Multi-domain Adaptation in Brain MRI Through Paired Consistency and Adversarial Learning

Orbes-Arteaga, Mauricio; Varsavsky, Thomas; Sudre, Carole H.; Eaton-Rosen, Zach; Haddow, Lewis; Sørensen, Lauge; Nielsen, Mads; Pai, Akshay; Ourselin, Sébastien; Modat, Marc; Nachev, Parashkev; Cardoso, M. Jorge

doi:10.1007/978-3-030-33391-1_7

Cited by 38 publications

(33 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distribution dissimilarity can be assessed using correlation distances ( Sun et al, 2016 ) or maximum mean discrepancy ( Pan et al, 2011 ; Long et al, 2015 ). However, more recent techniques are mostly focused on adversarial methods, achieving promising results in medical imaging ( Kamnitsas et al, 2017 ; Dou et al, 2018 ; Orbes-Arteaga et al, 2019 ). However, these methods are usually focus on solving a single task across domain.…”

Section: Related Workmentioning

confidence: 99%

“…Note that (16) stands for features extracted at any level of h θ . In this work and similarly to Orbes-Arteaga et al (2019 ), the contracting path features from the U-Net are used as input of the discriminator.…”

Section: Matching Feature Distributions Across Datasetsmentioning

confidence: 99%

“…As in Kamnitsas et al (2017 ); Orbes-Arteaga et al (2019 ), the DA loss is skipped for few epochs (20 in our experiments) in order to initialise the discriminator. The segmentation loss L s eg is the same as in (18) .…”

Section: Implementation Of the Joint Model Optimisationmentioning

confidence: 99%

“…As suggested by Ulyanov et al (2016 ), we used instance normalisation. We used the same discriminator as in Orbes-Arteaga et al (2019 ).…”

Section: Implementation Of the Joint Model Optimisationmentioning

confidence: 99%

See 3 more Smart Citations

Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets

Dorent

Booth

et al. 2021

Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

Brain tissue segmentation from multimodal MRI is a key building block of many neuroimaging analysis pipelines. Established tissue segmentation approaches have, however, not been developed to cope with large anatomical changes resulting from pathology, such as white matter lesions or tumours, and often fail in these cases. In the meantime, with the advent of deep neural networks (DNNs), segmentation of brain lesions has matured significantly. However, few existing approaches allow for the joint segmentation of normal tissue and brain lesions. Developing a DNN for such a joint task is currently hampered by the fact that annotated datasets typically address only one specific task and rely on task-specific imaging protocols including a task-specific set of imaging modalities. In this work, we propose a novel approach to build a joint tissue and lesion segmentation model from aggregated task-specific hetero-modal domain-shifted and partially-annotated datasets. Starting from a variational formulation of the joint problem, we show how the expected risk can be decomposed and optimised empirically. We exploit an upper bound of the risk to deal with heterogeneous imaging modalities across datasets. To deal with potential domain shift, we integrated and tested three conventional techniques based on data augmentation, adversarial learning and pseudo-healthy generation. For each individual task, our joint approach reaches comparable performance to task-specific and fully-supervised models. The proposed framework is assessed on two different types of brain lesions: White matter lesions and gliomas. In the latter case, lacking a joint ground-truth for quantitative assessment purposes, we propose and use a novel clinically-relevant qualitative assessment methodology.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Matching Feature Distributions Across Datasetsmentioning

confidence: 99%

Section: Implementation Of the Joint Model Optimisationmentioning

confidence: 99%

“…As suggested by Ulyanov et al (2016 ), we used instance normalisation. We used the same discriminator as in Orbes-Arteaga et al (2019 ).…”

Section: Implementation Of the Joint Model Optimisationmentioning

confidence: 99%

See 2 more Smart Citations

Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets

Dorent

Booth

et al. 2021

Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, one major drawback of these methods is their applicability in a clinical setting, as many models rely on the assumption that the source and target domains are drawn from the same distribution. As a result, the efficiency of these models may drop drastically when applied to images which were acquired with acquisition protocols different than the ones used to train the models (Kamnitsas et al, 2017; Orbes-Arteaga et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Harmonised segmentation of neonatal brain MRI

