Skip to main content
. 2015 Apr 13;10(4):e0123044. doi: 10.1371/journal.pone.0123044

Table 2. Results of the logistic regression differentiating high-grade from low-grade gliomas.

AUC HGG/LGG [a.u.] p HGG/LGG
rCBV enhancing tumor 0.776 0.015*
edema 0.565 0.376
enhancing tumor & edema 0.718 0.034*
rCTH enhancing tumor 0.775 0.014*
edema 0.811 0.003*
enhancing tumor & edema 0.807 0.006*
COV enhancing tumor 0.739 0.029*
edema 0.650 0.219
enhancing tumor & edema 0.693 0.070
rCBV & rCTH enhancing tumor 0.840 0.028* (rCBV), 0.023* (rCTH)
rCBV & rCTH enhancing tumor & edema 0.876 0.027* (rCBV), 0.006* (rCTH)
rCBV & rCTH & COV enhancing tumor 0.856 0.076 (rCBV), 0.038* (rCTH), 0.437 (COV)
rCBV & rCTH & COV enhancing tumor & edema 0.875 0.032* (rCBV), 0.010* (rCTH), 0.358 (COV)

The area under the curve (AUC) as an accuracy measure for discrimination between high-grade and low-grade gliomas. The AUC is calculated for the three parameters and their combinations in the enhancing tumor, the peri-focal edema, and a combined region of enhancing tumor & peri-focal edema. A p-value ≦ 0.05 (marked with *) allows the conclusion that the predictor has enough strength to distinguish between the two groups.