Scale selection for anisotropic scale-space: application to volumetric tumor characterization

1. A computer readable medium embodying instructions executable by a processor to perform method steps for determining a structure in volumetric data, the method steps comprising:

determining an anisotropic scale-space for a local region around a given spatial local maximum;

determining L-normalized scale-space derivatives in the anisotropic scale-space; and

determining the presence of noise in the volumetric data, upon determining noise in the volumetric data, determining the structure by a most-stable-over-scales determination, and upon determining noise below a desirable level, determining the structure by one of the most-stable-over-scales determination and a maximum-over-scales determination according to a size of a sampling range, wherein the maximum-over-scales is used when the sampling range is greater than a basin of attraction a spatial local maximum.

2. The method of claim 1 , wherein the most-stable-over-scales determination comprises:

determining a plurality of covariance estimates over an analysis scale set; and

determining a covariance estimate from among the plurality of covariance estimates having a minimum Jensen-Shannon divergence, wherein the covariance estimate defines a spread of the structure and the structure is determined in the volumetric data corresponding to the spread.

3. The method of claim 2 , wherein the analysis scale set is a given set of bandwidths over the volumetric data.

4. The method of claim 1 , wherein the a maximum-over-scales determination comprises:

determining Gamma- and L-normalized scale-space derivatives over an analysis scale set;

selecting a scale having a maximum normalized scale-space derivative, wherein the scale is a covariance defining a spread of the structure and the structure is determined in the volumetric data corresponding to the spread.

5. The method of claim 4 , wherein the Gamma- and L-normalized scale-space derivatives are determined with a constant normal having a Gamma- equal to ½.

6. The method of claim 4 , wherein the analysis scale set is a given set of bandwidths over the volumetric data.

7. The method of claim 1 , wherein the spatial local maximum indicates a location of the structure in the volumetric data.