請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/95198完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 藍俊宏 | zh_TW |
| dc.contributor.advisor | Jakey Blue | en |
| dc.contributor.author | 洪振倫 | zh_TW |
| dc.contributor.author | Zhen-Lun Hong | en |
| dc.date.accessioned | 2024-08-30T16:08:16Z | - |
| dc.date.available | 2024-08-31 | - |
| dc.date.copyright | 2024-08-30 | - |
| dc.date.issued | 2024 | - |
| dc.date.submitted | 2024-08-13 | - |
| dc.identifier.citation | [1] Bansal, H., & Rathore, A. (2017). Understanding and implementing CycleGAN in TensorFlow. Retrieved from https://hardikbansal.github.io/CycleGANBlog/ [Accessed: 6-August-2024].
[2] Ben-David, S., Blitzer, J., Crammer, K., & Pereira, F. C. (2006). Analysis of representations for domain adaptation. Neural Information Processing Systems. [3] Cao, H., Wang, Y., Chen, J., Jiang, D., Zhang, X., Tian, Q., & Wang, M. (2021). Swin-Unet: UNet-like pure transformer for medical image segmentation. ECCV Workshops. [4] Cardoso, M. J., Li, W., Brown, R., Ma, N., Kerfoot, E., Wang, Y., et al. (2022). MONAI: An open-source framework for deep learning in healthcare. ArXiv, abs/2211.02701. [5] Caussy, C., Reeder, S. B., Sirlin, C. B., & Loomba, R. S. (2018). Noninvasive, quantitative assessment of liver fat by MRI-PDFF as an endpoint in NASH trials. Hepatology, 68. [6] Chartsias, A., Joyce, T., Dharmakumar, R., & Tsaftaris, S. A. (2017). Adversarial image synthesis for unpaired multi-modal cardiac data. SASHIMI@MICCAI. [7] Chen, C., Dou, Q., Chen, H., Qin, J., & Heng, P.-A. (2020). Unsupervised bidirectional cross-modality adaptation via deeply synergistic image and feature alignment for medical image segmentation. IEEE Transactions on Medical Imaging, 39, 2494–2505. [8] Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., et al. (2021). TransUNet: Transformers make strong encoders for medical image segmentation. ArXiv, abs/2102.04306. [9] Choudhary, A., Tong, L., Zhu, Y., & Wang, M. D. (2020). Advancing medical imaging informatics by deep learning-based domain adaptation. Yearbook of Medical Informatics, 29, 129–138. [10] Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., et al. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. ArXiv, abs/2010.11929. [11] Dou, Q., Liu, Q., Heng, P.-A., & Glocker, B. (2020). Unpaired multi-modal segmentation via knowledge distillation. IEEE Transactions on Medical Imaging, 39, 2415–2425. [12] Fang, X., & Yan, P. (2020). Multi-organ segmentation over partially labeled datasets with multi-scale feature abstraction. IEEE Transactions on Medical Imaging, 39(11), 3619–3629. https://doi.org/10.1109/TMI.2020.3001036 [13] Feng, S., Zhou, Y., Zhang, X., Zhang, Y., & Wang, Y. (2021). MS-KD: Multi-organ segmentation with multiple binary-labeled datasets. ArXiv, abs/2108.02559. [14] Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. S. (2015). Domain-adversarial training of neural networks. Journal of Machine Learning Research. [15] Gao, Y., Li, Z., Liu, D., Zhou, M., Zhang, S., & Metaxas, D. N. (2023). Training like a medical resident: Universal medical image segmentation via context prior learning. ArXiv, abs/2306.02416. [16] Gibson, E., Giganti, F., Hu, Y., Bonmati, E., Bandula, S., Gurusamy, K. S., et