Jensen-Shannon Divergences Research Articles

Attribute vector guided groupwise registration

Groupwise registration has been recently introduced to simultaneously register a group of images by avoiding the selection of a particular template. To achieve this, several methods have been proposed to take advantage of information-theoretic entropy measures based on image intensity. However, simplistic utilization of voxelwise image intensity is not sufficient to establish reliable correspondences, since it lacks important contextual information. Therefore, we explore the notion of attribute vector as the voxel signature, instead of image intensity, to guide the correspondence detection in groupwise registration. In particular, for each voxel, the attribute vector is computed from its multi-scale neighborhoods, in order to capture the geometric information at different scales. The probability density function (PDF) of each element in the attribute vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure in that local pattern. Eventually, with the help of Jensen–Shannon (JS) divergence, a group of subjects can be aligned simultaneously by minimizing the sum of JS divergences across the image domain and all attributes. We have employed our groupwise registration algorithm on both real (NIREP NA0 data set) and simulated data (12 pairs of normal control and simulated atrophic data set). The experimental results demonstrate that our method yields better registration accuracy, compared with a popular groupwise registration method.

Divergence analysis of atomic ionization processes and isoelectronic series

Fisher divergences (FDs) and Jensen-Shannon divergences (JSDs) are used in this work to quantify the informational discrepancies between the one-particle electron densities of neutral atoms, singly charged ions, and isoelectronic series. These dissimilarity magnitudes, computed for a set of 319 atomic systems in both position and momentum spaces, provide relevant information concerning pattern, structure, and periodicity properties of the ionization processes. In particular an apparent correlation between extremal values of the atomic ionization potential and the divergences is found. Results are compared with those obtained by quantum similarity techniques.

Fisher and Jensen–Shannon divergences: Quantitative comparisons among distributions. Application to position and momentum atomic densities

The Fisher divergence (FD) and Jensen-Shannon divergence (JSD) are used in this work with the aim of providing quantitative measures of the discrepancies between two arbitrary D-dimensional distribution functions, the FD being of local character and the JSD of global one. In doing so, the concepts of Fisher information and Shannon entropy associated to a distribution are the essential quantities for building up these comparative functionals. This kind of relative measures are here applied to the study of the one-particle densities in both conjugated spaces (position and momentum) of neutral atoms, discussing the results as compared to those provided by other previous functional measures. It is clearly shown how these divergences provide relevant information on the atomic shell structure, up to a level which depends on the considered space and measure.