UCSF - "Compressed Sensing in Medical Physics," Lei Zhu, Georgia Institute of Technology
Compressed sensing (CS) is a theory of sparse-signal recovery from very few measurements, which recently gains heated attention in different fields of signal processing. This talk discusses the applications of CS in medical physics. New algorithms are proposed for fast CT reconstruction from 10%-dose projections. With prior knowledge assistance, the dose can be even lower (around 1%) without losing structural information. In the inverse planning of radiation therapy, CS is able to significantly reduce the total treatment time by eliminating dispensable radiation segments, showing great advantages in both conventional intensity modulated radiation and rotational arc therapy.
Wed Mar 21, 2012 5:30pm – 6:30pm GMT (no daylight saving)