Enhanced Target Domain Representation BasedUnsupervised Cross-Domain Medical Image Segmentation
LIU Kai1, LU Runuo1, ZHENG Xiaorou1, DONG Shoubin1
1. Guangdong Provincial Key Laboratory of Multimodal Big Data Intelligent Analysis, School of Computer Science and Engineering, South China University of Technology, Guangzhou 510641
Abstract Medical images produced by different imaging modality devices exhibit varying degrees of distribution differences. Unsupervised domain adaptation methods typically aim to generalize models trained in the source domain to the unlabeled target domain by minimizing these distribution differences and using shared features between the source and target domains for result prediction. However, they often neglect the private features of the target domain. To address this issue, a method for enhanced target domain representation based unsupervised cross-domain medical image segmentation(TreUCMIS) is proposed in this paper. First, TreUCMIS acquires common features through shared feature learning, and a target domain feature encoder is trained through image reconstruction to capture the complete features of the target domain. Second, unsupervised self-training of the target domain strengthens the shared characteristics of deep and shallow features. Finally, the predicted results obtained from the shared and complete features are aligned, enabling the model to utilize the complete features of the target domain for segmentation and thus improving the generalization in the target domain. Experiments on two medical image segmentation datasets involving bidirectional domain adaptation tasks with CT and MRI(abdominal and cardiac datasets) demonstrate the effectiveness and superiority of TreUCMIS.
Fund:National Natural Science Foundation of China(No.61976239), Innovation Fund of High-End Scientific Research Institutions of Zhongshan City(No.2019AG031)
Corresponding Authors:
DONG Shoubin, Ph.D., professor. Her research interests include high performance computing, bioinformatics and biocomputing, and natural language processing.
About author:: ZHENG Xiaorou, Ph.D., assistant professor. Her research interests include hyperspectral remote sensing imagery, semi-supervised learning and cross-domain transfer learning. LU Runuo, Master student. Her research interests include computer vision and remote sensing imagery. LIU Kai, Master student. His research interests include deep learning, computer vision and image recognition.
[1] YANG C, GUO X Q, CHEN Z, et al. Source Free Domain Adaptation for Medical Image Segmentation with Fourier Style Mining. Medical Image Analysis, 2022, 79. DOI: 10.1016/j.media.2022.102457. [2] GUAN H, LIU M X.Domain Adaptation for Medical Image Analysis: A Survey. IEEE Transactions on Biomedical Engineering, 2022, 69(3): 1173-1185. [3] GHAFOORIAN M, MEHRTASH A, KAPUR T, et al. Transfer Lear-ning for Domain Adaptation in MRI: Application in Brain Lesion Segmentation//Proc of the 20th International Conference on Medical Image Computing and Computer Assisted Intervention. Berlin, Germany: Springer, 2017: 516-524. [4] HOFFMAN J, TZENG E, PARK T, et al. CyCADA: Cycle-Consistent Adversarial Domain Adaptation. Journal of Machine Learning Research, 2018, 80: 1989-1998. [5] ZENG G D, LERCH T D, SCHMARANZER F, et al. Semantic Consistent Unsupervised Domain Adaptation for Cross-Modality Medi-cal Image Segmentation//Proc of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention. Berlin, Germany: Springer, 2021: 201-210. [6] ZHU J Y, PARK T, ISOLA P, et al. Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks//Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2017: 2242-2251. [7] DU X Q, LIU Y G.Constraint-Based Unsupervised Domain Adaptation Network for Multi-modality Cardiac Image Segmentation. IEEE Journal of Biomedical and Health Informatics, 2022, 26(1): 67-78. [8] LIU S L, YIN S Q, QU L H, et al. Reducing Domain Gap in Frequency and Spatial Domain for Cross-Modality Domain Adaptation on Medical Image Segmentation. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(2): 1719-1727. [9] MA X, YUAN J K, CHEN Y W, et al. Attention-Based Cross-Layer Domain Alignment for Unsupervised Domain Adaptation. Neurocomputing, 2022, 499. DOI: 10.1016/j.neucom.2022.04.086. [10] DOU Q, OUYANG C, CHEN C, et al. PnP-AdaNet: Plug-and-Play Adversarial Domain Adaptation Network at Unpaired Cross-Modality Cardiac Segmentation. IEEE Access, 2019, 7: 99065-99076. [11] TSAI Y H, HUNG W C, SCHULTER S, et al. Learning to Adapt Structured Output Space for Semantic Segmentation//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 7472-7481. [12] FENG W, JU L, WANG L, et al. Unsupervised Domain Adaptation for Medical Image Segmentation by Selective Entropy Con-straints and Adaptive Semantic Alignment. