Electronic Radiology Laboratory (ERL)

Advanced Algorithims for Security Applications

X-ray imaging and computed tomography (CT) are widely used tools in Radiology and in security applications such as baggage scanning at airports. CT images are a representation of the linear attenuation coefficient in each voxel of the tissue or object illuminated by the x-ray beam. Ordinarily, scattered radiation in the measurements is considered to be deleterious “noise” that degrades the quality of images interpreted by Radiologists. Researchers at the ERL and in Electrical and Systems Engineering (ESE) at the Washington University School of Engineering, along with a team of researchers at Duke University, have devised new methods for utilizing measurements of scattered radiation. Instead of estimating the attenuation coefficients of objects, the chemical composition of the material is determined, which allows dangerous materials to be readily identified. This work is sponsored by the Department of Homeland Security.

In another related project, also sponsored by the Department of Homeland Security, researchers in the ERL and in ESE are applying image-reconstruction algorithms originally devised for emission tomography and CT to the baggage scanning problem. In the medical context, these algorithms yield improved images with more accurate quantification compared to standard algorithms.

Collaborators: David Brady, Lawrence Carin
Grant: Department of Homeland Security, BAA 12-07-EXD.01-0007-I, Pi: Joseph O’Sullivan,Ph.D. and Industry subcontract, SureScan, Inc.,  Contract HSHQDC-14-C-B0011