Joint Progressive and Coarse-to-Fine Registration of Brain MRI via Deformation Field Integration and Non-Rigid Feature Fusion

Lv, Jinxin; Wang, Zhiwei; Shi, Hongkuan; Zhang, Haobo; Wang, Sheng; Wang, Yilang; Li, Qiang

doi:10.1109/tmi.2022.3170879

Cited by 32 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While frequent retraining for different values of λ is impractical, recent advances in hyper-networks could enable training of a single λ -adaptive network, such that the user can specify the value of this parameter for each scan at test time 63 . Exploring this direction, along with other architectural improvements that may have a positive impact on EasyReg (e.g., progressive deformations 64 ), remains as future work.…”

Section: Discussionmentioning

confidence: 99%

EasyReg: A ready-to-use deep learning tool for symmetric affine and nonlinear brain MRI registration

Iglesias

2023

Preprint

View full text Add to dashboard Cite

Volumetric registration of brain MRI is routinely used in human neuroimaging, e.g., to align different MRI modalities, to measure change in longitudinal analysis, to map an individual to a template, or in registration-based segmentation. Classical registration techniques based on numerical optimization have been very successful in this domain, and are implemented in widespread software suites like ANTs, Elastix, NiftyReg, or DARTEL. Over the last 7-8 years, learning-based techniques have emerged, which have a number of advantages like high computational efficiency, potential for higher accuracy, easy integration of supervision, and the ability to be part of a meta-architectures. However, their adoption in neuroimaging pipelines has so far been almost inexistent. Reasons include: lack of robustness to changes in MRI modality and resolution; lack of robust affine registration modules; lack of (guaranteed) symmetry; and, at a more practical level, the requirement of deep learning expertise that may be lacking at neuroimaging research sites. Here, we present EasyReg, an open-source, learning-based registration tool that can be easily used from the command line without any deep learning expertise or specific hardware. EasyReg combines the features of classical registration tools, the capabilities of modern deep learning methods, and the robustness to changes in MRI modality and resolution provided by our recent work in domain randomization. As a result, EasyReg is: fast; symmetric; diffeomorphic (and thus invertible); agnostic to MRI modality and resolution; compatible with affine and nonlinear registration; and does not require any preprocessing or parameter tuning. We present results on challenging registration tasks, showing that EasyReg is as accurate as classical methods when registering 1mm isotropic scans within MRI modality, but much more accurate across modalities and resolutions. EasyReg is publicly available as part of FreeSurfer; see https://surfer.nmr.mgh.harvard.edu/fswiki/EasyReg.

show abstract

Section: Discussionmentioning

confidence: 99%

EasyReg: A ready-to-use deep learning tool for symmetric affine and nonlinear brain MRI registration

Iglesias

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…MIND-SSC features are used for the similarity term. https:// grand-challenge.org/algorithms/corrfield/ Driver : [43] uses a dual-encoder UNet backbone with separated multi-scale feature extractors that comprises Deformation Field Integration (DFI) and non-rigid feature fusion (NFF) modules. It produces multi-scale sub-fields that progressively align fixed and moving features.…”

Section: Challenge Entriesmentioning

confidence: 99%

Learn2Reg: Comprehensive Multi-Task Medical Image Registration Challenge, Dataset and Evaluation in the Era of Deep Learning

Hering

Hansen²,

Mok

et al. 2023

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Image registration is a fundamental medical image analysis task, and a wide variety of approaches have been proposed. However, only a few studies have comprehensively compared medical image registration approaches on a wide range of clinically relevant tasks. This limits the development of registration methods, the adoption of research advances into practice, and a fair benchmark across competing approaches. The Learn2Reg challenge addresses these limitations by providing a multitask medical image registration data set for comprehensive characterisation of deformable registration algorithms. A continuous evaluation will be possible at https:// learn2reg.grand-challenge.org. Learn2Reg covers a wide range of anatomies (brain, abdomen, and thorax), modalities (ultrasound, CT, MR), availability of annotations, as well as intra-and inter-patient registration evaluation. We established an easily accessible framework for training and validation of 3D registration methods, which enabled the compilation of results of over 65 individual method submissions from more than 20 unique teams. We used a complementary set of metrics, including robustness, accuracy, plausibility, and runtime, enabling unique insight into the current state-of-the-art of medical image registration. This paper describes datasets, tasks, evaluation methods and results of the challenge, as well as results of further analysis of transferability to new datasets, the importance of label supervision, and resulting bias. While no single approach worked best across all tasks, many methodological aspects could be identified that push the performance of medical image registration to new stateof-the-art performance. Furthermore, we demystified the common belief that conventional registration methods have to be much slower than deep-learning-based methods.

