Rationale for Correction:More GLCM texture measures are being used than are currently defined in the DICOM standard.E.g., Orlhac F, et al. Understanding Changes in Tumor Texture Indices in PET: A Comparison Between Visual Assessment and Index Values in Simulated and Patient Data. J Nucl Med. 2017 Mar 1;58(3):387–92. , which references Orlhac F, et al. Tumor Texture Analysis in 18F-FDG PET: Relationships Between Texture Parameters, Histogram Indices, Standardized Uptake Values, Metabolic Volumes, and Total Lesion Glycolysis. Journal of Nuclear Medicine. 2014 Mar 1;55(3):414–22. , which in turn describes differences in definitions and references Brooks FJ. On some misconceptions about tumor heterogeneity quantification. Eur J Nucl Med Mol Imaging. 2013 Aug 1;40(8):1292–4. , and does not use definitions from Tixier F et al. Intratumor Heterogeneity Characterized by
Textural Features on Baseline 18F-FDG PET Images Predicts Response to Concomitant Radiochemotherapy in Esophageal Cancer. J Nucl Med. 2011 Mar 1;52(3):369–78. or Tixier F et al. Reproducibility of Tumor Uptake Heterogeneity Characterization Through Textural Feature Analysis in 18F-FDG PET. J Nucl Med. 2012 May 1;53(5):693–700. , but rather those from Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. . Also, whether "energy" is ASM or root ASM can be confusing; clarify and add further references.[Ed.Note: Not all of Haralick's measures are included: do we need Sum Average, Sum Variance, Sum Entropy, Difference Variance, Difference Entropy, the two Information Measures of Correlation, or Maximal Correlation Coefficient? See also in Santos TA et al. MRI Texture Analysis Reveals Bulbar Abnormalities in Friedreich Ataxia. American Journal of Neuroradiology. 2015 Dec 1;36(12):2214–8. .][Ed.Note: Haralick et al defined "contrast" distinct from "sum of squares: variance" - we are not distinguishing the two. Why?][Ed.Note: For homegeneity, MATLAB describes use of absolute value of difference rather than square of it. Is this an issue?][Ed.Note: Do we also want to include Regional Heterogeneity Features described in Tixier' et al. from Gray Level Size Zone matrix described in Thibault G,et al. Texture Indexes and Gray Level Size Zone Matrix Application to Cell Nuclei Classification. In: 10th International Conference on Pattern Recognition and Information Processing. 2009. ?]
Correction Wording:
Amend DICOM PS3.16 as follows (changes to existing text are bold and underlined for additions and struckthrough for removals):
A statistical index of complexity comparing how detail in a fractal pattern changes with the scale at which it is measured; a ratio of the change in detail to the change in scale.
126060
Entropy of GLCM
The zero order entropy of a Gray Level Co-occurrence Matrix (GLCM). A measure of disorder.See , .See Haralick RM, Shanmugam K, Dinstein I. Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics. 1973 Nov;SMC-3(6):610–21. .
126061
EnergyRoot ASM of GLCM
The energy (uniformity) (square root of the Angular Second Moment (ASM)) of a Gray Level Co-occurrence Matrix (GLCM). A measure of orderliness.See .
Sometimes referred to as "energy", "uniformity" or "uniformity of energy" but then potentially confused with ASM.
126062
Homogeneity of GLCM
The Inverse Difference Moment of a Gray Level Co-occurrence Matrix (GLCM).See , , .See Haralick RM, Shanmugam K, Dinstein I. Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics. 1973 Nov;SMC-3(6):610–21. .
Homogeneity is 1 for a diagonal GLCM.MATLAB uses absolute value of difference rather than square of it.
126063
Contrast of GLCM
The sum of squares variance of a Gray Level Co-occurrence Matrix (GLCM).See , , .
Sometimes referred to as "variance" or "inertia".Contrast is 0 for a constant image.Haralick et al defined "contrast" distinct from "sum of squares: variance".
126064
Dissimilarity of GLCM
The dissimilarity of a Gray Level Co-occurrence Matrix (GLCM).See , .
126065
ASM of GLCM
The Angular Second Moment of a Gray Level Co-occurrence Matrix (GLCM).See , , .See Haralick RM, Shanmugam K, Dinstein I. Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics. 1973 Nov;SMC-3(6):610–21. .
Sometimes referred to as "energy", "uniformity" or "uniformity of energy" but then potentially confused with square root of ASM.ASM is 1 for a constant image.
126066
Correlation of GLCM
A measure of the linear dependency of grey levels on those of neighbouring pixels of a Gray Level Co-occurrence Matrix (GLCM).See , , .See Haralick RM, Shanmugam K, Dinstein I. Textural Features for Image Classification. IEEE Transactions on Systems, Man, and Cybernetics. 1973 Nov;SMC-3(6):610–21. .
Correlation is 1 or -1 for a perfectly positively or negatively correlated image, NaN for a constant image.
126067
Gray Level Co-occurrence Matrix (GLCM)
A tabulation of how often different combinations of pixel values (grey levels) occur in an image.See .
ddd001
Short Run Emphasis
A measure of the distribution of short runs in a gray level run length matrix. Abbreviated SRE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The SRE is highly dependent on the occurrence of short runs and is expected to be large for fine textures.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd002
Long Run Emphasis
A measure of the distribution of long runs in a gray level run length matrix. Abbreviated LRE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The LRE is highly dependent on the occurrence of long runs and is expected to be large for coarse structural textures.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd003
Low Gray Level Zone Emphasis
A measure of the distribution of low gray level values in a gray level run length matrix. Abbreviated LGRE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The LGRE is expected to be large for images with low gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd004
High Gray Level Zone Emphasis
A measure of the distribution of high gray level values in a gray level run length matrix. Abbreviated HGRE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The HGRE is expected lto be arge for images with high gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd005
Short Run Low Gray Level Emphasis
A measure of the joint distribution of short runs and low gray level values in a gray level run length matrix. Abbreviated SRLGE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The SRLGE is expected to be large for images with many short runs and lower gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd006
Short Run High Gray Level Emphasis
A measure of the joint distribution of short runs and high gray level values of in a gray level run length matrix. Abbreviated SRHGE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The SRHGE is expected to be large for images with many short runs and high gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd007
Long Run Low Gray Level Emphasis
A measure of the joint distribution of long runs and low gray level values of in a gray level run length matrix. Abbreviated LRLGE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The LRLGE is expected to be large for images with many long runs and low gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd008
Long Run High Gray Level Emphasis
A measure of the the joint distribution of long runs and high gray level values of in a gray level run length matrix. Abbreviated LRHGE.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The LRHGE is expected to be large for images with many long runs and high gray level values.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd009
Gray Level Nonuniformity
A measure of the similarity of gray level values through out the image of in a gray level run length matrix. Abbreviated GLNU.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The GLNU is expected to be small if the gray level values are alike throughout the image.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd010
Run Length Nonuniformity
A measure of the the similarity of the length of runs through out the image of in a gray level run length matrix. Abbreviated RLNU.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The RLNU is expected to be small if the run lengths are alike throughout the image.This measure may determined for runs in each direction (typically of 0°, 45°, 90°, and 135°) or by combining them into a global view.
ddd011
Run Percentage
A measure of the homogeneity and distribution of runs of an image in a specific direction of in a gray level run length matrix. Abbreviated RPC.See Xu D-H et al. Run-length encoding for volumetric texture. Heart. 2004. p. 25. .
The RPC is the largest when the length of runs is 1 for all gray levels in specific direction.