31 August 2026 to 4 September 2026
Queen Mary University of London, London, UK
Europe/London timezone

Dose-efficient edge-illumination X-ray phase-contrast imaging with EIGER photon-counting detection

Not scheduled
20m
Mile End Campus: Graduate Centre Foyer and Peston Lecture Theatre (Queen Mary University of London, London, UK)

Mile End Campus: Graduate Centre Foyer and Peston Lecture Theatre

Queen Mary University of London, London, UK

Poster X-ray & Gamma Ray Detectors

Speaker

Alexander Duncalf (University College London)

Description

Edge illumination X-ray phase-contrast imaging (EI-XPCI) can allow dose reductions at constant image SNR, with great prospective benefit to biomedical applications. Photon counting detectors offer a route to this by providing a Poissonian response while eliminating read-out noise with appropriate energy thresholds, unlike energy integrating detectors where additive dark noise becomes increasingly important at low fluence.

We investigated this detector advantage for propagation distance optimisation. In a double mask EI-XPCI system, the object-detector distance was increased with fixed geometrical magnification, so changes in retrieved contrast were dominated by propagation sensitivity rather than beamlet widths or projected sample size. Concurrently, the source-object distance increases, reducing the dose. Measurements were performed with an EIGER 500K Hybrid Photon Counting detector for several refractive and scattering samples, compared with simulations. Additional flux-variance measurements were acquired with EIGER count-rate correction to test whether detector nonlinearity or correction artefacts contribute to observed noise response.

Mean flux followed an inverse-square dependence with distance. The refraction signal stayed constant with distance whereas dark-field decreased. Temporal variance increased with distance more than expected from the flux reduction, so the expected constant SNR dose reduction was not fully realised at larger distances. Flux-variance analysis showed that detector response remained Poissonian with and without count-rate correction, indicating that excess variance originates from the wider EI-XPCI system rather than from intrinsic detector statistics. These results identify photon counting detection as a promising route for dose efficient EI-XPCI, while highlighting the system stability requirements needed to exploit it quantitatively.

Author

Alexander Duncalf (University College London)

Co-authors

Dr Alberto Astolfo (University College London) Charlotte Hagen (University College London) Alessandro Olivo (University College London)

Presentation materials

There are no materials yet.