Image Segmentation of Triple-Negative Breast Cancer by Incorporating Multiscale and Parallel Attention Mechanisms

Table 3

Results of comparative analysis of different models on public datasets.

Model	Dice	Iou	Acc	Spec	Sens
UNet	0.6535	0.485	0.8701	0.9255	0.8505
Att-UNet	0.6464	0.4775	0.8846	0.9278	0.8703
ResUNet	0.6408	0.4714	0.8631	0.9132	0.8677
SegNet	0.6395	0.4700	0.8883	0.9281	0.8618
MSPAMUNet	0.6852	0.5211	0.8914	0.9409	0.8249

Note: Bolded font is the optimal value for each line.