Skip to main content

Table 6 The ten top-ranked classification features according to Gini importance and minimum-redundancy-maximum-relevance

From: Automatic segmentation and classification of breast lesions through identification of informative multiparametric PET/MRI features

Rank Manual annotation Automatic segmentation
Gini importance Minimum-redundancy-maximum-relevance Gini importance Minimum-redundancy-maximum-relevance
1 fl-tex-post/peak entropy fl-tex-post/peak energy f l-kinetic T 1/2 fl-tex-post/peak energy
2 fl-tex-post/peak energy fl-δckc # 11 fl-tex-post/peak energy fl-δckc #10
3 f l-kinetic T 1/2 fl-δckc # 8 fl-kinetic MDER fl-tex-post/pre sum average
4 fl-ckc #11 fl-kinetic MDER f l-adc fl-tex-post/pre homogeneity
5 fl-ckc #9 fl-tex-post/pre homogeneity fl-δckc #18 fl-δckc #20
6 fl-tex-post information measure 1 fl-tex-post/peak correlation fl-ckc #11 fl-tex-post/peak information measure 2
7 fl-δckc #14 fl-tex-post/peak entropy fl-δckc #21 fl-δckc #21
8 fl-tex-post information measure 2 f l-kinetic T 1/2 fl-δckc #14 f l-kinetic β
9 fl-δckc #11 fl-tex-post/pre dva fl-tex-post/pre dva fl-tex-peak/pre homogeneity
10 fl-tex-peak entropy fl-tex-post/pre energy fl-δckc #10 fl-tex-post sum variance
  1. dva difference variance, MDER maximum derivative of kinetic regression function