Skip to main content

Table 4 ROC analysis of radiomic features in prediction of immunohistochemical status

From: Prediction of the clinicopathological subtypes of breast cancer using a fisher discriminant analysis model based on radiomic features of diffusion-weighted MRI

IHC status

Radiomic feature

AUROC

ER(+) VS. ER(−)

Histogram

0.973 (0.949–0.997)

Texture

0.762 (0.674–0.851)

GLCM

0.963 (0.929–0.998)

RLM

0.967 (0.937–0.997)

PR(+) VS. PR(−)

Histogram

0.925 (0.879–0.972)

Texture

0.731 (0.637–0.824)

GLCM

0.939 (0.892–0.986)

RLM

0.923 (0.875–0.971)

HER2(+) VS. HER2(−)

Histogram

0.902 (0.847–0.957)

Texture

0.722 (0.627–0.818)

GLCM

0.911 (0.860–0.962)

RLM

0.974 (0.944–1.000)

Ki-67 low VS. high

Histogram

0.926 (0.870–0.981)

Texture

0.718 (0.615–0.820)

GLCM

0.975 (0.949–1.000)

RLM

0.946 (0.905–0.988)