Evaluation of automatic exposure control in CT

Project: Research

Description

Using automatic exposure control the tube current (radiation dose) can be adjusted according to each patient's anatomy and the study area in three dimensions. This technique has shown to be effective at reducing absorbed dose to patients undergoing CT examinations.

Today, all manufacturers offer automatic exposure control systems. Each system works in different ways, but the basic principle is that the user specifies a level of required image quality. Tube current, i.e. intensity of the radiation is then adapted to each patient's size, shape and attenuation so that the selected image quality level is achieved with improved radiation efficiency. Tube current is proportional to the radiation dose, i.e., a 50% reduction in tube current reduces the radiation dose to the patient by half.

We have investigated the potential for reducing radiation dose and thereby resulting image quality when automatic exposure control system from four different manufacturers are used; Siemens, Philips, General Electric and Toshiba. Measurements were performed using anthropomorphic phantoms. The results of our studies shows that the dose regulation systems adjust tube current similar to the anatomy of the respective phantom. E.g. for less dense objects such as the air-filled lungs that did not attenuate the radiation quite as much as the shoulder area with a lot of bone, it is possible to reduce the tube current and thus the radiation dose. The studies show that there are good opportunities to obtain significant dose reductions, about 50% for a CT chest examination. A common result was that the image noise increased, but the noise level became more consistent between different anatomical regions.
StatusNot started

Participants

Related research output

Marcus Söderberg & La, S., 2013, Medical Imaging 2013: Physics of Medical Imaging. Robert M., N., Bruce R., W. & Christoph, H. (eds.). SPIE, Vol. 8668. p. 866833

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding

View all (4)