Attention U-Net: Learning Where to Look for the Pancreas

11 Apr 2018Ozan OktayJo SchlemperLoic Le FolgocMatthew LeeMattias HeinrichKazunari MisawaKensaku MoriSteven McDonaghNils Y HammerlaBernhard KainzBen GlockerDaniel Rueckert

We propose a novel attention gate (AG) model for medical imaging that automatically learns to focus on target structures of varying shapes and sizes. Models trained with AGs implicitly learn to suppress irrelevant regions in an input image while highlighting salient features useful for a specific task... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Pancreas Segmentation CT-150 Att U-Net Dice Score 0.840 # 1
Pancreas Segmentation CT-150 Att U-Net Precision 0.849 # 1
Pancreas Segmentation CT-150 Att U-Net Recall 0.841 # 1