Volume 28, Issue 8 p. 1562-1576
Radiation imaging physics

Correcting organ motion artifacts in x-ray CT medical imaging systems by adaptive processing. I. Theory

A. C. Dhanantwari

A. C. Dhanantwari

Biomedical Sciences Section, Defence and Civil Institute of Environmental Medicine, 1133 Sheppard Avenue West, Toronto, Ontario M3M 3B9, Canada

Search for more papers by this author
S. Stergiopoulos

S. Stergiopoulos

Biomedical Sciences Section, Defence and Civil Institute of Environmental Medicine, 1133 Sheppard Avenue West, Toronto, Ontario M3M 3B9, Canada

Department of Electrical and Computer Engineering, The University of Western Ontario, London, Ontario N6A 5B9, Canada

Search for more papers by this author
I. Iakovidis

I. Iakovidis

European Commission, DG-IST, Avenue de Beaulieu 29, B-1160 Brussels, Belgium

Search for more papers by this author
First published: 09 August 2001
Citations: 13

Abstract

X-ray CT scanners provide images of transverse cross sections of the human body from a large number of projections. During the data acquisition process, which usually takes about 1 s, motion effects such as respiration, cardiac motion, and patient restlessness produce artifacts that appear as blurring, doubling, and distortion in the reconstructed images, and may lead to inaccurate diagnosis. To address this problem several processing techniques have been proposed that require a priori knowledge of the motion characteristics. This paper proposes a method, which makes no assumptions about the properties of the motion, to eliminate the motion artifacts. The approach in this paper uses a spatial overlap correlator scheme to accurately track organ motion in computed tomography imaging systems. Then, it is shown that as optimum processing scheme to remove organ motion effects is to apply adaptive interference cancellation (AIC) methods, which treat the output of the spatial overlap correlator as noise interference at the input of the AIC process. Furthermore, an AIC method does not require any kind of periodicity of the motion effects. Synthetic data tests demonstrate the validity of this approach and show that hardware modifications are essential for its implementation in x-ray CT medical imaging systems.