Table 3 Features showing high reproducibility from two experiments
| Â | Category | Parameter | Description / Interpretation |
|---|---|---|---|
Experiment 1 | Histogram-based features | Maximum | Measures maximum intensity value of histogram |
Minimum | Measures minimum intensity value of histogram | ||
Shape-based features | Maximum 3d diameter | Measures maximum 3D ROI diameter as the largest pairwise Euclidean distance between surface voxels of the ROI | |
Spherical disproportion | Ratio of the surface area of the ROI to the surface area of a sphere with the same volume as the ROI | ||
Texture-based features (GLCM) | Custer tendency | Measures homogeneity of GLCM | |
Dissimilarity | Measures differences of entries in GLCM | ||
Entropy | Measures irregularity of GLCM | ||
Filter-based feature | Log Skewness (σ = 1) | Measurement of skewness of ROI image processed by log filter | |
Fractal-based feature | Lacunarity | Measure of the texture or distribution of gaps within an image | |
Experiment 2 | Histogram-based features | Maximum | Same as experiment 1 |
Minimum | Same as experiment 1 | ||
Entropy | Measures irregularity of histogram | ||
Texture-based features | Difference entropy | Measures entropy of processed GLCM matrix Px-y | |
Homogeneity | Measures closeness of GLCM |