Comparative analysis of proficiencies of various textures and geometric features in breast mass classification using k-nearest neighbor

Visual Computing for Industry, Biomedicine, and Art

Table 10 Rank-wise lists of the top 20 most discriminative textures and the top 20 most discriminative geometric features selected by the Relief-F feature selection algorithm

Rank	Rank wise top 20 texture features			Rank wise top 20 geometry features
Rank	Index	Feature name	Feature model	Index	Feature name	Feature model
1	F42	H1	FRACTAL	F99	FD26	Fourier descriptor
2	F43	H2	FRACTAL	F103	FD30	Fourier descriptor
3	F33	SL	LTEM	F82	FD9	Fourier descriptor
4	F27	Roughness	SFM	F84	FD11	Fourier descriptor
5	F7	Sum variance	SGLCM	F66	ZM3–1	Zernike moments
6	F26	Periodicity	SFM	F67	ZM31	Zernike moments
7	F25	Contrast	SFM	F53	Euler number	Shape descriptor
8	F4	Sum squares	SGLCM	F87	FD14	Fourier descriptor
9	F20	Mean	FOS	F86	FD13	Fourier descriptor
10	F6	Sum average	SGLCM	F97	FD24	Fourier descriptor
11	F24	Coarseness	SFM	F70	ZM4–2	Zernike moments
12	F32	EL	LTEM	F72	ZM42	Zernike moments
13	F16	Contrast	GLDS	F74	FD1	Fourier descriptor
14	F15	Homogeneity	GLDS	F92	FD19	Fourier descriptor
15	F19	Entropy	GLDS	F55	Solidity	Shape descriptor
16	F17	Mean	GLDS	F56	Extent	Shape descriptor
17	F37	WE	LTEM	F100	FD27	Fourier descriptor
18	F22	Skewness	FOS	F119	FD46	Fourier descriptor
19	F23	Kurtosis	FOS	F115	FD42	Fourier descriptor
20	F9	Entropy	SGLCM	F104	FD31	Fourier descriptor