Real-time multimodal image registration with partial intraoperative point-set data

被引:23
作者
Baum, Zachary M. C. [1 ,2 ]
Hu, Yipeng [1 ,2 ]
Barratt, Dean C. [1 ,2 ]
机构
[1] UCL, Ctr Med Image Comp, London, England
[2] UCL, Wellcome EPSRC Ctr Intervent & Surg Sci, London, England
基金
英国惠康基金;
关键词
medical image registration; point-set registration; image-guided interventions; prostate cancer; MR; ROBUST; FUSION; OPTIMIZATION; ALGORITHM; HAMMER;
D O I
10.1016/j.media.2021.102231
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present Free Point Transformer (FPT) - a deep neural network architecture for non-rigid point-set registration. Consisting of two modules, a global feature extraction module and a point transformation module, FPT does not assume explicit constraints based on point vicinity, thereby overcoming a com-mon requirement of previous learning-based point-set registration methods. FPT is designed to accept unordered and unstructured point-sets with a variable number of points and uses a "model-free" ap-proach without heuristic constraints. Training FPT is flexible and involves minimizing an intuitive unsu-pervised loss function, but supervised, semi-supervised, and partially-or weakly-supervised training are also supported. This flexibility makes FPT amenable to multimodal image registration problems where the ground-truth deformations are difficult or impossible to measure. In this paper, we demonstrate the application of FPT to non-rigid registration of prostate magnetic resonance (MR) imaging and sparsely-sampled transrectal ultrasound (TRUS) images. The registration errors were 4.71 mm and 4.81 mm for complete TRUS imaging and sparsely-sampled TRUS imaging, respectively. The results indicate superior accuracy to the alternative rigid and non-rigid registration algorithms tested and substantially lower com-putation time. The rapid inference possible with FPT makes it particularly suitable for applications where real-time registration is beneficial. (c) 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ )
引用
收藏
页数:14
相关论文
共 71 条
[1]  
Abadi Martin, 2016, Proceedings of OSDI '16: 12th USENIX Symposium on Operating Systems Design and Implementation. OSDI '16, P265
[2]  
ang Y., Advances in Neural Inf. Processing Systems, V32, P8814
[3]   PointNetLK: Robust & Efficient Point Cloud Registration using PointNet [J].
Aoki, Yasuhiro ;
Goforth, Hunter ;
Srivatsan, Rangaprasad Arun ;
Lucey, Simon .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :7156-7165
[4]   Multimodality Biomedical Image Registration Using Free Point Transformer Networks [J].
Baum, Zachary M. C. ;
Hu, Yipeng ;
Barratt, Dean C. .
MEDICAL ULTRASOUND, AND PRETERM, PERINATAL AND PAEDIATRIC IMAGE ANALYSIS, ASMUS 2020, PIPPI 2020, 2020, 12437 :116-125
[5]   A METHOD FOR REGISTRATION OF 3-D SHAPES [J].
BESL, PJ ;
MCKAY, ND .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1992, 14 (02) :239-256
[6]  
Cao XG, 2016, 2016 INTERNATIONAL CONFERENCE ON DIGITAL IMAGE COMPUTING: TECHNIQUES AND APPLICATIONS (DICTA), P751, DOI 10.1007/978-3-319-46726-9_1
[7]  
Chollet F., 2015, Keras
[8]   A new point matching algorithm for non-rigid registration [J].
Chui, HL ;
Rangarajan, A .
COMPUTER VISION AND IMAGE UNDERSTANDING, 2003, 89 (2-3) :114-141
[9]   Robust 2-D-3-D Registration Optimization for Motion Compensation During 3-D TRUS-Guided Biopsy Using Learned Prostate Motion Data [J].
De Silva, Tharindu ;
Cool, Derek W. ;
Yuan, Jing ;
Romagnoli, Cesare ;
Samarabandu, Jagath ;
Fenster, Aaron ;
Ward, Aaron D. .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2017, 36 (10) :2010-2020
[10]   2D-3D rigid registration to compensate for prostate motion during 3D TRUS-guided biopsy [J].
De Silva, Tharindu ;
Fenster, Aaron ;
Cool, Derek W. ;
Gardi, Lori ;
Romagnoli, Cesare ;
Samarabandu, Jagath ;
Ward, Aaron D. .
MEDICAL PHYSICS, 2013, 40 (02)