SBIR Phase II: AI-Powered Low-dose, Low-cost, High-Quality Computed Tomography (CT) Imaging

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is to expand access to advanced three-dimensional X-ray imaging during surgery without requiring hospitals to purchase expensive computed tomography scanners. Many surgical procedures, particularly spine operations, rely on two-dimensional X-ray images that can make it difficult to fully visualize anatomy and implanted hardware. Limited access to affordable three-dimensional imaging can increase procedure time, complication rates, and overall healthcare costs. This project seeks to enable existing mobile X-ray systems to produce high-quality three-dimensional images, allowing more procedures to be safely performed in outpatient surgical centers and community hospitals. If successful, the technology could reduce healthcare expenditures, improve patient safety, and lower radiation exposure by avoiding repeat scans. Commercially, the approach supports a scalable software-based model that upgrades widely deployed imaging equipment rather than replacing it, creating a large potential market across surgical centers in the United States and globally. Broader societal benefits include improved access to high-quality surgical care in rural and underserved regions, workforce development in advanced manufacturing and artificial intelligence, and strengthened national leadership in medical imaging innovation. This Small Business Innovation Research (SBIR) Phase II project aims to develop and clinically validate a new method for generating three-dimensional images from limited-angle X-ray data acquired by standard mobile surgical imaging systems. Conventional mobile systems primarily produce flat, two-dimensional images because they rotate over a small angle and operate under radiation dose constraints, limiting their ability to create accurate three-dimensional reconstructions. The project will refine artificial intelligence models that incorporate physical principles of X-ray imaging to reconstruct volumetric images from limited data. Research objectives include improving image quality and reliability across different imaging systems and patient anatomies, developing real-time calibration methods to correct for mechanical motion and geometric distortion, and validating performance in realistic surgical environments. The anticipated technical results are rapid, low-dose three-dimensional reconstructions with image clarity and geometric accuracy comparable to conventional computed tomography for specific surgical tasks. Successful completion of this work would demonstrate a practical pathway to deliver advanced three-dimensional guidance and navigation using equipment that is already widely available in operating rooms. 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.