Unsupervised Deformable Image Registration Using Polyphase UNet for 3D Brain MRI Volumes
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Purpose: Image registration is a fundamental task in various medical imaging studies and clinical image analyses, such as comparison of patient data with anatomical structures. In order to solve the problems of conventional image registration approaches, such as long computational time, recent deep-learning supervised and unsupervised methods have been extensively studied because of their excellent performance and fast computational time. In this study, we propose a deep-learningbased network for deformable medical image registration using unsupervised learning. Materials and Methods: In this paper, we solve the image-registration optimization problem by modelling a function using a convolutional neural network with polyphase decomposition to learn the spatial transformable parameters based on the input images and to generate the registration field. A spatial transformer is used to reconstruct the output warped image while imposing smoothness constraints on the registration field. With polyphase decomposition, our proposed method learns more features based on the input image pairs without the need for any ground-truth registration field. Results: Experimental results using 3D T1 brain MRI volume scans and compared with state-of-the-art image-registration methods demonstrated that our method provides better 3D-image registration. Conclusion: Our proposed method uses less computational time in registering unseen pairs of input images during inference and can be applied for other unimodal image registration tasks, and the hyper-parameters can be adjusted for the specific task.
Abstract
Purpose: Image registration is a fundamental task in various medical imaging studies and clinical image analyses, such as comparison of patient data with anatomical structures. In order to solve the problems of conventional image registration approaches, such as long computational time, recent deep-learning supervised and unsupervised methods have been extensively studied because of their excellent performance and fast computational time. In this study, we propose a deep-learningbased network for deformable medical image registration using unsupervised learning. Materials and Methods: In this paper, we solve the image-registration optimization problem by modelling a function using a convolutional neural network with polyphase decomposition to learn the spatial transformable parameters based on the input images and to generate the registration field. A spatial transformer is used to reconstruct the output warped image while imposing smoothness constraints on the registration field. With polyphase decomposition, our proposed method learns more features based on the input image pairs without the need for any ground-truth registration field. Results: Experimental results using 3D T1 brain MRI volume scans and compared with state-of-the-art image-registration methods demonstrated that our method provides better 3D-image registration. Conclusion: Our proposed method uses less computational time in registering unseen pairs of input images during inference and can be applied for other unimodal image registration tasks, and the hyper-parameters can be adjusted for the specific task.
참고문헌 (28)
- Boveiri HR, "Medical image registration using deep neural networks: a comprehensive review", ArXiv 2002.03401, 1~45 (2002)
- Haskins G, "Deep learning in medical image registration : a survey", Mach Vision Appl, 31, pp.8~ (2020)
- Wu G, "Unsupervised deep feature learning for deformable registration of MR brain images", Med Image Comput Comput Assist Interv, 16, pp.649~656 (2013)
- Ghosal S, "Deep deformable registration : enhancing accuracy by fully convolutional neural net", Pattern Recognit Lett, 94, pp.81~86 (2017)
- Blendowski M, "3D-CNNs for deep binary descriptor learning in medical volume data", Informatik Aktuell, pp.23~28 (2018)
- Liu X, "Image synthesis-based multi-modal image registration framework by using deep fully convolutional networks", Med Biol Eng Comput, 57, pp.1037~1048 (2019)
- Cheng X, "Comparable electrode impedance and speech perception at 12months after cochlear implantation using round window versus cochleostomy : an analysis of 40 patients", ORL J Otorhinolaryngol Relat Spec, 80, pp.248~258 (2018)
- Yang X, "Fast predictive image registration", Lect Notes Comput Sc, pp.48~57 (2016)
- Sokooti H, "Nonrigid image registration using multi-scale 3D convolutional neural networks", Lect Notes Comput Sc, pp.232~239 (2017)
- Eppenhof KAJ, "Deformable image registration using convolutional neural networks", Med Imaging 2018: Image Processing, pp.105740S~ (2018)
- Cao X, "Deformable image registration based on similarity-steered CNN regression", Med Image Comput Comput Assist Interv, 10433, pp.300~308 (2017)
- Sun L, "Deformable MRI-ultrasound registration using 3D convolutional neural network", Lect Notes Comput Sc, pp.152~158 (2018)
- Balakrishnan G, "VoxelMorph: a learning framework for deformable medical image registration", IEEE Trans Med Imaging (2019)
- Jaderberg M, "Spatial transformer networks", Advances in Neural Information Processing Systems, pp.2017~2025 (2015)
- Li H, "Non-rigid image registration using selfsupervised fully convolutional networks without training data", IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pp.1075~1078 (2018)
- de Vos BD, "End-to-end unsupervised deformable image registration with a convolutional neural network", Lect Notes Comput Sc, pp.204~212 (2017)
- Dalca AV, "Unsupervised learning for fast probabilistic diffeomorphic registration", Lect Notes Comput Sc, pp.729~738 (2018)
- Yoo I, "ssEMnet : serial-section electron microscopy image registration using a spatial transformer network with learned features", Lect Notes Comput Sc, pp.249~257 (2017)
- Shu C, "An unsupervised network for fast microscopic image registration", Medical Imaging 2018:Digital Pathology, 10581, pp.105811D~ (2018)
- Kearney V, "An unsupervised convolutional neural network-based algorithm for deformable image registration", Phys Med Biol, 63, pp.185017~ (2018)
- Balakrishnan G, "An unsupervised learning model for deformable medical image registration", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.9252~9260 (2018)
- Ronneberger O, "U-net: convolutional networks for biomedical image segmentation", MICCAI 2015: Medical Image Computing and Computer-Assisted Intervention, pp.234~241 (2015)
- Kim B, "Cycle-consistent adversarial network with polyphase U-Nets for liver lesion segmentation", 1st Conf Med Imaging with Deep Learn (MIDL 2018), pp.1~3 (2018)
- Ye JC, "Deep convolutional framelets : a general deep learning framework for inverse problems", SIAM J Imaging Sci, 11, pp.991~1048 (2018)
- Kim B, "Mumford-shah loss functional for image segmentation with deep learning", IEEE T Image Process, 29, pp.1856~1866 (2019)
- LaMontagne PJ, "OASIS-3 : longitudinal neuroimaging, clinical, and cognitive dataset for normal aging and Alzheimer disease", Alzheimer’s Dement, 14, pp.P1097~ (2018)
- Fischl B, "FreeSurfer", Neuroimage, 62, pp.774~781 (2012)
- Avants BB, "Symmetric diffeomorphic image registration with crosscorrelation : evaluating automated labeling of elderly and neurodegenerative brain", Med Image Anal, 12, pp.26~41 (2008)