Skip to main content

Table 2 Radiomic features extracted from the SUV and Ki images

From: Radiomics predictive modeling from dual-time-point FDG PET Ki parametric maps: application to chemotherapy response in lymphoma

Category

Features

Gray-level co-occurrence matrix (GLCM)

Homogeneity, Energy, Contrast, Correlation, Entropy, Entropy log2, Dissimilarity

Neighboring gray-level dependence matrix (NGLDM)

Coarseness, Contrast, Busyness

Gray-level run length matrix (GLRLM)

SRE, LRE, LGRE, HGRE, SRLGE, SRHGE, LRLGE, LRHGE, GLNU, RLNU, RP

Gray-level zone length matrix (GLZLM)

SZE, LZE, LGZE, HGZE, SZLGE, SZHGE, LZLGE, LZHGE, GLNU, ZLNU, ZP

Shape

Sphericity, Compacity, Surface, Volume (ml, vx)

First-order features from the histogram

Entropy, Entropy log2, Uniformity, AUC_CSH

Conventional and discretized indices

Q1, Q2, Q3, min, mean, max, Standard Deviation, peak, TLG, Skewness, Kurtosis, Excess Kurtosis