A New Approach to Image Reconstruction from Projections Using a Recurrent Neural Network

Robert Cierniak

doi:10.2478/v10006-008-0014-y

.blurhash-client-img { display: none !important; }

A New Approach to Image Reconstruction from Projections Using a Recurrent Neural Network

International Journal of Applied Mathematics and Computer Science

Volume 18 (2008): Issue 2 (June 2008)

By: Robert Cierniak

Open Access

|Jun 2008

Abstract

A new neural network approach to image reconstruction from projections considering the parallel geometry of the scanner is presented. To solve this key problem in computed tomography, a special recurrent neural network is proposed. The reconstruction process is performed during the minimization of the energy function in this network. The performed computer simulations show that the neural network reconstruction algorithm designed to work in this way outperforms conventional methods in the obtained image quality.

References

Averbuch A., Coifman R.R., Donoho D.L., Israeli M. and Waldén J. (2001). A notion of Radon transform for data in a Cartesian grid, which is rapidly computable, algebraically exact, geometrically faithful and invertible, Technical report TR No. 2001-11, Department of Statistics, Stanford University, USA.
Search in Google Scholar
Censor Y. (1983). Finite series-expansion reconstruction methods, Proceeding of the IEEE 71(3): 409-419.10.1109/PROC.1983.12598
Search in Google Scholar
Cichocki A., Unbehauen R., Lendl M. and Weinzierl K. (1995). Neural networks for linear inverse problems with incomplete data especially in application to signal and image reconstruction, Neurocomputing 8: 7-41.10.1016/0925-2312(94)E0063-W
Search in Google Scholar
Cierniak R. and Rutkowski L. (2000). On image compression by competitive neural networks and optimal linear predictors, Signal Processing: Image Communication 15(6): 559-565.10.1016/S0923-5965(99)00013-2
Search in Google Scholar
Cierniak R. (2002). Image reconstruction from projection using unsupervised neural network, Proceedings of the Joint 1-st International Conference on Soft Computing and Intelligent Systems and 3-rd International Symposium on Advanced Intelligent Systems, Tsukuba, Japan.
Search in Google Scholar
Cierniak R. (2006). A novel approach to image reconstruction from projections using Hopfield-type neural network, in Rutkowski L., Tadeusiewicz R., Zadeh L. A., Żurada J. (Eds.), Proceedings of the 8-th International Conference on Artificial Intelligence and Soft Computing, Zakopane, Poland, LNAI, Springer, Berlin, pp. 890-898.
Search in Google Scholar
Cormack A.M. (1963). Representation of a function by its line integrals with some radiological application, Journal of Applied Physics 34: 2722-2727.10.1063/1.1729798
Search in Google Scholar
Gordon R., Bender R. and Herman G.T. (1970). Algebraic reconstruction techniques (ART) for three-dimensional electron microscopy and X-ray photography, Journal of Theoretical Biology 29: 471-481.10.1016/0022-5193(70)90109-8
Search in Google Scholar
Hopfield J.J. (1982). Neural networks and physical systems with emergent collective computational abilities, Proceedings of the National Academy of Sciences of the USA 79: 2554-2558.10.1073/pnas.79.8.2554
Search in Google Scholar
Ingman D. and Merlis Y. (1992). Maximum entropy signal reconstruction with neural networks, IEEE Transactions on Neural Networks 3: 195-201.10.1109/72.125860
Search in Google Scholar
Jaene B. (1991). Digital Image Processing - Concepts, Algoritms and Scientific Applications, Springer, Berlin, Heidelberg.
Search in Google Scholar
Jain A.K. (1989). Fundamentals of Digital Image Processing, Prentice Hall, Englewood Cliffs, NJ.
Search in Google Scholar
Kaczmarz S. (1937). Angeneaherte Aufloesung von Systemen Linearer Gleichungen, Bulletin de l'Académie Polonaise des Sciences et Lettres 35: 355-357.
Search in Google Scholar
Kak A.C. and Slanley M. (1988). Principles of Computerized Tomographic Imaging, IEEE Press, New York.
Search in Google Scholar
Kerr J.P. and Barlett E.B. (1995a). A statistically tailored neural network approach to tomographic image reconstruction, Medical Physics 22: 601-610.10.1118/1.597586
Search in Google Scholar
Kerr J.P. and Barlett E.B. (1995b). Medical image processing utilizing neural networks trained on a massively parallel computer, Computers in Biology and Medicine 25: 393-403.10.1016/0010-4825(95)00017-X
Search in Google Scholar
Kerr J.P. and Barlett E.B. (1995c). Neural network reconstruction of single-photon emission computed tomography images, Journal of Digital Imaging 8: 116-126.10.1007/BF031680857488654
Search in Google Scholar
Kingston A. and Svalbe I. (2003). Mapping between digital and continuous projections via the discrete Radon transform in Fourier space, Proceedings of the 7-th Conference on Digital Image Computing: Techniques and Applications, Sydney, pp. 263-272.
Search in Google Scholar
Knoll P., Mirzaei S., Muller A., Leitha T., Koriska K., Kohn H. and Neumann M. (1999). An artificial neural net and error backpropagation to reconstruct single photon emission computerized tomography data, Medical Physics 26: 244-248.10.1118/1.598511
Search in Google Scholar
Lewitt R.M. (1983). Reconstruction algorithms: Transform methods, Proceeding of the IEEE 71(3): 390-408.10.1109/PROC.1983.12597
Search in Google Scholar
Luo F.-L. and Unbehauen R. (1998). Applied Neural Networks for Signal Processing, Cambridge University Press, Cambridge, UK.
Search in Google Scholar
Munllay M.T., Floyd C.E., Bowsher J.E. and Coleman R. E. (1994). An artificial neural network approach to quantitative single photon emission computed tomographic reconstruction with collimator, attenuation and scatter compensation, Medical Physics 21: 1889-1899.10.1118/1.597167
Search in Google Scholar
Ramachandran G.N. and Lakshminarayanan A.V. (1971). Three-dimensional reconstruction from radiographs and electron micrographs: II. Application of convolutions instead of Fourier transforms, Proceedings of the National Academy of Sciences of the USA 68: 2236-2240.
Search in Google Scholar
Srinivasan V., Han Y.K. and Ong S.H. (1993). Image reconstruction by a Hopfield neural network, Image and Vision Computing 11(5): 278-282.10.1016/0262-8856(93)90005-2
Search in Google Scholar
Wang Y. and Wahl F.M. (1997). Vector-entropy optimization-based neural-network approach to image reconstruction from projections, IEEE Transaction on Neural Networks 8(5): 1008-1014.10.1109/72.62320218255703
Search in Google Scholar