I-Corps: Spectral Computed Tomography for Material Identification and Threat Detection

University Of Delaware, Newark DE

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the ability to perform material characterization using X-ray illumination at lower costs than conventional spectral computed tomography (CT) systems. Spectral CT aids in applications ranging from bomb detection to identification of cancerous tissue by identifying the chemical composition of the internal structure of objects. The approach that is proposed is based on the newly introduced concept of colored X-ray filters. Particularly, it relies on X-ray coded projections and computational algorithms to circumvent many of the current limitations of spectral CT systems, such as high costs and sensing complexity. Rather than measuring objects by the attenuated X-rays, the X-rays are first coded in amplitude and spectra by blocking or unblocking a known pattern, effectively creating a coded aperture on the path of the X-ray measurements. Coded apertures create lower-dose structured X-ray bundles without compromising the image reconstruction quality. The aim is to provide a cost-effective solution using conventional X-ray imaging systems with the addition of the colored X-ray filters and computational algorithms to obtain material-specific images. Lower cost systems for spectral CT images would allow a broader deployment of such systems in cost-sensitive venues. This I-Corps project uses structured illumination across time, space, and spectra to interrogate the underlying objects and attain coded projections. Computational algorithms then combine the coded measurements to reconstruct material-based images useful for diagnosis and analysis in security and medical applications. Extensive simulations of the proposed spectral CT system and explored optimal architectures for structured illumination patterns have been performed. For the simulation of the CT system, the software was used to simulate the X-ray source spectrum with the proposed colored X-ray filters and included the NIST database. Simulations using diverse materials have been performed to study the feasibility and performance of the proposed systems. The results have been published in multiple peer-reviewed publications by the EL and TL. Additionally, an X-ray imaging test-bed was designed to do experimental validation of the proposed spectral CT architectures. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.