al. (2018). Automatic multi-organ segmentation on abdominal CT with dense V-networks. IEEE Transactions on Medical Imaging, 37, 1822–1834. [17] Gretton, A., Borgwardt, K. M., Rasch, M. J., Scholkopf, B., & Smola, A. (2012). A kernel two-sample test. Journal of Machine Learning Research, 13, 723–773. [18] Gutmann, M. U., & Hyvärinen, A. (2010). Noise-contrastive estimation: A new estimation principle for unnormalized statistical models. International Conference on Artificial Intelligence and Statistics. [19] Hastie, T., Tibshirani, R., & Friedman, J. (2001). The elements of statistical learning. Springer New York Inc. [20] He, K., Fan, H., Wu, Y., Xie, S., & Girshick, R. B. (2019). Momentum contrast for unsupervised visual representation learning. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9726–9735. [21] Hoffman, J., Tzeng, E., Park, T., Zhu, J.-Y., Isola, P., Saenko, K., Efros, A. A., & Darrell, T. (2017). CyCADA: Cycle-consistent adversarial domain adaptation. ArXiv, abs/1711.03213. [22] Hoyer, L., Dai, D., & Gool, L. V. (2021). DAFormer: Improving network architectures and training strategies for domain-adaptive semantic segmentation. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9914–9925. [23] Hsieh, K.-C. (2018). 腹部電腦斷層掃描影像之骨骼肌自動判別與區分方法及臨床應用研究 [Master’s thesis, 國立臺灣大學]. https://doi.org/10.6342/NTU201803394 [24] Huang, R., Zheng, Y., Hu, Z., Zhang, S., & Li, H. (2020). Multi-organ segmentation via co-training weight-averaged models from few-organ datasets. In A. L. Martel, P. Abolmaesumi, D. Stoyanov, et al. (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 (pp. 146–155). Springer International Publishing. [25] Ji, Y., Bai, H., Yang, J., Ge, C., Zhu, Y., Zhang, R., et al. (2022). AMOS: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation. ArXiv, abs/2206.08023. [26] Jiang, J., Hu, Y.-C., Tyagi, N., Zhang, P., Rimner, A., Mageras, G. S., et al. (2018). Tumor-aware, adversarial domain adaptation from CT to MRI for lung cancer segmentation. MICCAI, 11071, 777–785. [27] Kaur, H., Kaur, N., & Neeru, N. (2022). Evolution of multiorgan segmentation techniques from traditional to deep learning in abdominal CT images: A systematic review. Displays, 73, 102223. https://doi.org/10.1016/j.displa.2022.102223 [28] Kobayashi, H., Lei, C., Wu, Y., Mao, A., Jiang, Y., Guo, B., et al. (2017). Label-free detection of cellular drug responses by high-throughput bright-field imaging and machine learning. Scientific Reports, 7. https://doi.org/10.1038/s41598-017-12378-4 [29] Li, J., Zhou, P., Xiong, C., Socher, R., & Hoi, S. C. H. (2020). Prototypical contrastive learning of unsupervised representations. ArXiv, abs/2005.04966. [30] Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., et al. (2021). Swin Transformer: Hierarchical vision transformer using shifted windows. 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 9992–10002. [31] Luo, X., Liao, W., Xiao, J., Chen, J., Song, T., Zhang, X., et al. (2021). WORD: A large-scale dataset, benchmark, and clinical applicable study for abdominal organ segmentation from CT image. Medical Image Analysis, 82, 102642. [32] Majchrowska, S., Mikołajczyk, A., Ferlin, M. A., Klawikowska, Z., Plantykow, M. A., Kwasigroch, A., & Majek, K. (2021). Waste detection in Pomerania: Non-profit project for detecting waste in environment. Waste Management, 138, 274–284. [33] Milletari, F., Navab, N., & Ahmadi, S.