Noise reduction in computed tomography using a multiplicative continuous-time image reconstruction method

Yusaku Yamaguchi; Takeshi Kojima; Tetsuya Yoshinaga

doi:10.1117/12.2216439

30 March 2016 Noise reduction in computed tomography using a multiplicative continuous-time image reconstruction method

Yusaku Yamaguchi, Takeshi Kojima, Tetsuya Yoshinaga

Proceedings Volume 9783, Medical Imaging 2016: Physics of Medical Imaging; 97834T (2016) https://doi.org/10.1117/12.2216439
Event: SPIE Medical Imaging, 2016, San Diego, California, United States

Abstract

In clinical X-ray computed tomography (CT), filtered back-projection as a transform method and iterative reconstruction such as the maximum-likelihood expectation-maximization (ML-EM) method are known methods to reconstruct tomographic images. As the other reconstruction method, we have presented a continuous-time image reconstruction (CIR) system described by a nonlinear dynamical system, based on the idea of continuous methods for solving tomographic inverse problems. Recently, we have also proposed a multiplicative CIR system described by differential equations based on the minimization of a weighted Kullback–Leibler divergence. We prove theoretically that the divergence measure decreases along the solution to the CIR system, for consistent inverse problems. In consideration of the noisy nature of projections in clinical CT, the inverse problem belongs to the category of ill-posed problems. The performance of a noise-reduction scheme for a new (previously developed) CIR system was investigated by means of numerical experiments using a circular phantom image. Compared to the conventional CIR and the ML-EM methods, the proposed CIR method has an advantage on noisy projection with lower signal-to-noise ratios in terms of the divergence measure on the actual image under the same common measure observed via the projection data. The results lead to the conclusion that the multiplicative CIR method is more effective and robust for noise reduction in CT compared to the ML-EM as well as conventional CIR methods.

Citation Download Citation

Yusaku Yamaguchi, Takeshi Kojima, and Tetsuya Yoshinaga "Noise reduction in computed tomography using a multiplicative continuous-time image reconstruction method", Proc. SPIE 9783, Medical Imaging 2016: Physics of Medical Imaging, 97834T (30 March 2016); https://doi.org/10.1117/12.2216439

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available