Comparison of Prospective Head Motion Correction with NMR Field Probes and an Optical Tracking System

Subject motion is a major problem in functional and anatomical head MRI. The resulting artifacts such as ghosting and blurring may complicate image interpretation, or in the worst case, render acquired images useless. Thus, measurements have to be repeated or entire patient populations, such as elderly or pediatric patients, have to be excluded from certain studies. Furthermore, the high spatial resolution achievable with ultra-high field MRI scanners might be limited by the range of involuntary motion even in trained, cooperative subjects.
Many different methods have been proposed to overcome this obstacle. Retrospective motion correction methods use the acquired data and try to eliminate motion artifacts with approaches such as image co-registration or entropy-based methods. While there are manifold powerful retrospective motion correction methods available, inconsistently sampled k-space data and spin-history effects can limit the ability of those methods to eliminate motion artifacts in the images. In addition, especially the coregistration-based methods require multiple shots of the same volume to work which is usually only applicable for functional imaging. Prospective motion correction (PMC), on the other hand, uses a motion tracking modality of choice to continuously update the orientation and position of the FOV during data acquisition. Those tracking modalities include MR-based methods such as navigators or external tracking devices such as NMR marker and optical tracking devices.
Optical tracking with camera(s) have become an established tool in prospective head motion correction. Those devices demonstrated a high accuracy, fast sampling rate and are almost independent from the imaging sequence. However, they require direct line of sight to an optical target or the subject's face and a cross-calibration to convert camera coordinates to scanner coordinates. In contrast, NMR markers such as NMR field probes (FPs) are an alternative that does not require a line of sight and naturally operates in the scanner coordinate system. However, the tracking of NMR markers usually requires additional tracking gradients in all three gradient axes, which can be implemented as a separate gradient block, as a modification of imaging gradients or as recently demonstrated, even with the unmodified native imaging gradients. NMR FPs have already successfully been used for prospective head motion correction, but to the authors' knowledge, there has not been any comparison of FP-based PMC against other methods available. This work compares the prospective head motion correction capabilities of an optical Moiré phase tracking (MPT) camera-marker system and ¹⁹F NMR FPs in healthy volunteers.