Table 7. Performance of different conditional embedding methods on the BraTS 2020 dataset.

Methods	WT			TC			ET			Average
	DSC↑	HD95↓ (mm)		DSC↑	HD95↓ (mm)		DSC↑	HD95↓ (mm)		DSC↑	HD95↓ (mm)
Conditional fusion	0.912	3.190		0.840	5.611		0.780	4.014		0.844	4.272
Feature encoder	0.915*	1.866*		0.847	3.018		0.783*	2.800*		0.848	2.561
Conditional embedding	0.910	2.040		0.850	2.816		0.780	2.772		0.847	2.543
FCFDiff-Net	0.914	1.910		0.854*	2.727*		0.782	2.912		0.850*	2.516*

Higher DSC scores (↑) indicate better segmentation, while lower HD95 values (↓) indicate better performance. The top result is marked with asterisk (*). BraTS, brain tumor segmentation; DSC, Dice similarity coefficient; ET, enhancing tumor; FCFDiff-Net, full-conditional feature diffusion embedded network; HD95, Hausdorff distance at the 95th percentile; TC, tumor core; WT, whole tumor.