Grigorescu

Vanes

Uus

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Deep learning based medical image segmentation has shown great potential in becoming a key part of the clinical analysis pipeline. However, many of these models rely on the assumption that the train and test data come from the same distribution. This means that such methods cannot guarantee high quality predictions when the source and target domains are dissimilar due to different acquisition protocols, or biases in patient cohorts. Recently, unsupervised domain adaptation (DA) techniques have shown great potential in alleviating this problem by minimizing the shift between the source and target distributions, without requiring the use of labelled data in the target domain. In this work, we aim to predict tissue segmentation maps on T2-weighted (T2w) magnetic resonance imaging (MRI) data of an unseen preterm-born neonatal population, which has both different acquisition parameters and population bias when compared to our training data. We achieve this by investigating two unsupervised DA techniques with the objective of finding the best solution for our problem. We compare the two methods with a baseline fully-supervised segmentation network and report our results in terms of Dice scores obtained on our ground truth test dataset. Moreover, we analyse tissue volumes and cortical thickness (CT) measures of the harmonised data on a subset of the population matched for gestational age (GA) at birth and postmenstrual age (PMA) at scan. Finally, we demonstrate the applicability of the harmonised cortical gray matter maps with an analysis comparing term and preterm-born neonates and a proof-of-principle investigation of the association between CT and a language outcome measure.

show abstract

Unsupervised shape‐and‐texture‐based generative adversarial tuning of pre‐trained networks for carotid segmentation from 3D ultrasound images

Chen,

Jiang,

Chiu

2024

Medical Physics

View full text Add to dashboard Cite

BackgroundVessel‐wall volume and localized three‐dimensional ultrasound (3DUS) metrics are sensitive to the change of carotid atherosclerosis in response to medical/dietary interventions. Manual segmentation of the media‐adventitia boundary (MAB) and lumen‐intima boundary (LIB) required to obtain these metrics is time‐consuming and prone to observer variability. Although supervised deep‐learning segmentation models have been proposed, training of these models requires a sizeable manually segmented training set, making larger clinical studies prohibitive.PurposeWe aim to develop a method to optimize pre‐trained segmentation models without requiring manual segmentation to supervise the fine‐tuning process.MethodsWe developed an adversarial framework called the unsupervised shape‐and‐texture generative adversarial network (USTGAN) to fine‐tune a convolutional neural network (CNN) pre‐trained on a source dataset for accurate segmentation of a target dataset. The network integrates a novel texture‐based discriminator with a shape‐based discriminator, which together provide feedback for the CNN to segment the target images in a similar way as the source images. The texture‐based discriminator increases the accuracy of the CNN in locating the artery, thereby lowering the number of failed segmentations. Failed segmentation was further reduced by a self‐checking mechanism to flag longitudinal discontinuity of the artery and by self‐correction strategies involving surface interpolation followed by a case‐specific tuning of the CNN. The U‐Net was pre‐trained by the source dataset involving 224 3DUS volumes with 136, 44, and 44 volumes in the training, validation and testing sets. The training of USTGAN involved the same training group of 136 volumes in the source dataset and 533 volumes in the target dataset. No segmented boundaries for the target cohort were available for training USTGAN. The validation and testing of USTGAN involved 118 and 104 volumes from the target cohort, respectively. The segmentation accuracy was quantified by Dice Similarity Coefficient (DSC), and incorrect localization rate (ILR). Tukey's Honestly Significant Difference multiple comparison test was employed to quantify the difference of DSCs between models and settings, where was considered statistically significant.ResultsUSTGAN attained a DSC of % in LIB and % in MAB, improving from the baseline performance of % in LIB (p ) and % in MAB (p ). Our approach outperformed six state‐of‐the‐art domain‐adaptation models (MAB: , LIB: ). The proposed USTGAN also had the lowest ILR among the methods compared (LIB: 2.5%, MAB: 1.7%).ConclusionOur framework improves segmentation generalizability, thereby facilitating efficient carotid disease monitoring in multicenter trials and in clinics with less expertise in 3DUS imaging.

show abstract

Multi-domain Adaptation in Brain MRI Through Paired Consistency and Adversarial Learning

Cited by 38 publications

References 14 publications

Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets

Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets

Harmonised segmentation of neonatal brain MRI

Unsupervised shape‐and‐texture‐based generative adversarial tuning of pre‐trained networks for carotid segmentation from 3D ultrasound images

Contact Info

Product

Resources

About