8000 GitHub - pejmanS21/CAAS-Net
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

pejmanS21/CAAS-Net

Repository files navigation

CAAS-Net

Fully automated anatomic segmentation for coronary artery angiography

#final evaluation set sample for DeepLabV3 model in RAO Caudal view: alt text

#final evaluation set sample for DeepLabV3 model in LAO Cranial view: alt text

#final evaluation set sample for ResNet50 model in LAO Cranial view: alt text

#final evaluation set sample for ResNet50 model in LAO Cranial view: alt text

#final evaluation set sample for DenseNet121 model in LAO Cranial view: alt text

#final evaluation set sample for DenseNet121 model in LAO Caudal view: alt text

#final evaluation set sample for DenseNet121 model in LAO Straight view: alt text

#final evaluation set sample for Simple U-Net model in LAO Cranial view: alt text

#final evaluation set sample for AttentionEfficientWNet model in LAO Cranial view: alt text

#final evaluation set sample for AttentionDenseWNet model in LAO Cranial view: alt text

These models use dice score to optimize a decoder, which detects important areas and send the results to the multi class decoder as an input. This works as similar to attention and makes the classification easier.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

0