Skip to main content
. 2022 Feb 22;12:802234. doi: 10.3389/fonc.2022.802234

Table 8.

Delong Test between each two models of LR-classifier in the training and validation cohorts.

Radiomic model T1WI T2WI CE-T1WI T1WI+T2WI T1WI+CE-T1WI T2WI+CE-T1WI T1WI+T2WI+CE-T1WI
T1WI 0.0063* 0.0725 0.0117* 0.4114 0.0063* 0.0110*
T2WI 0.1391 0.2395 0.6474 0.0318* 1.0000 0.6744
CE-T1WI 0.1048 0.7088 0.4144 0.1713 0.2395 0.3033
T1WI+T2WI 0.0695 0.5149 0.4838 0.0613 0.6474 0.8620
T1WI+CE-T1WI 0.6829 0.2054 0.2238 0.0771 0.0318* 0.0545
T2WI+CE-T1WI 0.1391 1.0000 0.7088 0.5149 0.2054 0.6744
T1WI+T2WI+
CE-T1WI
0.0465* 0.2542 0.3188 0.6750 0.0720 0.2542
Training cohort Validation cohort

T1WI, T1-weighted imaging; T2WI, T2-weighted imaging; CE, contrast-enhanced. *p < 0.05.