Speaker
Description
The Positron Emission Particle Tracking (PEPT) technique enables the tracking of a moving
radioactive tracer particle at high spatial and temporal resolution, from which its trajectory can
be accurately reconstructed. The uncertainty budget is complex and poorly understood,
particularly for derived quantities such as momenta, energies, and forces, which are typically
calculated using numerical differentiation techniques.
We report a filtering and data processing method based on a local polynomial least squares
fitting approach known as the Savitzky-Golay filter. The method is adapted to incorporate the
propagation of measurement uncertainties, maintaining them within useful bounds. The
method is benchmarked against several systems of known particle motion, including constant
velocity and constant acceleration, to place confidence limits on the results. Across all tested
regimes the Savitzky-Golay filter resolves higher precision than existing methods, providing
notable improvements to the uncertainty budget in PEPT analysis. We demonstrate tracking of
a particle moving up to 3 m/s with location precision within its diameter, and a 60% and 40%
average reduction in uncertainty bounds for velocity and acceleration respectively.
These results have motivated development of a high-resolution detector array for PEPT,
enabling measurements on the micro-scale by making immediate use of the improvements in
precision. Successful implementation will allow the meaningful application of PEPT to identified
problems in diagnostic medicine and in the study of micro-fluidic devices.
Apply to be considered for a student ; award (Yes / No)?
Yes
Level for award;(Hons, MSc, PhD, N/A)?
MSc
