Hexagonal image grids improve CT scan reconstruction quality
Hexagonal Basis Projectors for Model-Based Iterative CT Reconstruction
Computer Vision and Pattern Recognition
Summary
CT scans create pictures of the inside of the body by measuring X-rays from many angles. The way the image is built from the data affects how clear and accurate these pictures are. The authors show that using a hexagonal grid for this image building, instead of the usual square grid, can make the process more efficient and capture details better. Their method works well both in computer simulations and when tested on real CT data. This approach offers a practical way to improve CT images used for diagnosis.
What this means in practice
- •For medical imaging engineers: Implement hexagonal pixel models for more accurate and efficient CT image reconstruction from projection data in fan-beam geometries.
- •For gpu software developers: Develop high-performance forward and backprojection algorithms using hexagonal basis functions to accelerate CT iterative reconstruction.
Authors
Ke Chen, Alireza Entezari
Abstract
Image representation affects both the forward model and quantitative feature recovery in model-based iterative CT reconstruction. For isotropically bandlimited signals, the hexagonal lattice is the most sampling-efficient lattice in two dimensions, with greater directional isotropy and a larger alias-free bandwidth than the Cartesian grid at equal density. We develop constant and linear hexagonal projectors for two-dimensional fan-beam CT using box spline calculus. Both bases form a partition of unity on common coefficient centers; the linear tent also reproduces affine functions. We derive analytic ray footprints and effective detector-blur expressions and implement efficient GPU forward and backprojection. On simulated FORBILD data, the linear model improves on Cartesian and hexagonal pixels, with its best noisy-data accuracy at the coarser resolution. Real-world 2DeteCT comparisons use high-dose, high-resolution Cartesian reconstructions as pseudo ground truth and show competitive image and feature agreement. These results establish a practical hexagonal representation for quantitative CT.