MOTIVATION - Motion remains one of the most frequent contributors to image artifacts in MR studies. The motion susceptibility of MRI is well-known and has spawned a number of elegant navigation techniques. These methods, however, are tailored to specific MR acquisitions that require modified k-space trajectories or the acquisition of additional MR data, and most are unable to correct certain types of motion, for example, through-plane motion. Moreover, motion correction has been focused on specific families of sequences, but no generally applicable approach currently exists. Certain patient populations, such as pediatric or geriatric patients, are more likely to move than others. In pediatric imaging, anesthesia is used to control motion, adding substantially to exam costs and patient risks. A sequence-independent, autonomous and prospective motion correction system could greatly improve image quality for a wide spectrum of MR examinations. For pediatric imaging, in particular, we anticipate reduced reliance on anesthesia to control patient motion.
AIMS - We will be focusing on three independent specific aims (carried out in parallel and completed within 4 years), with corresponding subaims that we believe are important for establishing the technical/scientific merit and to demonstrate the feasibility of the proposed R&D efforts for our real-time adaptive motion correction approach. Specifically, these aims are: (1) to develop and evaluate a coil-mounted MR-compatible tracking device for routine clinical use;(2) to integrate pose tracking into real-time MRI;and (3) to validate our real-time motion correction system in volunteers and patients. METHODS -In Aim 1 we will improve the methods for computer-vision-based pose estimation inside an MR scanner and build an MR-compatible coil-mounted pose tracker that can be used in clinical routine examinations.
In Aim 2 we will focus on reducing the latency between pose changes happening and the MR scanner reacting to these pose changes, and on building a software library for the MR pulse sequence development that allows one to implement real-time motion correction into all MR pulse sequences.
In Aim 3 we will perform a thorough evaluation of our system on 60 volunteers (30 adults and 30 children) and 120 patients (80 adults and 40 children). SIGNIFICANCE - The impact of our technology has several facets. First, it will improve patient care by reducing the number of MR images with compromised quality because of motion artifacts. Especially because of the increasing reliance on MR Images as a primary means of diagnosis, this will reduce the number of misdiagnoses. Secondly, this technology will help to lower the high national spending on imaging by dramatically improving the efficiency of MRI scanners. Finally, it will improve patient comfort by reducing the need for repeat sequences, as well as reduce the necessity of sedation aimed at keeping the patient still. Overall, this technology will have a significant impact on MRI both in clinical practice and basic science research.
Synopsis Patient motion during an MRI exam can result in major degradation of image quality and decreased efficiency in a large portion of the 40 million MRI procedures performed annually. This is not only of increasing concern due to the aging population and its associated diseases, but also in children and patients who are simply too sick to remain still during an exam. This proposal aims to build an optical tracking system and methods to adapt MRI pulse sequences to changes in patient pose in real time. The project leverages on previous work from an R21 project in which a prototype system was successfully built. Specifically, the aims are to (i) introduce innovative improvements to the existing prototype, making the system suitable for use in clinical routine, (ii) modify the optical tracking system to be compatible with multiple MR sequences, and (iii) perform clinical evaluation. Our real-time technique is a unique and innovative solution that will improve MRI image quality and thus patient care, and will address the escalating burden of imaging costs on the health care system.
|Aksoy, Murat; Maclaren, Julian; Bammer, Roland (2017) Prospective motion correction for 3D pseudo-continuous arterial spin labeling using an external optical tracking system. Magn Reson Imaging 39:44-52|
|Vos, Sjoerd B; Aksoy, Murat; Han, Zhaoying et al. (2016) Trade-off between angular and spatial resolutions in in vivo fiber tractography. Neuroimage 129:117-132|
|O'Halloran, Rafael; Aksoy, Murat; Aboussouan, Eric et al. (2015) Real-time correction of rigid body motion-induced phase errors for diffusion-weighted steady-state free precession imaging. Magn Reson Med 73:565-76|
|Van, Anh T; Aksoy, Murat; Holdsworth, Samantha J et al. (2015) Slab profile encoding (PEN) for minimizing slab boundary artifact in three-dimensional diffusion-weighted multislab acquisition. Magn Reson Med 73:605-13|
|Zaitsev, Maxim; Maclaren, Julian; Herbst, Michael (2015) Motion artifacts in MRI: A complex problem with many partial solutions. J Magn Reson Imaging 42:887-901|
|Maclaren, Julian; Aksoy, Murat; Bammer, Roland (2015) Contact-free physiological monitoring using a markerless optical system. Magn Reson Med 74:571-7|
|Van, Anh T; Holdsworth, Samantha J; Bammer, Roland (2014) In vivo investigation of restricted diffusion in the human brain with optimized oscillating diffusion gradient encoding. Magn Reson Med 71:83-94|
|Kopeinigg, Daniel; Bammer, Roland (2014) Time-resolved angiography using inflow subtraction (TRAILS). Magn Reson Med 72:669-78|
|Yeom, Kristen W; Holdsworth, Samantha J; Van, Anh T et al. (2013) Comparison of readout-segmented echo-planar imaging (EPI) and single-shot EPI in clinical application of diffusion-weighted imaging of the pediatric brain. AJR Am J Roentgenol 200:W437-43|
|Ooi, Melvyn B; Muraskin, Jordan; Zou, Xiaowei et al. (2013) Combined prospective and retrospective correction to reduce motion-induced image misalignment and geometric distortions in EPI. Magn Reson Med 69:803-11|
Showing the most recent 10 out of 29 publications