IRUS Total

Attention Gated Networks: Learning to Leverage Salient Regions in Medical Images

File Description SizeFormat 
1808.08114v1.pdfWorking paper5.54 MBAdobe PDFView/Open
Title: Attention Gated Networks: Learning to Leverage Salient Regions in Medical Images
Authors: Schlemper, J
Oktay, O
Schaap, M
Heinrich, M
Kainz, B
Glocker, B
Rueckert, D
Item Type: Working Paper
Abstract: We propose a novel attention gate (AG) model for medical image analysis 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. This enables us to eliminate the necessity of using explicit external tissue/organ localisation modules when using convolutional neural networks (CNNs). AGs can be easily integrated into standard CNN models such as VGG or U-Net architectures with minimal computational overhead while increasing the model sensitivity and prediction accuracy. The proposed AG models are evaluated on a variety of tasks, including medical image classification and segmentation. For classification, we demonstrate the use case of AGs in scan plane detection for fetal ultrasound screening. We show that the proposed attention mechanism can provide efficient object localisation while improving the overall prediction performance by reducing false positives. For segmentation, the proposed architecture is evaluated on two large 3D CT abdominal datasets with manual annotations for multiple organs. Experimental results show that AG models consistently improve the prediction performance of the base architectures across different datasets and training sizes while preserving computational efficiency. Moreover, AGs guide the model activations to be focused around salient regions, which provides better insights into how model predictions are made. The source code for the proposed AG models is publicly available.
Issue Date: 31-Dec-2018
URI: http://hdl.handle.net/10044/1/63021
Copyright Statement: © 2018 The Author(s).
Sponsor/Funder: Engineering & Physical Science Research Council (E
Wellcome Trust
Wellcome Trust/EPSRC
Wellcome Trust
Engineering & Physical Science Research Council (E
Funder's Grant Number: RTJ5557761-1
PO :RTJ5557761-1
Nvidia Hardware donation
Keywords: cs.CV
Notes: Submitted to Medical Image Analysis (Special Issue on Medical Imaging with Deep Learning). arXiv admin note: substantial text overlap with arXiv:1804.03999, arXiv:1804.05338
Appears in Collections:Computing
Faculty of Engineering