Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer

Table 5 Foreground segmentation statistics on the TMUH test set

Model	Training dataset	Foreground precision	Foreground recall
CNN-based	KCGMH	0.2673	0.6859
	TMUH	0.5203	0.4898
	Mix	0.3538	0.6123
Transformer-based	KCGMH	0.2896	0.6787
	TMUH	0.5304	0.4561
	Mix	0.3131	0.654

ISSN: 1471-2377