Recently it has been shown that fluoroscopy sequences of noise image frames strongly interact with human-observer spatio-temporal processing to determine equivalent perception dose savings from low acquisition rate pulsed fluoroscopy and from simple digital temporal filters. Using image perception studies and observer modeling, the applicants proposed to continue to optimize acquisition and digital processing of x-ray fluoroscopy images. Effects of digital spatial and temporal processing on observer detection of moving objects will be measured. A new reference/test adaptive forced choice method will greatly improve efficiency and reliability of experiments. As object velocity increases, there is a conflicting need for the observer to low-pass filter the noise and to visualize the object without blurring. The applicants proposed to characterize the potential degradation of dynamic visual acuity in fluoroscopy using new experiments that both dissect mechanisms nd demonstrate the practical relationship to digital processing. They proposed to close the loop between perception experiments and digital processing by extending current patio-temporal filters to include perception results. This research also applies to compression of medical image sequences and to optimization of MRI fluoroscopy.
Showing the most recent 10 out of 13 publications