show abstract

“…However, these methods ignored that it is difficult to accurately align the given two images at once, especially in the brain image with complex tissue structure. To address this, some methods (Hu et al 2019;Wang et al 2021a;Kang et al 2022;Zhao et al 2019c,b;Mok and Chung 2020;Wang et al 2021b;Lv et al 2022;Hu et al 2022) proposed to decompose the target deformation field by multi-scale CNNs or multiple cascaded CNNs models. Despite achieving good registration performance, such atlas-based segmentation methods are susceptible to tissue gray-scale blurring, resulting in inaccurate segmentation results, since they only rely on the similarity between images and lack guidance of the anatomical structures.…”

Section: Pointed Out That the Atlas-basedmentioning

confidence: 99%

“…We adopt the current state-of-the-art registration method (Lv et al 2022) as our reg-model, and a 3D-UNet (C ¸ic ¸ek et al 2016) with a strategy of deep supervision as our segmodel. For both the unsupervised (initial) and weakly supervised (iterative) training phase of the reg-model, the learning rate was set to 1×10 −4 , and the training was performed for 40,000 steps.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…Another well-known solution is called atlas-based segmentation (Wang et al 2021a(Wang et al , 2020Lv et al 2022;Wang et al 2019;Coupé et al 2011), which enables a labelefficient segmentation with the help of medical image registration. The basic idea of these methods is to 'spread' the segmentation label of an atlas (i.e., a special sample with high image quality for accurate manual annotation) onto a target image that is about to be segmented.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust One-Shot Segmentation of Brain Tissues via Image-Aligned Style Transformation

Zeng

Wang

et al. 2023

AAAI

View full text Add to dashboard Cite

One-shot segmentation of brain tissues is typically a dual-model iterative learning: a registration model (reg-model) warps a carefully-labeled atlas onto unlabeled images to initialize their pseudo masks for training a segmentation model (seg-model); the seg-model revises the pseudo masks to enhance the reg-model for a better warping in the next iteration. However, there is a key weakness in such dual-model iteration that the spatial misalignment inevitably caused by the reg-model could misguide the seg-model, which makes it converge on an inferior segmentation performance eventually. In this paper, we propose a novel image-aligned style transformation to reinforce the dual-model iterative learning for robust one-shot segmentation of brain tissues. Specifically, we first utilize the reg-model to warp the atlas onto an unlabeled image, and then employ the Fourier-based amplitude exchange with perturbation to transplant the style of the unlabeled image into the aligned atlas. This allows the subsequent seg-model to learn on the aligned and style-transferred copies of the atlas instead of unlabeled images, which naturally guarantees the correct spatial correspondence of an image-mask training pair, without sacrificing the diversity of intensity patterns carried by the unlabeled images. Furthermore, we introduce a feature-aware content consistency in addition to the image-level similarity to constrain the reg-model for a promising initialization, which avoids the collapse of image-aligned style transformation in the first iteration. Experimental results on two public datasets demonstrate 1) a competitive segmentation performance of our method compared to the fully-supervised method, and 2) a superior performance over other state-of-the-art with an increase of average Dice by up to 4.67%. The source code is available at: https://github.com/JinxLv/One-shot-segmentation-via-IST.

show abstract

Joint Progressive and Coarse-to-Fine Registration of Brain MRI via Deformation Field Integration and Non-Rigid Feature Fusion

Cited by 32 publications

References 44 publications

EasyReg: A ready-to-use deep learning tool for symmetric affine and nonlinear brain MRI registration

EasyReg: A ready-to-use deep learning tool for symmetric affine and nonlinear brain MRI registration

Learn2Reg: Comprehensive Multi-Task Medical Image Registration Challenge, Dataset and Evaluation in the Era of Deep Learning

Robust One-Shot Segmentation of Brain Tissues via Image-Aligned Style Transformation

Contact Info

Product

Resources

About