Skip to main content

Table 2 Descriptors retained for the creation of the model to predict malignant or benign masses

From: Radiomics model to classify mammary masses using breast DCE-MRI compared to the BI-RADS classification performance

Type of dynamic MRI card

Shape features

First-order features

Second-order features

Importance coefficient

DCE1

  

Inverse difference normalized

97.9

  

Inverse difference moment normalized

62.3

DCE2

Maximum 3D diameter

  

18.3

  

Inverse difference normalized

91.9

  

Inverse difference moment normalized

100.0

DCE4

Maximum 3D diameter

  

13.1

DCE5

  

Informal measure of correlation 1

46.3

AUC

  

Low gray level run emphasis

67.2

  

Long run low gray level run emphasis

59.4

  

Short run low gray level run emphasis

53.5

  

Inverse difference moment normalized

40.7

  

Long run high gray level run emphasis

33.9

PE

  

Long run high gray level emphasis

32.2

SER

 

Total energy

 

13.3

 

Energy

 

4.4

WO

Major axis

  

7.2

Maximum 3D diameter

  

20.6

  1. AUC Area under curve; DCE Dynamic contrast enhancement; PE Peak enhancement; SER Signal enhancement ratio; WO Washout