Making Binary Decisions Based on the Posterior Probability Distribution Associated with Tomographic Reconstructions

Abstract

This report was prepared as an account of work sponsored by an agency of the United States Government, Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or .!'i:. process disclosed, or represents that its use would not infringe privately owned rights. Refer' ,',:, ence herein to any specific commercial product, process, or service by trade name, trademark, ' _, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom-_" , ,_ mendation, or favoring by the United States Government or any agency thereof, The views /_i[_[/ [) i: ,,_,,. and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. ' /_._],_ By acceptance of this article, the p_,'bli_i_errec,_gnlzesthat the U.S. Government retains a nonexclusive, royalty.fre_ license to publish or reproduce the published form of this contribution, or to allow others to do so, for U.S. Government purposes.

Bayesian Maximum Entropy Based Algorithm for Digital X-ray Mammogram Processing

“…The best suited prior probability distribution, P (f |α, H), should incorporate as much as possible of the available statistical characteristics of the ensemble of images to which the original image, f , is assumed to belong. Apart from the entropic one [28], there are more different choices to be considered and evaluated for prior probability distribution of an image.…”

<title>Introduction to Bayesian image analysis</title>

“…This point has been emphasized by the author 33,34 by displaying the posterior probability associated with a tomographic reconstruction as a function of two pixel values. Even when a pixel value is zero in a constrained reconstruction, the posterior probability for a positive value of that pixel can be readily calculated.…”

“…Binary decisions have been made using the posterior probability density functions associated with reconstructed images. 20,[33][34][35] To date, this more complicated approach does not perform any better than basing the decision on the mean square difference between the reconstruction and the two test images.…”

“…This issue has been investigated briefly 33,34 in regard to improving the performance of binary tasks without any firm conclusion. It is apparently not a large effect.…”

Performance of the Rayleigh task based on the posterior probability of tomographic reconstructions

“…The sum is over ali pixels in the in]age and ii is aptly theory [7,10]. We designate this decision method as PP.…”