FULL PAPER The Effects of SENSE on PROPELLER Imaging Yuchou Chang, 1 * James G. Pipe, 1 John P. Karis, 1 Wende N. Gibbs, 1 Nicholas R. Zwart, 1 and Michael Sch€ ar 1,2,3 Purpose: To study how sensitivity encoding (SENSE) impacts periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) image quality, including signal- to-noise ratio (SNR), robustness to motion, precision of motion estimation, and image quality. Methods: Five volunteers were imaged by three sets of scans. A rapid method for generating the g-factor map was proposed and validated via Monte Carlo simulations. Sensitivity maps were extrapolated to increase the area over which SENSE can be performed and therefore enhance the robustness to head motion. The precision of motion estimation of PROPELLER blades that are unfolded with these sensitivity maps was investigated. An interleaved R-factor PROPELLER sequence was used to acquire data with similar amounts of motion with and without SENSE acceleration. Two neuroradiologists inde- pendently and blindly compared 214 image pairs. Results: The proposed method of g-factor calculation was similar to that provided by the Monte Carlo methods. Extrapo- lation and rotation of the sensitivity maps allowed for contin- ued robustness of SENSE unfolding in the presence of motion. SENSE-widened blades improved the precision of rotation and translation estimation. PROPELLER images with a SENSE fac- tor of 3 outperformed the traditional PROPELLER images when reconstructing the same number of blades. Conclusion: SENSE not only accelerates PROPELLER but can also improve robustness and precision of head motion correction, which improves overall image quality even when SNR is lost due to acceleration. The reduction of SNR, as a penalty of acceleration, is characterized by the proposed g-factor method. Magn Reson Med 74:1598–1608, 2015. V C 2014 Wiley Periodicals, Inc. Key words: PROPELLER; SENSE; parallel imaging; motion correction; g-factor; sensitivity map extrapolation and rotation INTRODUCTION Periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) (1,2) is inher- ently motion-insensitive and corrects bulk in-plane motion, at the expense of a 60% increase in minimum scan time. Several parallel imaging methods have been developed to accelerate PROPELLER to shorten scan time and additionally widen each blade, increasing the area in k-space used for self-navigation, which can improve motion compensation. These methods generally reconstruct the undersampled blades by data-driven par- allel imaging methods (3–6) or iterative reconstruction methods (7,8). For calculation of the generalized autoca- librating partially parallel acquisitions (GRAPPA) (9) weights, autocalibration signal lines are acquired in each blade acquisition for image reconstruction (6), requiring more data in the center of each blade; this in turn reduces the effective acceleration factor. Alternatively, the GRAPPA kernel is trained by source and target lines acquired from orthogonal blades, and the missing data are reconstructed on the undersampled blades (3,4). In these references, the orthogonal blades were acquired in the same echo train, reducing the blade width by a factor of 2. Holmes et al. (5) used a variable density sampling pattern in each blade in combination with an extra exter- nal calibration blade to increase the effective accelera- tion, but this may introduce inconsistent training data if motion occurred between acquisitions of calibration blade and imaging blades. Other methods (7,8) iteratively reconstruct blades without requiring orthogonal blades but are computationally expensive. Another popular parallel imaging technique, sensitiv- ity encoding (SENSE) (10), has been less studied with PROPELLER MRI. SENSE has been used to accelerate data acquisition by reconstructing undersampled PRO- PELLER EPI blades (11). Either an estimate of the sensi- tivity map for each blade or special coil geometry, such that the sensitivity maps are circularly symmetric, was used to reconstruct the blades. Unlike data-driven auto- calibration methods, SENSE reconstruction does not uti- lize autocalibration data, but requires external coil calibration data based on an additionally acquired sensi- tivity map. The external calibration in SENSE provides a large effective acceleration for each blade without need- ing autocalibration signal lines, orthogonal blades, or variable density sampling. Skare et al. (12) demonstrated that a mismatch between external calibration of coil sen- sitivity maps and imaging data may occur due to motion, and imaging data may fall outside of the supported region defined by the sensitivity map of Cartesian SENSE. Potential patient motion between acquisition of the sensitivity map and the acquired blades needs to be accounted for when using SENSE reconstruction of PRO- PELLER MRI. Blades widened by SENSE may be able to improve the precision of the motion estimation. Wide blades have been quantitatively proven to reduce errors in rotation estimates, but translation estimation has been shown to 1 Neuroimaging Research, Barrow Neurological Institute, Phoenix, Arizona, USA. 2 Philips Healthcare, Cleveland, Ohio, USA. 3 Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, Maryland, USA. Grant sponsor: Philips Healthcare. *Correspondence to: Yuchou Chang, Ph.D., Neuroimaging Research, Bar- row Neurological Institute, 350 W. Thomas Road, Phoenix, AZ 85013. E-mail: yuchou.chang@gmail.com Twitter: @yuchou_chang Received 13 August 2014; revised 27 October 2014; accepted 8 November 2014 DOI 10.1002/mrm.25557 Published online 17 December 2014 in Wiley Online Library (wileyonlineli- brary.com). Magnetic Resonance in Medicine 74:1598–1608 (2015) V C 2014 Wiley Periodicals, Inc. 1598