Adaptive kernel-based scatter correction for multi-source stationary CT with non-circular geometry

Abstract

Tomographic systems based on stationary arrangements of compact x-ray sources coupled to curved panel detectors have shown great potential for point-of-care brain imaging, but suffer from large, non-isotropic x-ray scatter. This work presents an adaptive kernel strategy to efficiently estimate scatter in stationary multi-source CT. The adaptive scatter estimation handles non-circular geometries, by the addition of pre- and post-processing steps to projection domain scatter estimators. The method was calibrated and evaluated on simulated data for a previously presented system with 31 x-ray sources on a circular arc coupled to a curved detector. Further assessment was obtained on experimental data obtained with an imaging testbench including a compact CNT-based x-ray source and simulating the scanner geometry. The method achieved accurate air-normalized scatter distributions across x-ray source positions and detector pixels, yielding a mean absolute error of 1.98𝑥10−3 with respect to the Monte-Carlo ground truth. Air-gap compensation had the largest impact on final accuracy. Image quality for simulated data showed consistent mitigation of scatter artifacts and reduction in non-uniformity from NU = 109 HU to 24 HU, with comparable performance for variations in cranium size, ranging in length from 161 mm (NU =14 HU) to 246 mm (NU = 15 HU). The experimental data showed comparable performance with error attributable to slight simulation infidelity. This work presents an adaptive approach to scatter compensation in multi-source, non-circular geometries using warping and weighting operations coupled to kernel-based scatter estimation on a virtual circular geometry, with immediate extension to other projection-based scatter compensation strategies.

Multi-source semi-stationary CT for brain imaging: development and assessment of a prototype system and image formation algorithms

“…Compensation of scatter in the view-variant geometry posed by the proposed system was achieved via an adaptive deep scatter estimation approach (ADSE): projection domain scatter estimators were first trained in a circular, invariant CBCT geometry and a set of warping and weighting operations was developed to adapt the invariant scatter estimation to the noncircular, stationary geometry of the MXA scanner. In the proposed approach, which follows a similar methodology to our previous work [6], scatter corrupted projections from the curved panel were first registered to a virtual flat-panel, via normalized forward projection onto a plane orthogonal to the central ray of the beam. The virtual projection was then used as the input to a conventional projection-based scatter estimator.…”

“…Our previous work showed mitigation of sparse and limited sampling effects in stationary CT for brain imaging via novel posterior sampling strategies within unsupervised learning-based diffusion generative models [5]. Feasibility of accurate estimation of scatter in stationary MXA geometries was demonstrated via adaptive deep scatter estimation methods that effectively combined estimation of scatter in a virtual, invariant, CBCT geometry with adaptive warping and ray-dependent weighting operators [6].…”

Stationary x-ray tomography for hemorrhagic stroke imaging: sampling and resolution properties

“…The x-ray technique was 102 kV (+2 mm Cu filtration) and 0.5 mAs/projection. Prior to reconstruction, raw projection data were flat-field corrected, followed by projection-domain scatter compensation [10] and log conversion. PWLS and DPS reconstructions were performed on a 256 x 256 x 200 voxel grid (1 mm isotropic).…”

“…Lighter and more compact systems have been proposed by leveraging stationary CT (sCT) configurations with multi-x-ray source arrays (MXAs) based on cold-cathode technology -e.g., Carbon Nanotubes (CNT) [8]. Previous work showed the feasibility of using an ultra-portable sCT for ICH detection [9], [10], with a configuration including a MXA and a curved area detector arranged in a cone beam geometry suitable for pointof-care deployment in conventional emergency vehicles. However, its imaging geometry presents an unconventional sampling pattern that challenges image reconstruction [9].…”