-A. (2016). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. 2016 Fourth International Conference on 3D Vision (3DV), 565–571. https://doi.org/10.1109/3DV.2016.79 [34] Qiu, Y.-X. (2020). Dixon 核磁共振影像之自動分割及放射學研究及其骨骼肌與肝臟組織之應用 [Master’s thesis, 國立臺灣大學]. https://doi.org/10.6342/ntu202003009 [35] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. ArXiv, abs/1505.04597. [36] Saito, K., Watanabe, K., Ushiku, Y., & Harada, T. (2017). Maximum classifier discrepancy for unsupervised domain adaptation. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3723–3732. [37] Shi, G., Xiao, L., Chen, Y., & Zhou, S. K. (2021). Marginal loss and exclusion loss for partially supervised multi-organ segmentation. Medical Image Analysis, 70, 101979. https://doi.org/10.1016/j.media.2021.101979 [38] Snell, J., Swersky, K., & Zemel, R. S. (2017). Prototypical networks for few-shot learning. Neural Information Processing Systems. [39] Sun, B., Feng, J., & Saenko, K. (2016). Correlation alignment for unsupervised domain adaptation. ArXiv, abs/1612.01939. [40] Sun, B., & Saenko, K. (2016). Deep CORAL: Correlation alignment for deep domain adaptation. ECCV Workshops. [41] Tsai, Y.-H., Hung, W.-C., Schulter, S., Sohn, K., Yang, M.-H., & Chandraker, M. (2018). Learning to adapt structured output space for semantic segmentation. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7472–7481. [42] Tzeng, E., Hoffman, J., Saenko, K., & Darrell, T. (2017). Adversarial discriminative domain adaptation. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2962–2971. [43] Tzeng, E., Hoffman, J., Zhang, N., Saenko, K., & Darrell, T. (2014). Deep domain confusion: Maximizing for domain invariance. ArXiv, abs/1412.3474. [44] Valindria, V. V., Pawlowski, N., Rajchl, M., Lavdas, I., Aboagye, E. O., Rockall, A. G., et al. (2018). Multi-modal learning from unpaired images: Application to multi-organ segmentation in CT and MRI. 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), 547–556. https://doi.org/10.1109/WACV.2018.00066 [45] van den Oord, A., Li, Y., & Vinyals, O. (2018). Representation learning with contrastive predictive coding. ArXiv, abs/1807.03748. [46] Vaswani, A., Shazeer, N. M., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., et al. (2017). Attention is all you need. Neural Information Processing Systems. [47] Wang, J., & Chen, Y. (2023). Introduction to transfer learning: Algorithms and practice. Springer Nature Singapore. [48] Wang, P., Shen, C., Wang, W., Oda, M., Fuh, C.-S., Mori, K., & Roth, H. R. (2023). CondistFL: Conditional distillation for federated learning from partially annotated data. ArXiv, abs/2308.04070. [49] Wei, G., Lan, C., Zeng, W., & Chen, Z. (2021). ToAlign: Task-oriented alignment for unsupervised domain adaptation. ArXiv, abs/2106.10812. [50] Wu, J., Kamath, M., Noseworthy, M. D., Boylan, C., & Poehlman, S. (2008). Segmentation of images of abdominal organs. Critical Reviews™ in Biomedical Engineering, 36(5-6), 305–334. [51] Yu, H., Yang, L. T., Zhang, Q., Armstrong, D., & Deen, M. J. (2021). Convolutional neural networks for medical image analysis: State-of-the-art, comparisons, improvement and perspectives. Neurocomputing, 444, 92–110. https://doi.org/10.1016/j.neucom.2020.04.157 [52] Zhang, P., Zhang, B., Zhang, T., Chen, D., Wang, Y., & Wen, F. (2021). Prototypical pseudo label denoising and target structure learning for domain adaptive semantic segmentation. