Marc Chamberland and Paul Johns

Printer-friendly version
Date: 
Thursday, November 20, 2014

Time:   3:30 - 5:00 pm

Location:   Hospital Auditorium, 2nd floor, General Campus (escalator from main lobby to 2nd floor, follow signage)

Presentations:
1.       "List-mode motion tracking and correction for positron emission tomography imaging using low-activity fiducial markers"
Marc Chamberland - Carleton University
Abstract: Positron emission tomography (PET) imaging suffers from artifacts caused by patient body motion. We propose a method of tracking three-dimensional (3D) patient body motion during dynamic PET imaging by placing low-activity positron-emitting markers on a patient and using a tracking algorithm to extract the 3D motion information from the raw list-mode PET data. This information can then be used to perform motion correction on the raw list-mode data. Monte Carlo techniques were used to simulate a 92.5-kBq Na-22 marker moving sinusoidally in 3D. The simulated events were combined with list-mode data from patients undergoing cardiac PET imaging in order to test the algorithm. In experimental studies, three external Na-22 markers were placed on a dynamic torso phantom with an initial activity of approximately 680 MBq of Rb-82 in its cardiac insert. We tracked the motion of those markers while simulating breathing motion and patient drift with the phantom.  Results show that the tracking can achieve submillimetre precision and accuracy. In addition, the motion information was used to correct the raw list-mode data. Reconstructed images showed no perceivable translational motion compared to the original non-motion-corrected images. We conclude that this technique can potentially replace the need for additional and expensive respiratory-motion/triggering systems used for respiratory-gated or motion-free PET image reconstruction.

2.       "X-ray scatter imaging:  Cross sections, collimation, and signal extraction"
Paul Johns - Carleton University
Abstract: In diagnostic radiology, scattered photons comprise up to 90% of the radiation downstream of the patient and can provide useful information above and beyond that provided by the transmitted primary x rays.  Our development work on x-ray scatter imaging will be described.  First, since coherent scatter cross sections are too complicated to calculate, we have measured them for some normal tissues and phantom materials using an energy-dispersive system.  Second, step-and-shoot scatter imaging using multiple pencil beams has been demonstrated using 33.2 keV x rays at the Canadian Light Source synchrotron.  Collimation design options range from scanning a single pencil beam in tandem with a pixelated scatter detector over the patient, which is prohibitively slow but which captures all scatter information unambiguously, to multibeam geometries which speed acquisition but are reliant on pattern untangling algorithms since the diffraction ring patterns overlap.  These innovations in radiography are applicable both in medicine and in industrial nondestructive testing, security imaging, and other areas.