Radiation Dose Reduction in X-ray Computed Tomography

Washington University, Saint Louis MO

Investigators

Linked publications & trials

Abstract

X-ray computed tomography (CT) is invaluable in modern health care. Because of its relatively high radiation dose, CT now accounts for a disproportionate amount of medical-based radiation. Although a reduction in dose for CT examinations is possible and would reduce somatic and genetic risk, dose reduction results in the loss of some image quality. The degree of dose reduction possible without deleteriously affecting diagnostic performance for CT is unknown. This is because it is either practical nor ethical to perform multiple exposures of patients to study how diagnostic performance varies with dose reduction. We plan to overcome this problem by developing a software-based simulator for evaluating dose reduction. With this simulation environment, we will use existing CT scans to retrospectively generated reduced-dose CT images, allowing expert observers to determine optimal dose levels. Our long- term goal is to reduce the radiation dose involved with CT. Our central hypothesis is that simulations will produce realistic low-dose CT images that can support a scientific framework for establishing lower-dose clinical protocols. Our objective in this application is to determine the feasibility of using our dose-reduction simulation software in studies of diagnostic performance. The following specific aims are proposed: 1) implement and validate dose-reduction simulation software for clinical CT imaging; 2) assess potential radiation dose reduction for CT while maintaining diagnostic confidence; 3) perform a pilot ROC study to establish the parameters required for measurements of diagnostic performance versus radiation dose level. The use of our dose-reduction software simulator is innovative in that currently there is no practical method available for studying the effects of CT dose reduction on diagnostic performance with clinical patients. Accomplishing these objectives will enable us to define a systematic research program for establish clinical dose-reduction protocols.