Magnetic Resonance Image Reconstruction

For our December meetup, we are pleased to have a very interesting discussion about using open source tools and distributed computing for MRI image reconstruction. Please join us on Tuesday 12/16 at 6:30 (location to be confirmed soon),

6:30 - 7:00 Food & Networking
7:00 - 7:10 Intros and announcements
7:15 - 8:15ish Presentation / Q&A
Adjourn to Data Drinks

Title: Magnetic Resonance Image Reconstruction 1. 5 minute introduction to MRI and MRI reconstruction: where does the signal come from, how does it get turned into an image (that's the reconstruction part), what are the typical constraints in terms of imaging speed, etc.
http://onlinelibrary.wiley.com/doi/10.1002/jmri.24687/abstract

2. Image reconstruction in units of signal to noise (SNR): This would be me taking a more statistical view of how we reconstruct images such that at the end we know what the signal to noise is in each pixel.

3. The extension of the SNR reconstruction to region of interest analysis in MRI images: pixels in the image are not entirely independent, what happens to ROI analysis? Extension to flow measuremens, how do we measure flow with MRI and how do we assign a confidence interval on such flow measurements.
   http://onlinelibrary.wiley.com/doi/10.1002/mrm.25194/abstract
   http://www.jcmr-online.com/content/16/1/46

4. Open source software tools that we develop/use. I will focus on our open raw data standard (ismrmrd.github.io) and a high performance reconstruction pipeline that I have designed (gadgetron.github.io)
   http://onlinelibrary.wiley.com/doi/10.1002/mrm.24389/abstract

5. Using cloud computing in MRI reconstruction. Interfacing clinical MRI systems directly to Gadgetron instances in the cloud.
   http://www.ncbi.nlm.nih.gov/pubmed/24687458

Michael Hansen is a biomedical engineer with a PhD from University of Aarhus, Denmark. After completing his PhD, he was a research fellow at University College, London and Great Ormond Street Hospital for Children, London. He also worked at Novartis Institutes for BioMedical Research in Cambridge, MA before moving to the National Heart, Lung, and Blood Institute, NIH, where he heads the Magnetic Resonance Technology Program. His program focuses on MRI techniques for cardiac imaging, pediatric imaging, and real-time guidance of interventional procedures. His particular areas of interest are fast pulse sequences, non-Cartesian imaging, real-time image reconstruction, GPU based reconstruction, Distributed Computing, and motion correction. He has extensive experience in the area of high throughput real-time processing and reconstruction of imaging data.