Pyradiomics (GLCM) calculation

Community Radiomics
,
1

Hi developers,

I am working on implementing a past study on radiomics. The extract from the published report includes the definition of the GLCM radiomic features as follows in their matlab software:

“Neighbouring voxels were paired in four spatial directions within the 2D axial image planes [(-1, 0), (-1, -1), (0, -1), (1, -1)]. All second-order texture measures were computed based on GLCMs calculated using each of these four directions, and were averaged over all directions.”

I am using pyRadiomics, is there a setting somewhere where I can implement this particular GLCM technique? I used symmetrical GLCM in the past but I am not sure if it resembles this particular way of calculating GLCM.

best
Tenzin

2

Ensure force2D is set to true (default false) and symmetricalGLCM is set to true (default).

Be aware that other preprocessing steps (especially pixel size resampling and discretization methods) can heavily influence the resultant feature values.

1 Like
3

Great! Thank you so much! thats all I need!