuated Inversion Recovery) fat suppression=90 ms, FOV_nav = 200 mm, res_nav = 64, Flip angle_nav = 15^º, TR_nav=4.22ms. The labeling pulse train and readout were triggered by the pulse signal and played out only during the systolic period. The renal blood flow was selectively tagged using a tagging plane placed above to the diaphragm, labeling the arterial blood in the descending aorta.
Before motion analysis, the interleaved navigator signals during image acquisition were Fourier transformed and truncated to exclude RF saturation along IS direction from the EPI readout (Fig. 2). The position for this truncation was calculated based on peaks fitted from the averaged training navigator and the peaks from the first 32 interleaved navigators. The diaphragm position was derived by calculating the phase difference of the interleaved navigator signals at each acquisition after Fourier transform and truncation. The unwrapped data from different coils were then combined by using coil clustering (2) based on the first 32 interleaved navigators. The motion information was then directly sent back to the sequence and slice positioning was adjusted in real-time. This motion analysis and real-time feedback was performed on the scanner, implemented in ICE (Image calculation environment, Siemens Healthineers AG, Erlangen, Germany).

RESULTS

The respiratory motion from navigator signals could be precisely calculated (Fig. 2) and slice positioning was changed in real-time based on the motion information. Comparing to the case without, motion compensation reduced liver movement in a selected slice (Fig. 2). Fig. 3 shows spatial maps of the coefficient of variation (CoV) of control images from a represent volunteer in kidney. The value of CoV was calculated as the standard deviation of the constituent voxels divided by their mean. Motion greatly increased the CoV, but to a lesser extent when the motion correction was on. The calculated perfusion maps show significantly higher perfusion in the renal cortex from navigator-based motion compensated pCASL than from no motion compensated pCASL (Fig. 4, 5). More detailed structure in the perfusion maps could be observed in the motion compensated case (Fig. 4, 5).

CONCLUSION

This study demonstrates the feasibility of navigator-based slice tracking technique in kidney pCASL using EPI readout. The sequence may improve the evaluation of the perfusion in kidney diseases.

Acknowledgements

No acknowledgement found.