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 12409–12419. [53] Zhang, Q., Zhang, J., Liu, W., & Tao, D. (2019). Category anchor-guided unsupervised domain adaptation for semantic segmentation. ArXiv, abs/1910.13049. [54] Zheng, Z., & Yang, Y. (2020). Rectifying pseudo label learning via uncertainty estimation for domain adaptive semantic segmentation. International Journal of Computer Vision, 129, 1106–1120. [55] Zhou, T., Wang, W., Konukoglu, E., & Gool, L. V. (2022). Rethinking semantic segmentation: A prototype view. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2572–2583. [56] Zhou, Y., Li, Z., Bai, S., Chen, X., Han, M., Wang, C., Fishman, E., & Yuille, A. (2019). Prior-aware neural network for partially-supervised multi-organ segmentation. 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 10671–10680. https://doi.org/10.1109/ICCV.2019.01077 [57] Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017). Unpaired image-to-image translation using cycle-consistent adversarial networks. 2017 IEEE International Conference on Computer Vision (ICCV), 2242–2251. [58] Zou, Y., Yu, Z., Kumar, B. V. K. V., & Wang, J. (2018). Unsupervised domain adaptation for semantic segmentation via class-balanced self-training. European Conference on Computer Vision. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/95198 | - |
| dc.description.abstract | 醫療影像分割在臨床實踐中是一件相當重要的任務。相較於傳統的電腦視覺技術，透過深度神經網路進行醫療影像分割，我們可以得到更為精準的分割結果。然而，醫療影像資料集仰賴專業人工標註，因此在醫學領域中獲取大量標註的數據集仍然具有相當的挑戰性。為了提升資料的易得性，過去方法透過加入相同模態的不完整標註資料進行訓練，或者其他模態完整標記的資料利用領域自適應 (Domain Adaptation) 來補充原有資料的資訊。若能同時考慮二項限制，則更能資加資料來源的豐富度，而在現有文獻中這個問題依然是研究領域的缺口。本研究因此提出了一種使用不完整標註資料集進行跨模態分割的新架構，這些資料集來自電腦斷層掃描 (CT) 和核磁共振成像 (MRI)。該框架同時解決了兩個關鍵挑戰：模態之間的領域遷移 (Domain Shift) 和目標器官/組織的不完整標註。
本研究介紹了一種兩階段的方法：首先在第一階段透過 CycleGAN 生成另一模態的影像來進行像素級別的對齊，在第二階段則使用原型對比學習 (Prototypical Contrastive Learning) 再次進行特徵級別的對齊。所提出的原型域自適應噪聲對比估計 (Prototypical Domain Adaptive Noise Contrastive Estimation, ProdaNCE) 擴展了傳統的對比學習，適用於領域自適應和部分標註場景。它通過將來自另一模態的原型視為正樣本，而將來自同一模態的原型視為負樣本，鼓勵神經網路對跨模態的標註和未標註類別的表示進行對齊。這種新穎的方法使神經網路能夠學習與領域無關的特徵表示，並利用每種模態的部分標註，實現全面的多器官分割。 我們將提出的架構在一個私有的腹部 CT 和 MRI 資料集上進行評估。該資料集的任務是用於分割皮下脂肪、骨骼肌和內臟脂肪。結果顯示，與現有的領域自適應方法相比，特別是在未標註器官方面，性能更為優越。所提出的方法優於 CycleGAN、CyCADA 和 AdaptSegNet 基準模型，展示了 ProdaNCE 損失在處理領域遷移和不完整標註方面的有效性。 本研究旨在提出了一個具有前景的研究方向，即利用跨模態的不完整標註醫學影像集來產生完整標註的影像分割。通過更有效地利用現有的標註數據，我們的架構減少在臨床實踐中部署深度學習分割模型的標註負擔，為更便捷和精確的醫療影像分析鋪平道路。 | zh_TW |
| dc.description.abstract | Medical image segmentation is a crucial task in clinical practice, but obtaining large annotated datasets across multiple imaging modalities remains challenging. This thesis proposes a novel framework for cross-modality segmentation using partially labeled datasets from computed tomography (CT) and magnetic resonance imaging (MRI). The framework addresses two key challenges simultaneously: domain shift between modalities and incomplete annotations for target organs/tissues.
