Skip to main content
. 2020 Jul 18;10(7):492. doi: 10.3390/diagnostics10070492

Table 3.

Detailed results of group-wise radiomic feature-based cancer: diagnostic accuracies for radiomic features based on cancer invasiveness, hormone subtype, and grading for contrast-enhanced mammography (CEM) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).

Imaging Modality Cancer Subtype
Invasive Cancers HR Positive a Low Grade
CEM Non-Invasive Cancers 6 best: 92% (COM/Fisher) 90% (COM/POE) 88% (COM/MI) - -
HR negative b - 7 best: 91.3% (WAV-RUN/Fisher) 95.6% (RUN/POE) 86.8% (COM/MI) -
High grade - - 9 best: 75.6% (WAV + RUN + COM/Fisher) 77.8% (RUN/POE) 64.4% (WAV + COM/MI)
MRI Non-Invasive Cancers 6 best: 90% (COM/Fisher) 88% (COM/POE) 88% (COM/MI) - -
HR negative b - 6 best: 76.1% (COM/Fisher) 80.4% (COM/POE) 82.6% (COM/MI) -
High grade - - 6 best: 77.8% (RUN/Fisher) 71.1% (COM/POE) 73.3% (COM/MI)

Notes: a HR positive/HER2 negative or HR positive/HER2 positive. b HR negative/HER2 positive or HR negative/HER2 negative (triple negative). Abbreviations: MI, mutual information; POE, Probability of error.