Literature DB >> 33484431

Performance-aware programming for intraoperative intensity-based image registration on graphics processing units.

Martin C W Leong1, Kit-Hang Lee1, Bowen P Y Kwan2, Yui-Lun Ng1, Zhiyu Liu1, Nassir Navab3, Wayne Luk2, Ka-Wai Kwok4.   

Abstract

PURPOSE: Intensity-based image registration has been proven essential in many applications accredited to its unparalleled ability to resolve image misalignments. However, long registration time for image realignment prohibits its use in intra-operative navigation systems. There has been much work on accelerating the registration process by improving the algorithm's robustness, but the innate computation required by the registration algorithm has been unresolved.
METHODS: Intensity-based registration methods involve operations with high arithmetic load and memory access demand, which supposes to be reduced by graphics processing units (GPUs). Although GPUs are widespread and affordable, there is a lack of open-source GPU implementations optimized for non-rigid image registration. This paper demonstrates performance-aware programming techniques, which involves systematic exploitation of GPU features, by implementing the diffeomorphic log-demons algorithm.
RESULTS: By resolving the pinpointed computation bottlenecks on GPU, our implementation of diffeomorphic log-demons on Nvidia GTX Titan X GPU has achieved ~ 95 times speed-up compared to the CPU and registered a 1.3-M voxel image in 286 ms. Even for large 37-M voxel images, our implementation is able to register in 8.56 s, which attained ~ 258 times speed-up. Our solution involves effective employment of GPU computation units, memory, and data bandwidth to resolve computation bottlenecks.
CONCLUSION: The computation bottlenecks in diffeomorphic log-demons are pinpointed, analyzed, and resolved using various GPU performance-aware programming techniques. The proposed fast computation on basic image operations not only enhances the computation of diffeomorphic log-demons, but is also potentially extended to speed up many other intensity-based approaches. Our implementation is open-source on GitHub at https://bit.ly/2PYZxQz .

Entities:  

Keywords:  Demons algorithm; Image-guided treatment; Non-rigid registration; Parallel computing; Surgical guidance

Mesh:

Year:  2021        PMID: 33484431      PMCID: PMC7946684          DOI: 10.1007/s11548-020-02303-y

Source DB:  PubMed          Journal:  Int J Comput Assist Radiol Surg        ISSN: 1861-6410            Impact factor:   2.924


  22 in total

1.  Engineering and algorithm design for an image processing Api: a technical report on ITK--the Insight Toolkit.

Authors:  Terry S Yoo; Michael J Ackerman; William E Lorensen; Will Schroeder; Vikram Chalana; Stephen Aylward; Dimitris Metaxas; Ross Whitaker
Journal:  Stud Health Technol Inform       Date:  2002

2.  A Demons algorithm for image registration with locally adaptive regularization.

Authors:  Nathan D Cahill; J Alison Noble; David J Hawkes
Journal:  Med Image Comput Comput Assist Interv       Date:  2009

3.  3D Slicer as an image computing platform for the Quantitative Imaging Network.

Authors:  Andriy Fedorov; Reinhard Beichel; Jayashree Kalpathy-Cramer; Julien Finet; Jean-Christophe Fillion-Robin; Sonia Pujol; Christian Bauer; Dominique Jennings; Fiona Fennessy; Milan Sonka; John Buatti; Stephen Aylward; James V Miller; Steve Pieper; Ron Kikinis
Journal:  Magn Reson Imaging       Date:  2012-07-06       Impact factor: 2.546

4.  A fast diffeomorphic image registration algorithm.

Authors:  John Ashburner
Journal:  Neuroimage       Date:  2007-07-18       Impact factor: 6.556

5.  Diffeomorphic demons: efficient non-parametric image registration.

Authors:  Tom Vercauteren; Xavier Pennec; Aymeric Perchant; Nicholas Ayache
Journal:  Neuroimage       Date:  2008-11-07       Impact factor: 6.556

6.  Symmetric log-domain diffeomorphic Registration: a demons-based approach.

Authors:  Tom Vercauteren; Xavier Pennec; Aymeric Perchant; Nicholas Ayache
Journal:  Med Image Comput Comput Assist Interv       Date:  2008

Review 7.  Review of intraoperative imaging and planning techniques in permanent seed prostate brachytherapy.

Authors:  Alfredo Polo; Carl Salembier; Jack Venselaar; Peter Hoskin
Journal:  Radiother Oncol       Date:  2010-01-13       Impact factor: 6.280

8.  elastix: a toolbox for intensity-based medical image registration.

Authors:  Stefan Klein; Marius Staring; Keelin Murphy; Max A Viergever; Josien P W Pluim
Journal:  IEEE Trans Med Imaging       Date:  2009-11-17       Impact factor: 10.048

9.  Multiresolution elastic image registration.

Authors:  P J Kostelec; J B Weaver; D M Healy
Journal:  Med Phys       Date:  1998-09       Impact factor: 4.071

10.  Deep Adaptive Log-Demons: Diffeomorphic Image Registration with Very Large Deformations.

Authors:  Liya Zhao; Kebin Jia
Journal:  Comput Math Methods Med       Date:  2015-05-18       Impact factor: 2.238

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.