Physical optimization of acquisition and visualization parameters for the management of complex metallic artifacts in CT

Abstract

Managing metallic artifacts in Computed Tomography (CT) represents a persistent challenge that degrades diagnostic quality due to physical phenomena such as beam hardening and scattering. This study proposes an optimized protocol on a Philips MX16 console, based on the manipulation of physical and perceptual variables to overcome the limitations of automatic reconstruction software. Methodologically, the "Hard" approach involves increasing photon penetrance (140 kV), reducing collimation (8x0.75 mm / 10x0.75 mm), and employing high spatial frequency kernels (Bone). These choices are supported by an increase in tube load (250-300 mAs) to stabilize the signal-to-noise ratio. The results highlight how the use of extended windowing (WW 3000+ HU) allows for the discrimination of cortical bone from hyperdense streaks, preserving the definition of periprosthetic bone trabeculation. Manual optimization of parameters proves superior to software algorithms alone, restoring centrality to the technical expertise of the Radiology Technologist.