Yong Long receives Best Poster Award for work in medical imaging

Long’s work describes a new algorithm for performing model-based methods in a way that requires less computation yet provides improved image quality.

Long and Fessler Enlarge

Yong Long, doctoral student in Electrical Engineering:Systems, and co-authors Prof. Jeffrey A. Fessler(Professor of EECS, Biomedical Engineering, and Radiology) and Prof. James M. Balter (Professor of Radiation Oncology) received a Best Poster award at the 10th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, held in Beijing, China. Their paper was titled, “A 3D Forward and Back-Projection Method for X-Ray CT Using Separable Footprint.”

Model-based methods for 3D tomographic image reconstruction have the potential to significantly improve image quality and reduce patient X-ray dose in medical computed tomography (CT) scans. The primary drawback of model-based methods compared to conventional image reconstruction methods is the increased computation time. This work describes a new algorithm for performing the most computationally expensive portion of model-based methods in a way that requires less computation than traditional methods yet provides improved image quality. The method is being evaluated by U-M researchers on patient X-ray CT scans for possible use in next generation CT systems. It is also being investigated for applications in radiation therapy.