Beyond breast density: a review on the advancing role of parenchymal texture analysis in breast cancer risk assessment

Table 2 Parenchymal texture descriptors for breast cancer risk assessment; texture descriptors which have been examined in association with breast cancer risk, classified to five feature groups

Grey-level histogram features [51, 53–56, 60, 66, 71, 72, 77, 80, 81, 84, 85]
min intensity	skewness	5^th percentile	energy
max intensity	kurtosis	5^th percentile mean	root mean square variation
standard deviation	entropy	95^th percentile
mean intensity	sum intensity	95^th percentile mean
Co-occurrence features [51, 53–56, 66, 71, 77, 78, 80, 81, 84, 85]
cluster shade	entropy	inverse difference moment	difference entropy
correlation	inertia	sum variance	homogeneity
Haralick correlation	difference moment	sum average	product moment
energy	coarseness	difference variance	triangular symmetry
Run-length measures [51, 56, 66, 73, 76–78]
long run emphasis	gray-level non-uniformity	high gray level run emphasis	run percentage
short run emphasis	run-length non-uniformity	low gray level run emphasis	number of runs
Structural/Pattern measures [51, 54, 56, 60, 61, 64, 66, 72, 74, 77, 78, 81, 82, 84, 85, 89]
fractal dimension	local binary pattern	Hessian matrix	Weber local descriptors
lacunarity	Law’s masks	edge enhancing index	directional gradient
Multi-resolution/Spectral features [53–56, 61, 64, 66, 71, 75, 78, 80, 81, 83–85, 89]
Fourier power spectrum	wavelet/Gabor	Gaussian Kernels	power-law spectrum

ISSN: 1465-542X