The extended MENT algorithm: a maximum entropy type algorithm using prior knowledge for computerized tomography

Unlike MENT (maximum entropy algorithm), the extended MENT algorithm can process prior information and deal with incomplete projections or limited angle data. The reconstruction problem is formulated for solving linear systems involving the Fredholm integral equation. To develop the extended MENT algorithm, maximum entropy is substituted by a more general optimization criterion, that of minimizing the discriminatory function. The a priori knowledge of the shape of the object is easily incorporated in the algorithm by using the discriminatory function. Useful mathematical properties that make the discriminatory function attractive are derived. The sensitivity of the minimum discriminatory solution is derived to determine the characteristics of the noise in the reconstructed images. The extended MENT algorithm is developed for a parallel geometry, and its convergence properties are given. Its image processing performance is better than that for other maximum entropy algorithms such as multiplicative algebraic reconstruction techniques (MART) or more standard methods such as ART and the convolution backprojection. >

[1]  B. R. Frieden,et al.  Computational methods of probability and statistics , 1980 .

[2]  Y. Censor.  Finite series-expansion reconstruction methods , 1983, Proceedings of the IEEE.

[3]  B. Frieden.  Dice, entropy, and likelihood , 1985, Proceedings of the IEEE.

[4]  R. Lewitt.  Reconstruction algorithms: Transform methods , 1983, Proceedings of the IEEE.

[5]  M. Reha Civanlar,et al.  Error bounds for iterative reprojection methods in computerized tomography , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[6]  E. Jaynes.  On the rationale of maximum-entropy methods , 1982, Proceedings of the IEEE.

[7]  John Skilling,et al.  Data analysis: The maximum entropy method , 1984, Nature.

[8]  Abraham Charnes,et al.  Computation of minimum cross entropy spectral estimates: An unconstrained dual convex programming method , 1986, IEEE Trans. Inf. Theory.

[9]  G. N. Minerbo,et al.  MENT: A maximum entropy algorithm for reconstructing a source from projection data , 1979 .

[10]  J. P. Ponstein.  Approaches to the theory of optimization: List of symbols , 1980 .

[11]  W. B. Gilboy,et al.  Recent developments in the industrial application of computerized tomography with ionizing radiation , 1984 .

[12]  N. J. Dusaussoy,et al.  Performance of maximum entropy algorithms for image reconstruction from projections , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[13]  Y. Censor,et al.  On some optimization techniques in image reconstruction from projections , 1987 .

[14]  R. A. Armistead.  CT: Quantitative 3-D inspection , 1988 .

[15]  J. Darroch,et al.  Generalized Iterative Scaling for Log-Linear Models , 1972 .

[16]  D. Titterington,et al.  The maximum entropy method for data analysis , 1984, Nature.

[17]  Rodney W. Johnson,et al.  Axiomatic derivation of the principle of maximum entropy and the principle of minimum cross-entropy , 1980, IEEE Trans. Inf. Theory.

[18]  R.H.T. Bates,et al.  Overview of computerized tomography with emphasis on future developments , 1983, Proceedings of the IEEE.