SBIR Phase II: Deep-learning Driven Single-shot X-ray Phase-contrast Computed Tomography for Security Screening

The broader impact of this Small Business Innovation Research (SBIR) Phase II project is to improve security screening while reducing delays, costs, and stress for travelers and inspectors. The project will advance an X-ray imaging upgrade that gives computed tomography scanners three complementary types of information from a single scan, helping separate dangerous items from ordinary belongings and shipped goods more accurately than today’s absorption-only systems. The technology will reduce false alarms, decrease unnecessary manual inspections, shorten airport screening lines, and improve throughput without requiring full replacement of installed scanner fleets. The project will also enhance scientific and technological understanding by translating phase-contrast X-ray imaging and artificial intelligence from laboratory demonstrations into a practical, high-energy, tri-modal CT system suitable for real-world deployment. This Small Business Innovation Research (SBIR) Phase II project will address a major limitation of current X-ray security systems: many threat materials and benign items produce similar absorption signals, making accurate differentiation difficult and contributing to false alarms. The research objective is to validate an AI-driven X-ray phase-contrast CT method using a single-shot approach that can simultaneously produce absorption, differential phase-contrast, and dark-field information on a commercial-style scanner. The proposed research includes fabrication of high-energy X-ray gratings, integration of the hardware into a prototype matching commercial CT geometry, development of artificial-intelligence-based reconstruction methods for real security-relevant data, and testing with representative benign items and explosive simulants at relevant imaging speeds. The technical result will be a prototype demonstration showing improved material differentiation while preserving practical CT workflow, providing the technical basis for later operational testing and commercialization in security screening. 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.