Skip to main content

Table 3 Calculated image region-of-interest features, as defined in the Supplementary Data of Aerts et al. [26]

From: “Real-world” radiomics from multi-vendor MRI: an original retrospective study on the prediction of nodal status and disease survival in breast cancer, as an exemplar to promote discussion of the wider issues

Shape features

 Compactness 1

Spherical disproportion

 Compactness 2

Surface area

 Maximum diameter

Surface-to-volume ratio

 Sphericity

Volume

Statistical features

 Energy

Minimum

 Entropy

Range

 Kurtosis

RMS

 Maximum

Skewness

 Mean

Standard deviation

 Mean absolute deviation

Variance

 Median

Uniformity

Texture features

 Autocorrelation

Inverse difference moment normalised

 Cluster prominence

Inverse difference normalised

 Cluster shade

Inverse variance

 Cluster tendency

Long run emphasis

 Contrast

Long run low grey level emphasis

 Correlation

Low grey level run emphasis

 Difference entropy

Maximum probability

 Dissimilarity

Run length non-uniformity

 Energy

Run percentage

 Entropy

Short run emphasis

 Grey level nonuniformity

Short run low grey level emphasis

 High grey level run emphasis

Short run high grey level emphasis

 Homogeneity 1

Sum average

 Homogeneity 2

Sum entropy

 Informational measure correlation 1

Sum variance

 Informational measure correlation 2

Short run emphasis