Statistical Image Reconstruction Process for XRay CT with Compensation for Polyenergetic Source Spectrum

Researchers in the electrical engineering department at The University of Michigan have developed a novel method for image reconstruction that improves image clarity and quantitative accuracy. This method should increase use of CT scanning for quantitative measurements and improve the diagnostic accuracy of X-ray CT scanning.||Current CT scanning image reconstruction algorithms are based on over-simplified models for the nonlinearities in CT imaging caused by the wide spectrum of X-ray photon energies. The resulting beam hardening effect causes artifacts (streaks and shadows) in CT images that impair diagnostic performance and limits quantitative accuracy of the CT scans. The algorithms used here properly correct for these effects using a proprietary methodology. The iterative algorithm used for this technology also significantly reduces the computer processing time, overcoming a limitation that has hampered previous iterative methods. The result is clearer, more precise CT images and improved accuracy of quantitative CT data, achieved with feasible computation time.|
US and foreign patent applications have been filed. The university seeks a development and marketing partner interested in rights to bring the technology to the marketplace. |

Inventor(s): Jeffrey A. Fessler,v

Type of Offer: Licensing



« More Medical Patents

---