Sparse Reconstruction Techniques
Anthony G Christodoulou1
1Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States

Synopsis

MRI is a powerful but slow imaging modality, presenting challenges for scanner throughput, motion corruption, and observation of fast dynamic processes. Modern sparse reconstruction techniques break the classical speed limits of MRI, opening new opportunities and solving several long-standing problems. These approaches exploit the redundancy within images and across image sequences, representing images more efficiently than classical approaches to allow efficient acquisition. This talk will provide an overview of various sparse reconstruction techniques for static and dynamic imaging, with particular focus on compressed sensing and low-rank approaches.

Overview

MRI is an incredibly powerful imaging modality; however, it is notoriously slow, presenting challenges for scanner throughput, motion corruption, and observation of fast dynamic processes in the body. The 21st century has seen the advent of sparse reconstruction techniques that break the classical speed limits of MRI, opening new opportunities and solving several long-standing problems in the field. Sparse reconstruction approaches exploit the redundancy within images and across image sequences, representing images more efficiently than classical approaches and in turn allowing more efficient acquisition. This talk will provide an overview of various sparse reconstruction techniques for static and dynamic imaging, with particular focus on compressed sensing and low-rank approaches.

Target Audience

The target audience is any attendee interested in learning how high-quality images can be produced from seemingly "incomplete" measurements. Although sparse reconstruction techniques rely on several advanced mathematical concepts, these concepts will be discussed in terms suitable for a broad audience; no specialized mathematical background is required.

Objectives

As a result of attending this talk, participants should be able to:
  • Understand the basics of efficient, sparse image representation
  • Understand the difference between reconstruction approaches
  • Understand the pros and cons of sparse reconstruction
  • Understand the current and potential applications of sparse reconstruction techniques

Acknowledgements

No acknowledgement found.

References

No reference found.
Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)