Abstract | ||
---|---|---|
•Propose an ultrasonic breast tumor segmentation network combining antagonism mechanism and contour optimization.•Using active contour modules to highlight ultrasound tumor edges, forming a bi-directional optimization with a Deformed U-Net.•Antagonism mechanism is adopted to improve the precision of segmentation network.•The loss of discriminator, Deformed U-net and active contour module is used for collaborative optimization of the whole network.•The result of this method is better than that of other segmentation networks of breast lesions. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1016/j.cmpb.2022.107052 | Computer Methods and Programs in Biomedicine |
Keywords | DocType | Volume |
Segmentation of tumor lesions,Conditional adversarial network,Deformed U-Net,Active contour module,Markov discriminator | Journal | 225 |
ISSN | Citations | PageRank |
0169-2607 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jinhong Wang | 1 | 0 | 0.34 |
Guiqing Chen | 2 | 0 | 0.34 |
Shiqiang Chen | 3 | 0 | 0.34 |
Alex Noel Joseph Raj | 4 | 0 | 0.34 |
Zhemin Zhuang | 5 | 0 | 0.34 |
Lei Xie | 6 | 0 | 0.34 |
Shuhua Ma | 7 | 0 | 0.34 |