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(1): 623-631. [13] SURUCU M, ISLER Y, PERC M, et al. Convolutional Neural Networks Predict the Onset of Paroxysmal Atrial Fibrillation: Theory and Applications. Chaos: An Interdisciplinary Journal of Nonlinear Science, 2021, 31(11). DOI: 10.1063/5.0069272. [14] YANG Y, WANG B, ZHANG D W, et al. Self-Supervised Interactive Embedding for One-Shot Organ Segmentation. IEEE Tran-sactions on Biomedical Engineering, 2023, 70(10): 2799-2808. [15] BLEKER J, KWEE T C, ROUW D, et al. A Deep Learning Masked Segmentation Alternative to Manual Segmentation in Biparametric MRI Prostate Cancer Radiomics. European Radiology, 2022, 32(9): 6526-6535. [16] ITO R, NAKAE K, HATA J, et al. Semi-supervised Deep Lear-ning of Brain Tissue Segmentation. Neural Networks, 2019, 116: 25-34. [17] ZHANG J, SAHA A, ZHU Z, et al. Hierarchical Convolutional Neural Networks for Segmentation of Breast Tumors in MRI with Application to Radiogenomics. IEEE Transactions on Medical Imaging, 2019, 38(2): 435-447. [18] DALMIŞ M U, LITJENS G, HOLLAND K, et al. Using Deep Learning to Segment Breast and Fibroglandular Tissue in MRI Vo-lumes. Medical Physics, 2017, 44(2): 533-546. [19] KAVUR A E, GEZER N S, BARIŞ M,et al. CHAOS Challenge-Combined(CT-MR) Healthy Abdominal Organ Segmentation. Me-dical Image Analysis, 2021, 69. DOI: 10.1016/j.media.2020.101950. [20] MORRIS E D, GHANEM A I, DONG M, et al. Cardiac Substructure Segmentation with Deep Learning for Improved Cardiac Spa-ring. Medical Physics, 2020, 47(2): 576-586. [21] CHEN C, DOU Q, CHEN H, et al. Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentation. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 865-872. [22] ZHANG Y, MIAO S, MANSI T, et al. Task Driven Generative Modeling for Unsupervised Domain Adaptation: Application to X-Ray Image Segmentation//Proc of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Berlin, Germany: Springer, 2018: 599-607. [23] ZHANG Z Z, YANG L, ZHENG Y F.Translating and Segmenting Multimodal Medical Volumes with Cycle-and Shape-Consistency Generative Adversarial Network//Proc of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 9242-9251. [24] TZENG E, HOFFMAN J, SAENKO K, et al. Adversarial Discri-minative Domain Adaptation//Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 2962-2971. [25] GANIN Y, USTINOVA E, AJAKAN H, et al. Domain-Adversarial Training of Neural Networks. Journal of Machine Learning Research, 2016, 17(1): 1-35. [26] HU S S, LIAO Z H, XIA Y.Domain Specific Convolution and High Frequency Reconstruction Based Unsupervised Domain Adaptation for Medical Image Segmentation//Proc of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Berlin, Germany: Springer, 2022: 650-659. [27] CHEN C, DOU Q, CHEN H, et al. Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation. IEEE Transactions on Medical Imaging, 2020, 39(7): 2494-2505. [28] YANG Y C, SOATTO S.FDA: Fourier Domain Adaptation for Semantic Segmentation//Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2020: 4084-4094. [29] ZHAO Z Y, ZHOU F C, XU K X, et al. LE-UDA: Label-Efficient Unsupervised Domain Adaptation for Medical Image Segmen-tation. IEEE Transactions on Medical Imaging, 2023, 42(3): 633-646. [30] SOHN K, BERTHELOT D, LI C L, et al. FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence//Proc of the 34th International Conference on Neural Information Proce-ssing Systems. Cambridge, USA: MIT Press, 2020: 596-608. [31] CHEN T, KORNBLITH S, NOROUZI M, et al. A Simple Framework for Contrastive Learning of Visual Representations. Journal of Machine Learning Research, 2020, 119: 1597-1607. [32] LIU X F, HU B, LIU X C, et al. Energy-Constrained Self-Trai-ning for Unsupervised Domain Adaptation//Proc of the 25th International Conference on Pattern Recognition. Washington, USA: IEEE, 2021: 7515-7520. [33] KUANG S, WOODRUFF H C, GRANZIER R, et al. MSCDA: Multi-level Semantic-Guided Contrast Improves Unsupervised Domain Adaptation for Breast MRI Segmentation in Small Datasets. Neural Networks, 2023, 165: 119-134. [34] ZHAO Z Y, XU K X, LI S M, et al. MT-UDA: Towards Unsupervised Cross-Modality Medical Image Segmentation with Limited Source Labels//Proc of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention. Berlin, Germany: Springer, 2021: 293-303. [35] BOUSMALIS K, TRIGEORGIS G, SILBERMAN N, et al. Domain Separation Networks//Proc of the 30th International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2016: 343-351. [36] LANDMAN B, XU Z, IGLESIAS J.E, et al. Multi-atlas Labeling beyond the Cranial Vault[EB/OL].[2024-07-17]. https://www.synapse.org/Synapse:syn3193805/wiki/217789. [37] ZHUANG X H, SHEN J.Multi-scale Patch and Multi-modality Atlases for Whole Heart Segmentation of MRI. Medical Image Analysis, 2016, 31: 77-87.
2024, Vol. 37 
