Jordan Muraskin¹, Paul Sajda¹, Robin I Goldman¹, William J Thomas¹, Melvyn B Ooi², and Truman R Brown^3
1Biomedical Engineering, Columbia University, New York, New York, United States, ²Stanford University, Stanford, California, United States, ³Medical University of South Carolina, Charleston, South Carolina, United States

 

Maximilian Haeberlin¹, Lars Kasper¹, David Otto Brunner¹, Christoph Barmet¹, and Klaas Paul Pruessmann^1
1Electrical Engineering, Institute for Biomedical Engineering, Zurich, Zurich, Switzerland

We propose to combine NMR field probes and gradient reference tones as a means to autocalibrate real-time slice tracking for head MRI. Being fully autocalibrated, the method needs no extra sequence parts for position determination and is very sensitive for typical readouts.

Michael Herbst¹, Julian Maclaren¹, Matthias Weigel¹, and Maxim Zaitsev^1
1Department of Radiology, University Medical Center Freiburg, Freiburg, Germany

Turbo spin echo (TSE) sequences are useful for rapid, high-resolution structural imaging. However, even small movements during acquisition can cause severe artifacts, due to the motion sensitivity of this technique. This work shows that prospective motion correction can improve results of high resolution TSE imaging even in cooperative volunteers by correcting for position changes between echo trains. When longer readouts are used (e.g. SPACE), a correction once per excitation is not sufficient and continuous correction during the echo train shows further improvement, while preserving the desired sequence timing.

Peter Schulze¹, Daniel Stucht¹, K. A. Danishad¹, Ilja Y. Kadashevich¹, Michael Herbst², Cris Lovell-Smith², Julian Maclaren², Robb T. Barrows³, Todd P. Kusik³, Brian S. Armstrong³, Tom Prieto⁴, Thomas Ernst⁵, Maxim Zaitsev², and Oliver Speck^1
1Department of Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany, ²Department of Radiology, University Medical Center, Freiburg, Germany, ³Department of Electrical Engineering and Computer Science, University of Wisconsin-Milwaukee, Milwaukee, United States,⁴Department of Neurology, Medical College of Wisconsin, Milwaukee, United States, ⁵Department of Medicine, University of Hawaii, Honolulu, United States

Ultra high resolution MRI suffers from long acquisition times, which increase the risk of involuntary subject motion. In this study we present ultra high resolution (0.25 mm in-plane) in vivo MR brain images acquired at 7T using real-time prospective motion correction. A comparison between conventional and prospectively-corrected MRI shows a clear improvement in image quality. To our knowledge it is the first demonstration of such improvements at ultra high spatial resolutions in co-operative and trained subjects. The technique is well explored and ready to be established as a standard feature of new MRI systems.

R. Grimm¹, K. T. Block², J. Hutter¹, C. Forman¹, C. Hintze³, B. Kiefer⁴, and J. Hornegger^1
1Pattern Recognition Lab, University of Erlangen-Nuremberg, Erlangen, Germany, ²Department of Radiology, NYU Langone Medical Center, New York, NY, United States, ³German Cancer Research Center, Heidelberg, Germany, ⁴MR Application & Workflow Development, Siemens AG, Healthcare Sector, Erlangen, Germany

Radial sampling of k-space allows the computation of an intrinsic self-gating signal for free-breathing MRI examinations of the thorax. However, it is common practice to capture respiratory motion and contrast-enhanced perfusion in separate acquisitions. Here, we show how different aspects of dynamic imaging can be extracted retrospectively from a single acquisition, based on a 3D stack-of-stars GRE sequence. A combination of sliding-window reconstructions and self-gating yields three valuable image sequences: (1) Images of motion at high spatial resolution, (2) dynamic perfusion with minimal motion artifacts, and (3) perfusion at high temporal resolution.

Christian Buerger¹, Claudia Prieto^1,2, Andrew Peter King¹, and Tobias Schaeffter^1
1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, Westminster, United Kingdom, ²Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Santiago, Chile

We propose a method to reconstruct motion compensated 3D MRI of the abdomen acquired during free-breathing with nearly 100% scan efficiency. The approach is based on a self-gated Golden-Radial Phase Encoding sampling that allows reconstruction of multiple undersampled 3D images at different respiratory positions. Non-rigid image registrations are employed to combine all images into one high quality motion compensated dataset. The highly efficient technique allows reconstruction of 3D liver MRI with a high isotropic resolution of 1.75mm. Compared to a commonly employed gating technique overall scan time is reduced by 56% while similar image quality is preserved.

Nadine Gdaniec¹, Holger Eggers², Peter Boernert², Mariya Doneva², and Alfred Mertins^1
1University of Luebeck, Luebeck, Germany, ²Philips Research, Hamburg, Germany

A sampling strategy is proposed that minimizes artifacts due to respiratory motion in abdominal imaging acquired during a breathhold. The predefined breathhold length is not required anymore because the sampling strategy supports the reconstruction of data acquired up to any point in time. The acquisition is segmented with global sampling density variation and local Poisson disk sampling and thus compatible with CS and PI. It is adaptable to target functions of the spatial resolution over time. Abdominal imaging of volunteers was performed that show significantly reduced motion artifacts compared to incomplete breathholds.

Olaf Kosch¹, Florian Thiel¹, Steffen Schneider¹, Bernd Ittermann¹, and Frank Seifert^1
1Physikalisch-Technische Bundesanstalt (PTB), Berlin, Germany

The motion detection by contactless multi-channel UWB radar for cardiac MRI navigator is verified by a simultaneously measured one dimensional ‘pencil-like’ MRI for cardiac and respiration motion. A blind source separation was applied in UWB and MR data sets to separate the cardiac and respiration motion components.

Christoph Kolbitsch¹, Jouke Smink², Tobias Schaeffter¹, and Claudia Prieto^1,3
1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom, ²Philips Healthcare, Best, Netherlands,³Escuela de Ingeniería, Pontificia Universidad Católica de Chile, Santiago, Chile

Retrospectively gated cardiac cine imaging is an important clinical tool to diagnose and analyze heart function. For this, data from several cardiac cycles are combined based on ECG signals to increase temporal and spatial resolution. Here we propose a new cardiac navigator technique which derives gating signals from real-time (Golden-Ratio radial) image data by analyzing contraction and expansion of the left ventricle. A volunteer study has shown that the obtained gating signal has an accuracy of less than 17ms compared to standard ECG. The derived gating signal can be used to combine the real-time data to retrospective gated CINE images.

R. Reeve Ingle¹, Taehoon Shin¹, and Dwight G Nishimura^1
1Electrical Engineering, Stanford University, Stanford, California, United States

The TR-perturbed bSSFP sequence can be used to acquire fat-only images and bSSFP-like images during the same steady state. We propose an application of this sequence for respiratory motion correction for 3D cardiac imaging that utilizes the epicardial fat signal for motion estimation. A series of projection fat navigator images are acquired throughout the scan and used for motion estimation and retrospective correction of the image data. The steady-state contrast of the TR-perturbed bSSFP sequence is used for image and navigator acquisition without the need for interruption and subsequent re-catalyzation of the steady-state magnetization.