EE369C: Medical Image Reconstruction

Course Description

Image reconstruction methods are central to many of the new applications of medical imaging. This course will provide an introduction these techniques in a consistant framework by developing a sequence of software tools for the reconstruction of medical imaging data. This course will concentrate on magnetic resonance imaging (MRI), but will also examine x-ray computed tomography (CT) and discuss positron emission tomography (PET).

Some of the reconstruction problems that will be studied include reconstuction from non-uniform frequency domain data, automatic focusing, phase unwrapping, reconstruction from incomplete frequency domain data, backprojection, and reconstruction of time series of images. Examples will be drawn from fast MRI methods such as spiral, echo-planar, multi-coil/parallel and partial k-space acquisitions.

Instructor

John Pauly
Information Systems Laboratory
Packard Electrical Engineering 258
(650) 723-4569
(650) 723-8473 (FAX)
pauly@stanford.edu

Class Location:

Building 380, room 380W

Office Hours

Wednesday 1-2, Thursday 11-12.

Required Text

Handbook of MRI Pulse Sequences
Bernstein, King, and Zhou
Elsevier/Wiley, 2004

This should be in the bookstore. You can also get it from Amazon here.

Optional Text

Principles of Computerized Tomographic Imaging
Avinash C. Kak and Malcolm Slaney
SIAM

This should also be in the bookstore, and is available from Amazon at here . It is also available online here .

Other Useful Texts

Computed Tomography, Principles, Design, Artifacts, and Recent Advances
Jian Hsieh
SPIE Press, 2003
Amazon link

Positron Emission Tomography
Valk, Bailey, Townsend, and Maisey
Springer-Verlag, 2003
Amazon link for the first volume of the new 2005 edition

Grading

  • Weekly assignments consisting of problem sets and matlab programming.(50% of the grade)
  • A final project. This will be a one page abstract and a 10-15 minute oral presentation, or a 10-15 page report. (50% of the grade)

    Announcements

    Class Handouts (pdf)

    Sept 25: Class Overview

    Sept 27: Review of MR, and sources of error in MR data.
    Read Partial k-Space Reconstruction Notes , and Bernstein section 13.4, pages 546-558.
    Handwritten notes for Sept 25 class.

    Oct. 2: Direct algorithms for partial k-space reconstruction.
    Handwritten notes for Sept 27 class.

    Oct. 4: Iterative algorithms for partial k-space reconstruction.
    Assignment 1 is due Oct 11. Data files are here.

    Oct. 9: Non-Cartesian reconstruction.
    Read the notes here .
    This year we won't be covering field maps, or Dixon reconstruction, but the notes are here and here for reference.

    Oct. 11: Non-Cartesian reconstruction, continued.
    Assignment 2 is due Oct 18. Data files are here.

    Oct 16: Gridding Kernels and Density Compensation
    Read Rapid Gridding Reconstruction With a Minimal Oversampling Ratio by Beatty, et al.

    Oct 18: Choosing a gridding kernel.
    Assignment 3 due Oct 25.

    Oct 23: Off-Resonance Correction.
    Handwritten notes from class.
    Read Section 17.6.3 in the Bernstein book for a survey of a variety of off-resonsance correction methods.

    Oct 25: Multicoil Reconstruction
    Assignment 4 due Nov 1.
    Read Section 13.1.5 (Multicoil Reconstruction) and 13.3 (Parallel Image Reconstruction) from the Bernstein book.
    Optional: "SENSE: Sensitivity Encoding for Fast MRI", K.P. Pruessmann, M. Weiger, M. Scheidegger, and P. Boesinger, Magn. Reson. Med. Vol 42, No 5, pp 952-962, 1999.

    Oct 30: Parallel Image Reconstruction

    Nov 1: Parallel Image Reconstruction, continued
    Assignment 5 due Nov 8.

    Nov 6: Parallel Image Reconstruction, k-Space Algorithms
    Project topics due Nov 15.

    Nov 8: Compressed Sensing, Miki Lustig
    Sparse MRI: The Application of Compressed Sensing for Rapid MR Imaging
    Assignment 6 due Nov 20. The data file is here, and the m-files are here.

    Nov 13: CT, Backprojection
    Assignment 2 Solutions
    Assignment 3 Solutions
    Assignment 4 Solutions

    Nov 15: Fan Beam CT Reconstruction
    Read Sections 3.1-3.4 in Kak and Slaney
    Initial project ideas are due today

    Nov 20: No Class
    Assignment 7 is optional. The data files are here, and is limited to Stanford students.

    Nov. 27: Positron Emission Tomography
    Reading Assignment: "Positron Emission Tomography," J.M. Ollinger and J.A. Fessler, IEEE Signal Processing Magazine, Vol 14, No 1, pp 43-55, 1997.

    Nov. 29: 3D PET, 3D Filtered Backprojection
    Assignment 8 is to write a brief one page abstract for your presentation or report.

    Dec. 4: Iterative PET Reconstruction

    Dec. 6: Presentations
    Assignment 1 Solutions
    Assignment 5 Solutions

    Last updated Dec 6, 2007