Unsupervised domain adaptation in brain lesion segmentation with adversarial networks

By Konstantinos Kamnitsas, Christian F. Baumgartner, Christian Ledig, Virginia F. J. Newcombe, Joanna P. Simpson, Andrew D. Kane, David K. Menon, Aditya V. Nori, Antonio Criminisi, Daniel Rueckert and 1 more. 2016

544 citations recorded by the source corpus.

Significant advances have been made towards building accurate automatic segmentation systems for a variety of biomedical applications using machine learning. However, the performance of these systems often degrades when they are applied on new data that differ from the training data, for example, due to variations in imaging protocols. Manually annotating new data for each test domain is not a feasible solution. In this work we investigate unsupervised domain adaptation using adversarial neural networks to train a segmentation method which is more robust to differences in the input data, and which does not require any annotations on the test domain. Specifically, we derive domain-invariant features by learning to counter an adversarial network, which attempts to classify the domain of the input data by observing the activations of the segmentation network. Furthermore, we propose a multi-connected domain discriminator for improved adversarial training. Our system is evaluated using two MR databases of subjects with traumatic brain injuries, acquired using different scanners and imaging protocols. Using our unsupervised approach, we obtain segmentation accuracies which are close to the upper bound of supervised domain adaptation.

Subfields: Artificial Intelligence, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. P. Heng
2. Qi-Yan Dou
3. Jing Qin
4. Hao Chen
5. Lequan Yu
6. Arvid Lundervold
7. Daniel Rueckert
8. Sergey I. Nikolenko

In Radiology, Nuclear Medicine and Imaging

1. P. Heng
2. Arvid Lundervold
3. Qi-Yan Dou
4. Hao Chen
5. Lequan Yu
6. Jing Qin
7. Daniel Rueckert
8. B. van Ginneken

Citations in the corpus

• An overview of deep learning in medical imaging focusing on MRI (2018, background): “ multiple large companies are pursuing healthcare projects based on machine learning. Not only medical technology companies, but also for example Google Brain [29, 30, 31]5, DeepMind [32]6, Microsoft [33, 34]7 and IBM [35]8. There is also a plethora of small and medium-sized businesses in the eld9. ”
• Deep learning based brain tumor segmentation: a survey (2020, background): “In spite of their performance [69] and [68], one main drawback is that the loss function of the discriminator generates”
• End-to-End Adversarial Retinal Image Synthesis (2018, background): “sequences [5], to estimate cross-modality transformations [6], or to perform knowledge transfer by learning features invariant to the MR scanning protocol [7].”
• A Review of Domain Adaptation without Target Labels (2019, background): “They have also been applied to a variety of problem settings, such as speech recognition [163], medical image segmentation [41] and cross-language knowledge transfer [164].”
• GANs for Medical Image Analysis (2018, methodology): “Similarly, [39] utilizes a cGAN for synthesizing high-resolution PET images from pairs of CT images and binary labelmaps.”
• Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation (2020, background): “The large performance degradation of deep models has been observed between different MRI sequences in [6].”
• Deep Learning Applications in Medical Image Analysis (2018, methodology): “Aside from synthetic data generation, GANs have been used in brain MRI segmentation as well by Moeskops et al. [91], Kamnitsas et al. [92] and Alex et al. [93].”
• A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis (2019, background): “Thus, once trained models do not transfer seamlessly from one clinical task or site to another because of an often yawning domain gap (Kamnitsas et al., 2017; Ben-David et al., 2010).”
• Multi-Domain Adversarial Learning (2019, methodology): “DA has been applied to image-based datasets for improving or accelerating image segmentation tasks (Becker et al., 2015; van Opbroek et al., 2015; Bermúdez-Chacón et al., 2016; Kamnitsas et al., 2017).”
• Adversarial Domain Adaptation and Pseudo-Labeling for Cross-Modality Microscopy Image Quantification (2019, methodology): “In particular, adversarial domain adaptation has been applied to medical image segmentation and classification [12,17].”
• Synthetic Data for Deep Learning (2019, methodology): “[302] use unsupervised DA for brain lesion segmentation in 3D, switching from one type of MR images to another in domain adaptation.”
• Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (2017, unknown): “Multiple bifurcations are used to make domain classification possible from different levels of abstraction and to improve training stability as in [4].”