Evaluating reinforcement learning agents for anatomical landmark detection

File Description SizeFormat 
MIDL2018_journal_submission.pdfFile embargoed until 14 February 2020792.79 kBAdobe PDF    Request a copy
Title: Evaluating reinforcement learning agents for anatomical landmark detection
Authors: Alansary, A
Oktay, O
Li, Y
Folgoc, LL
Hou, B
Vaillant, G
Kamnitsas, K
Vlontzos, A
Glocker, B
Kainz, B
Rueckert, D
Item Type: Journal Article
Abstract: Automatic detection of anatomical landmarks is an important step for a wide range of applications in medical image analysis. Manual annotation of landmarks is a tedious task and prone to observer errors. In this paper, we evaluate novel deep reinforcement learning (RL) strategies to train agents that can precisely and robustly localize target landmarks in medical scans. An artificial RL agent learns to identify the optimal path to the landmark by interacting with an environment, in our case 3D images. Furthermore, we investigate the use of fixed- and multi-scale search strategies with novel hierarchical action steps in a coarse-to-fine manner. Several deep Q-network (DQN) architectures are evaluated for detecting multiple landmarks using three different medical imaging datasets: fetal head ultrasound (US), adult brain and cardiac magnetic resonance imaging (MRI). The performance of our agents surpasses state-of-the-art supervised and RL methods. Our experiments also show that multi-scale search strategies perform significantly better than fixed-scale agents in images with large field of view and noisy background such as in cardiac MRI. Moreover, the novel hierarchical steps can significantly speed up the searching process by a factor of 4-5 times.
Issue Date: 1-Apr-2019
Date of Acceptance: 12-Feb-2019
URI: http://hdl.handle.net/10044/1/67534
DOI: https://dx.doi.org/10.1016/j.media.2019.02.007
ISSN: 1361-8415
Publisher: Elsevier
Start Page: 156
End Page: 164
Journal / Book Title: Medical Image Analysis
Volume: 53
Copyright Statement: © 2019 Elsevier B.V. All rights reserved. This manuscript is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Licence http://creativecommons.org/licenses/by-nc-nd/4.0/
Sponsor/Funder: Engineering & Physical Science Research Council (E
Wellcome Trust
Wellcome Trust/EPSRC
Engineering & Physical Science Research Council (EPSRC)
Wellcome Trust
Engineering & Physical Science Research Council (E
Nvidia
Engineering and Physical Sciences Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
Innovate UK
Funder's Grant Number: RTJ5557761-1
PO :RTJ5557761-1
NS/A000025/1
EP/P001009/1
RTJ5557761
RTJ5557761-1
Nvidia Hardware donation
EP/S013687/1
EP/S013687/1
COVIP_P76965
Keywords: Automatic landmark detection
DQN
Deep learning
Reinforcement learning
09 Engineering
11 Medical And Health Sciences
Nuclear Medicine & Medical Imaging
Publication Status: Published
Conference Place: Netherlands
Embargo Date: 2020-02-14
Online Publication Date: 2019-02-14
Appears in Collections:Faculty of Engineering
Computing



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons