HIGHLIGHTS
SUMMARY
The proposed method can accurately segment both the more prominent structures (lungs and heart) and the smaller structures (clavicles) in chest X-rays.. The experimental results show that the proposed method outperforms the previous SOTA methods on single class anatomical structure segmentation as well as multiclass anatomical structure segmentation. The authors avoided the traditional methods, because the main aim was to compare the proposed model with the deep learning based approaches for lung segmentation. The extensive experimental evaluation affirmed that the proposed dual encoder-decoder CNN architectures produced the best results for anatomical structure . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.