Quantitative imaging

Patent metadata

Field	Value
Publication number	US-9600877-B2
Application number	US-201314649723-A
Country	US
Kind code	B2
Filing date	Dec 16, 2013
Priority date	Oct 31, 2012
Publication date	Mar 21, 2017
Grant date	Mar 21, 2017

Abstract

The following generally relates to scaling irregularity maps based at least on one of a histogram bin width, an image noise or a contrast agent concentration. A method includes obtaining a non-scaled irregularity map generated based on local weighted histograms of voxel distributions about voxels of interest from volumetric image data of a subject or object. The local weighted histograms include a plurality of bins having a predetermined bin width. The local weights are determined based on a predetermined cluster length. The method further includes scaling the non-scaled irregularity map, generating a scaled irregularity map. The non-scaled irregularity map is scaled based at least on the histogram bin width.

First claim

The invention claimed is:

1. A method, comprising:
obtaining a non-scaled irregularity map generated based on local weighted histograms of voxel distributions about voxels of interest from volumetric image data of a subject or object, wherein the local weighted histograms include a plurality of bins having a predetermined bin width, and wherein the local weights are determined based on a predetermined cluster length; and
scaling the non-scaled irregularity map, generating a scaled irregularity map, wherein the non-scaled irregularity map is scaled based at least on the histogram bin width.

2. The method of claim 1 , further comprising:
representing the scaled irregularity map through an entropy value.

3. The method of claim 1 , further comprising:
representing the scaled irregularity map through one or more of an inverse of a uniformity of the local weighted histogram or high-order statistics of local weighted high-order histogram, where the high-order histogram is based on co-occurrence matrix, and the high-order statistics include entropy or inverse of uniformity functions.

4. The method of claim 1 , further comprising:
scaling the non-scaled irregularity as a function of a logarithm of the histogram bin width.

5. The method of claim 1 , wherein the scaled irregularity map is independent of the histogram bin width.

6. The method of claim 1 , wherein the predetermined bin width and the predetermined cluster length are optimized to give the highest definition of the irregularity maps.

7. The method of claim 6 , further comprising:
estimating a minimum bin width such that a maximum measurable entropy is equal to or less than a maximum available entropy due to use of a specific histogram weight mask and a maximal range of image values.

8. The method of claim 6 , further comprising:
determining a maximum bin width of interest based on image noise.

9. The method of claim 6 , further comprising:
determining optimal bin width and optimal cluster length based on maximizing a product of a variance and a shifted mean of the irregularity map values.

10. The method of claim 1 , further comprising:
scaling the non-scaled irregularity map based on an image noise.

11. The method of claim 10 , further comprising:
scaling the non-scaled irregularity map by direct deconvolution of a given noise histogram from a total histogram, which is a real texture histogram with added noise.

12. The method of claim 1 , further comprising:
scaling the non-scaled irregularity map based a contrast agent concentration.

13. The method of claim 12 , further comprising:
scaling the non-scaled irregularity map as a function of a negative log of a difference between an average image values of a contrast scan and an average image values of a non-contrast scan.

14. The method of claim 1 , further comprising:
visually presenting at least one of the scaled irregularity map or the non-scaled irregularity map either side by side with the volumetric image data or fused with the volumetric image data.

15. The method of claim 1 , further comprising:
visually presenting at least one of the scaled irregularity map or the non-scaled irregularity map using a color map along with a color-bar scale.

16. The method of claim 1 , further comprising:
visually presenting a value indicative of at least one scale limits, a log of the bin-width, an estimated image noise entropy, or an upper dynamic range limit.

17. The method of claim 1 , further comprising:
performing a region or volume of interest analysis of the irregularity map values.

18. The method of claim 1 , further comprising:
calculating a mean irregularity map value and a corresponding standard deviation within a region or volume of interest.

19. An image data processing system, comprising:
a scaled irregularity map generator that obtains a non-scaled irregularity map generated based on local weighted histograms of voxel distributions about voxels of interest from volumetric image data of a subject or object, wherein the local weighted histograms include a plurality of bins having a predetermined bin width and a predetermined cluster length, the scaled irregularity map generator, including:
a histogram bin-width scaler that scales the non-scaled irregularity map, generating a scaled irregularity map, wherein the non-scaled irregularity map is scaled based at least on one of the histogram bin width, an image noise or a contrast agent concentration.

20. The system of claim 19 , wherein the scaled irregularity map is represented via one or more of an entropy or an inverse of uniformity of the local weighted histogram, or high-order statistics of local weighted high-order histogram.

Assignees

• Koninklijke Philips Nv

Inventors

• Carmi Raz

Classifications

• G06T7/11

  Region-based segmentation · CPC title

• G06T2207/30004

  Biomedical image processing · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06T5/40

  using histogram techniques · CPC title

• A61B6/032

  Transmission computed tomography [CT] · CPC title