Unsupervised domain adaptation for cross-modality liver segmentation via joint adversarial learning and self-learning

• Published in: Applied soft computing Vol. 121; p. 108729
• Main Authors: Hong, Jin, Yu, Simon Chun-Ho, Chen, Weitian
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.05.2022
• Subjects: Adversarial learning; Liver segmentation; Post-situ identification manner; Self-learning; Student-to-partner learning; Unsupervised domain adaptation; Unsupervised domain adaptation; Post-situ identification manner; Liver segmentation; Self-learning; Adversarial learning; Student-to-partner learning
• DOI: 10.1016/j.asoc.2022.108729
• ISSN: 1568-4946

Liver segmentation on images acquired using computed tomography (CT) and magnetic resonance imaging (MRI) plays an important role in clinical management of liver diseases. Compared to MRI, CT images of liver are more abundant and readily available. However, MRI can provide richer quantitative information of the liver compared to CT. Thus, it is desirable to achieve unsupervised domain adaptation for transferring the learned knowledge from the source domain containing labeled CT images to the target domain containing unlabeled MR images. In this work, we report a novel unsupervised domain adaptation framework for cross-modality liver segmentation via joint adversarial learning and self-learning. We propose joint semantic-aware and shape-entropy-aware adversarial learning with post-situ identification manner to implicitly align the distribution of task-related features extracted from the target domain with those from the source domain. In proposed framework, a network is trained with the above two adversarial losses in an unsupervised manner, and then a mean completer of pseudo-label generation is employed to produce pseudo-labels to train the next network (desired model). Additionally, semantic-aware adversarial learning and two self-learning methods, including pixel-adaptive mask refinement and student-to-partner learning, are proposed to train the desired model. To improve the robustness of the desired model, a low-signal augmentation function is proposed to transform MRI images as the input of the desired model to handle hard samples. Using the public datasets, our experiments demonstrated the proposed unsupervised domain adaptation framework reached four supervised learning methods with a Dice score 0.912 ± 0.037 (mean ± standard deviation). •An unsupervised domain adaptation method via joint adversarial learning and self-learning for medical image segmentation.•A post-situ identification manner for adversarial learning to focus on alignment of task-related features.•A joint semantic-aware and shape-entropy-aware adversarial learning to align the distribution of low-level features.•A mean completer approach for pseudo-label generation and a low-signal augmentation function to improve the model robustness.•A novel self-learning mechanism named student-to-partner learning for further improving the segmentation results.