Julian R. Maclaren, Ph.D.

Vera K. Kimbrell, B.S., R.T.(R)(MR)

Maryna Babayeva^1,2, Pavel Falkovskiy^1,2, Tom Hilbert^1,2, Guillaume Bonnier^1,2, Bénédicte Maréchal^1,2, Reto Meuli³, Jean-Philippe Thiran², Rolf Gruetter^3,4, Gunnar Krueger^1,2, and Tobias Kober^1,2
1Siemens ACIT - CHUV Radiology, Siemens Healthcare IM BM PI, & Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, ²LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ³Department of Radiology, University Hospital (CHUV), Lausanne, Switzerland, ⁴CIBM, École Polytechnique Fédérale de Lausanne and University of Geneva, Switzerland

In this work, we propose a method for prospective motion correction in MRI using a novel image navigator module, which is triggered by a free induction decay (FID) navigator. Only when motion occurs, the image navigator is run and new positional information is obtained through image registration. The image navigator was specifically designed to match the impact on the magnetization and the acoustic noise of the host sequence. This detection-correction scheme was implemented for an MP-RAGE sequence and 5 healthy volunteers were scanned at 3T while performing various head movements. The correction performance was demonstrated through automated brain segmentation and an image quality index whose results are sensitive to motion artifacts.

Enrico Avventi¹, Mathias Engström^1,2, Ola Norbeck¹, Magnus Mårtensson^2,3, and Stefan Skare^1,2
1Dept. of Neuroradiology, Karolinska University Hospital, Stockholm, Sweden, ²Dept. of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden,³EMEA Research & Collaboration, GE Science Laboratory, GE Healthcare, Stockholm, Sweden

In previous works we developed a novel and promising navigator technique aimed for prospective motion correction: cFatNav (collapsed FatNav). A cFatNav sub-sequence consists of three EPI readouts sampling orthogonal planes in k-space placed between a non space-selective, fat saturation pulse and the host sequence excitation. From each sampled k-space plane, via IFFT, we can obtain a view of the excited volume projected along three orthogonal direction. We have shown that 2D registration applied to cFatNav data produces precise motion estimates when the motion occur mostly along one of the three sampled planes. In this work we present a 2D/3D registration algorithm for cFatNav data that can handle out-of-plane motion. Specifically each of the three collapsed views are matched against a reference 3D volume simultaneously by Gauss-Newton method.

Gregory R. Lee^1,2, Yong Chen³, and Vikas Gulani^3,4
1Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States, ²University of Cincinnati, Cincinnati, OH, United States, ³Radiology, University Hospitals Case Medical Center, Cleveland, OH, United States, ⁴Radiolgoy, Case Western Reserve University, Cleveland, OH, United States

In this work an approach to respiratory self-navigation is proposed that is applicable to non-Cartesian dynamic contrast enhanced MRI sequences that repeatedly sample the k-space center. The approach starts by bandpass filtering to removes slow changes in the signal magnitude & phase due to contrast arrival as well as high frequency noise. Correlations among the complex central k-space point are then evaluated across coils to extract a respiratory waveform. This waveform enables retrospective respiratory binning for motion-compensated image reconstruction. Computation time is negligible and the resulting navigator compares favorably to those acquired with a respiratory bellows or image-domain navigator.

Jieying Luo¹, Nii Okai Addy¹, R. Reeve Ingle¹, Joseph Y. Cheng¹, Bob S. Hu², and Dwight G. Nishimura^1
1Electrical Engineering, Stanford University, Stanford, California, United States, ²Palo Alto Medical Foundation, Palo Alto, California, United States

An autofocusing nonrigid respiratory motion correction method for abdominal imaging was developed. This technique used motion trajectory candidates estimated from 3D image-based navigators that were acquired with a variable-density 3D cones trajectory. The performance of the proposed autofocusing motion correction was demonstrated with free-breathing scans of the abdominal vasculature.

Rasmus Ramsbøl Jensen^1,2, Claus Benjaminsen^1,2, Adam Espe Hansen², Rasmus Larsen¹, and Oline Vinter Olesen^1,2
1DTU Compute, Technical University of Denmark, Lyngby, Copenhagen, Denmark, ²Department of Clinical Physiology, Nuclear Medicine & PET, Rigshospitalet, Copenhagen, Denmark

We propose a method for accurate motion correction in brain MRI through markerless tracking. It is the first time markerless tracking is used to correct MRI images. The method has the potential to seamlessly fit the clinical workflow with no added complexity and no errors related to the generally attached markers. It is based on our Tracoline system; a unique remote surface scanner conveying images through optical fibers. The developed software continuously aligns the surface scans, which allows for accurate motion tracking. In this work, we demonstrate our system on the Siemens mMR with motion correction of an EPI sequence.

Guido P. Kudielka^1,2, Anne Menini¹, Pierre-André Vuissoz^2,3, Jacques Felblinger^3,4, and Florian Wiesinger^1
1GE Global Research, Munich, BY, Germany, ²Imagerie Adaptative Diagnostique et Interventionnelle, Université de Lorraine, Nancy, Lorraine, France,³U947, INSERM, Nancy, Lorraine, France, ⁴CIC-IT 1433, INSERM, Nancy, Lorraine, France

Acquistion of 3D information with dedicated cameras is becoming more popular in healthcare applications for motion and position registration as well as for diagnostic purposes. We present the technical feasibility of such a camera system in a MRI environment. Respiratory movement was acquired by the camera during free-breathing abdominal MRI acquistions. The acquired images were corrected retrospectively with the respiratory belt information, the depth information and intensity amplitude of the camera. The comparison with pneumatic belt data shows strong correlation and motion artifacts were greatly reduced with the camera data.