Premium
A theoretical comparison of x‐ray angiographic image quality using energy‐dependent and conventional subtraction methods
Author(s) -
Tanguay Jesse,
Kim Ho Kyung,
Cunningham Ian A.
Publication year - 2012
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.3658728
Subject(s) - imaging phantom , monte carlo method , image quality , detector , subtraction , energy (signal processing) , flat panel detector , digital subtraction angiography , noise (video) , physics , photon , signal to noise ratio (imaging) , optics , angiography , computer science , radiology , mathematics , artificial intelligence , medicine , image (mathematics) , statistics , arithmetic , quantum mechanics
Purpose: X‐ray digital subtraction angiography (DSA) is widely used for vascular imaging. However, the need to subtract a mask image can result in motion artifacts and compromised image quality. The current interest in energy‐resolving photon‐counting (EPC) detectors offers the promise of eliminating motion artifacts and other advanced applications using a single exposure. The authors describe a method of assessing the iodine signal‐to‐noise ratio (SNR) that may be achieved with energy‐resolved angiography (ERA) to enable a direct comparison with other approaches including DSA and dual‐energy angiography for the same patient exposure. Methods: A linearized noise‐propagation approach, combined with linear expressions of dual‐energy and energy‐resolved imaging, is used to describe the iodine SNR. The results were validated by a Monte Carlo calculation for all three approaches and compared visually for dual‐energy and DSA imaging using a simple angiographic phantom with a CsI‐based flat‐panel detector. Results: The linearized SNR calculations show excellent agreement with Monte Carlo results. While dual‐energy methods require an increased tube heat load of 2× to 4× compared to DSA, and photon‐counting detectors are not yet ready for angiographic imaging, the available iodine SNR for both methods as tested is within 10% of that of conventional DSA for the same patient exposure over a wide range of patient thicknesses and iodine concentrations. Conclusions: While the energy‐based methods are not necessarily optimized and further improvements are likely, the linearized noise‐propagation analysis provides the theoretical framework of a level playing field for optimization studies and comparison with conventional DSA. It is concluded that both dual‐energy and photon‐counting approaches have the potential to provide similar angiographic image quality to DSA.