OPTIMAL SPATIO-TEMPORAL SAMPLING FOR REAL-TIME MAGNETIC RESONANCE IMAGING
Professor Bresler is researching the mathematical theory and algorithms of computed imaging systems, with special emphasis on biomedical imaging. Such systems range from radio telescopes and imaging radars, through computed tomography (CT) and magnetic resonance imaging (MRI) scanners, to electron microscopes. He recently developed a complete theory of time-sequential sampling for the acquisition of time-varying images subject to the constraint that only one sample can be acquired at any time instant. This theory, which overcomes a more-than-decade-long impasse, provides the tools to design effective systems to image dynamic physical and biological phenomena, such as drifting wafts of atmospheric pollutants or the beating heart.
During his Center appointment, Professor Bresler will study the problem of dynamic MRI imaging for cardiac, functional, or interventional MRI. Despite considerable technological progress and intense research interest, a general theory for dynamic MRI is still lacking. Existing techniques are still largely governed by the conventional static-imaging theory, and thus restrict the acquisition speed. His goal in this project is to develop a new paradigm that provides a theoretical and practical framework for optimal dynamic MRI, allowing speedup by an order of magnitude or better. If successful, these developments will be a breakthrough in dynamic MRI, providing for the first time the capability for high-resolution 3D real-time imaging of the beating heart.