Altmetric

Nesterov accelerated ADMM for fast diffeomorphic image registration

File Description SizeFormat 
2109.12688v1.pdfFile embargoed until 20 September 20224.48 MBAdobe PDF    Request a copy
Title: Nesterov accelerated ADMM for fast diffeomorphic image registration
Authors: Thorley, A
Jia, X
Chang, HJ
Liu, B
Bunting, K
Stoll, V
De Marvao, A
O'Regan, DP
Gkoutos, G
Kotecha, D
Duan, J
Item Type: Conference Paper
Abstract: Deterministic approaches using iterative optimisation have been historically successful in diffeomorphic image registration (DiffIR). Although these approaches are highly accurate, they typically carry a significant computational burden. Recent developments in stochastic approaches based on deep learning have achieved sub-second runtimes for DiffIR with competitive registration accuracy, offering a fast alternative to conventional iterative methods. In this paper, we attempt to reduce this difference in speed whilst retaining the performance advantage of iterative approaches in DiffIR. We first propose a simple iterative scheme that functionally composes intermediate non-stationary velocity fields to handle large deformations in images whilst guaranteeing diffeomorphisms in the resultant deformation. We then propose a convex optimisation model that uses a regularisation term of arbitrary order to impose smoothness on these velocity fields and solve this model with a fast algorithm that combines Nesterov gradient descent and the alternating direction method of multipliers (ADMM). Finally, we leverage the computational power of GPU to implement this accelerated ADMM solver on a 3D cardiac MRI dataset, further reducing runtime to less than 2 s. In addition to producing strictly diffeomorphic deformations, our methods outperform both state-of-the-art deep learning-based and iterative DiffIR approaches in terms of dice and Hausdorff scores, with speed approaching the inference time of deep learning-based methods.
Editors: DeBruijne, M
Cattin, PC
Cotin, S
Padoy, N
Speidel, S
Zheng, Y
Essert, C
Issue Date: 21-Sep-2021
Date of Acceptance: 1-Sep-2021
URI: http://hdl.handle.net/10044/1/93260
DOI: 10.1007/978-3-030-87202-1_15
ISBN: 978-3-030-87201-4
ISSN: 0302-9743
Publisher: SPRINGER INTERNATIONAL PUBLISHING AG
Start Page: 150
End Page: 160
Journal / Book Title: MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2021, PT IVV
Volume: 12904
Copyright Statement: © Springer Nature Switzerland AG 2021. The final publication is available at Springer via https://doi.org/10.1007/978-3-030-87202-1_15
Sponsor/Funder: The Academy of Medical Sciences
Imperial College Healthcare NHS Trust- BRC Funding
Imperial College Healthcare NHS Trust- BRC Funding
British Heart Foundation
Funder's Grant Number: SGL015/1006
RDC04
RDB02
RG/19/6/34387
Conference Name: International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)
Keywords: Science & Technology
Technology
Life Sciences & Biomedicine
Computer Science, Artificial Intelligence
Computer Science, Software Engineering
Engineering, Biomedical
Imaging Science & Photographic Technology
Radiology, Nuclear Medicine & Medical Imaging
Surgery
Computer Science
Engineering
Image registration
Diffeomorphism
ADMM
FRAMEWORK
Science & Technology
Technology
Life Sciences & Biomedicine
Computer Science, Artificial Intelligence
Computer Science, Software Engineering
Engineering, Biomedical
Imaging Science & Photographic Technology
Radiology, Nuclear Medicine & Medical Imaging
Surgery
Computer Science
Engineering
Image registration
Diffeomorphism
ADMM
FRAMEWORK
cs.CV
cs.CV
math.DS
Artificial Intelligence & Image Processing
Publication Status: Published
Start Date: 2021-09-27
Finish Date: 2021-10-01
Conference Place: ELECTR NETWORK
Embargo Date: 2022-09-20
Online Publication Date: 2021-09-21
Appears in Collections:Institute of Clinical Sciences