Wolfram R Jarisch PhD, Presenter: Nothing to Disclose

The above-noted three key advantages result typically in a low iteration count. Completeness of information extraction can be verified by examination of residuals. References: [1] Wood, S.L., Morf, M., A fast implementation of a minimum variance estimator for computerized tomography image reconstruction, IEEE BME-28, No. 2, pp. 56-68, 1981. [2] Yu, Z.,Thibault, et al., Fast Model-Based X-Ray CT Reconstruction Using Spatially Nonhomogeneous ICD Optimization, IEEE TIP-20, NO. 1, pp. 161-175, Jan. 2011. [3] Ramani, S., J. A. Fessler, A Splitting-Based Iterative Algorithm for Accelerated Statistical X-Ray CT Reconstruction, IEEE Vol. MI-31, No 3, pp. 677-688, March 2012. Sample TEM data (from 72 views) provided by Christian Kuebel of KIT; prediction, residuals, and gain.

Efficient computation of high quality 3D densities from projections continues to be a challenge. Early formulations of minimum variance estimation, a form of “statistical reconstruction”, [1] are rarely pursued due their numerical challenges. Recent alternative approaches to reduce CT radiation levels while striving for good image quality include model-based reconstruction [2] and decomposition methods [3] to better account for the nature of the estimation problem. Here a new numerically efficient Bayesian-derived statistical method is evaluated.

Jarisch, W, Statistical Validation of HECTOR a Highly Efficient Bayesian-Derived CT Density Estimator with Optimized Recursions.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12043861.html