Skip to main content

Table 2 Results of group-wise radiomic feature-based cancer classifications for hormone receptor status (training dataset)

From: Radiomic signatures with contrast-enhanced magnetic resonance imaging for the assessment of breast cancer receptor status and molecular subtypes: initial results

 

HR positive

HER2 positive

HR negative

HER2 negative

Her2-enriched

TN

All others

HR positive

–

–

67%

(POE; 9)

–

79.4% (Fisher; 7)

71.3%

(Fisher; 8)

67%

(POE; 9)

HER2 positive

–

–

–

73.6%

(Fisher; 9)

–

59.5%

(Fisher; 4)

73.6%

(Fisher; 9)

HR negative

67%

(POE; 9)

–

–

–

–

–

67%

(POE; 9)

HER2 negative

–

73.6%

(Fisher; 9)

–

–

–

–

73.6%

(Fisher; 9)

HER2-enriched

79.4%

(Fisher; 7)

–

–

–

–

–

–

TN

71.3%

(Fisher; 8)

59.5%

(Fisher; 4)

–

–

–

–

–

All others

67%

(POE; 9)

73.6%

(Fisher; 9)

67%

(POE; 9)

73.6%

(Fisher; 9)

–

–

–

  1. HER2 human epidermal growth factor receptor 2, HR hormone receptor, MI mutual information, POE probability of error and average correlation, TN triple negative. The feature selection algorithm and number of features used for classification are given in parentheses