A two-stage approach is introduced, combining pixel-level alignment via CycleGAN with feature-level alignment using prototypical contrastive learning. The proposed Prototypical Domain Adaptive Noise Contrastive Estimation (ProdaNCE) loss extends traditional contrastive learning by adapting it for domain adaptation and partial labeling scenarios. It encourages the network to align representations of both labeled and unlabeled classes across modalities by treating prototypes from the other modality as positive samples and prototypes from the same modality as negative samples. This novel approach enables the network to learn domain-agnostic feature representations and leverage partial labels from each modality to achieve comprehensive multi-organ segmentation. The framework is evaluated on a private abdominal CT and MRI dataset for segmenting subcutaneous adipose tissue, skeletal muscle, and visceral adipose tissue. Results demonstrate superior performance compared to existing domain adaptation methods, especially for unlabeled organs. The proposed approach outperforms CycleGAN, CyCADA, and AdaptSegNet baselines, showcasing the effectiveness of the ProdaNCE loss in handling both domain shift and incomplete annotations. This work provides a promising direction for utilizing heterogeneous partially labeled medical imaging datasets across modalities. By enabling more effective use of existing annotated data, the framework potentially reduces the annotation burden for deploying deep learning segmentation models in clinical practice, paving the way for more accessible and accurate medical image analysis. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2024-08-30T16:08:15Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2024-08-30T16:08:16Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | Verification Letter from the Oral Examination Committee i
致謝 iii 摘要 v Abstract vii Contents ix List of Figures xi List of Tables xiii Chapter 1 Introduction 1 1.1 Background 1 1.2 Motivation 2 1.3 Research Contribution 4 Chapter 2 Related Work 5 2.1 Multiorgan Segmentation 5 2.1.1 Single-Modality Images 6 2.1.2 Unpaired Multi-Modality Images 6 2.1.3 Incomplete Annotations and Partially Labeled datasets 7 2.2 Contrastive Learning 9 2.2.1 Prototypes and Prototypical Contrastive Learning 10 2.3 Unsupervised Domain Adaptation 12 2.3.1 Pixel-level Distribution Alignment 13 2.3.2 Feature-level Distribution Alignment 15 2.3.3 Pseudo-labeling 18 Chapter 3 Method 21 3.1 Method Overview 21 3.2 Pixel-label Alignment with CycleGAN 21 3.3 Feature Alignment with Contrastive Learning 24 3.3.1 Minimizing ProdaNCE Loss for Domain-Agnostic Clusters 24 3.3.2 Contrastive Learning pairs: Pseudo Labeling 26 3.3.3 Total Objective for feature alignment 26 Chapter 4 Case Studies 29 4.1 Dataset 29 4.2 Implementation Details 30 4.2.1 Network Architecture 30 4.2.2 Training Details 31 4.3 Evaluation Protocal 32 4.3.1 Baselines 32 4.3.2 Metrics 33 4.4 Results 34 4.4.1 Scenario 1CT label = {VAT, TSM} and MRI label = {SAT} 35 4.4.2 Scenario 2CT label = {SAT} and MRI label = {VAT, TSM} 38 4.4.3 Scenario 3CT label = {VAT, TSM, SAT} and No MRI label 41 Chapter 5 Conclusion 45 5.1 Wrap Up 45 5.2 Limitations and Future Research 45 References 47 | - |
| dc.language.iso | en | - |
| dc.subject | 領域自適應 | zh_TW |
| dc.subject | 醫療影像分析 | zh_TW |
| dc.subject | 原型對比學習 | zh_TW |
| dc.subject | 跨模態 | zh_TW |
| dc.subject | 不完整標註 | zh_TW |
| dc.subject | Incomplete Annotations | en |
| dc.subject | Cross-Modality | en |
| dc.subject | Prototypical Contrastive Learning | en |
| dc.subject | Domain Adaptation | en |
| dc.subject | Medical Image Analysis | en |
| dc.title | 不完整標註之跨模態分割模型及其在醫療影像之分析應用 | zh_TW |
| dc.title | Cross-modality Segmentation Model with Incomplete Annotations for Multiorgan Medical Image Analysis | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 112-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 吳志宏;顏宏軒 | zh_TW |
| dc.contributor.oralexamcommittee | Chih-Horng Wu;Hung-Hsuan Yen | en |
| dc.subject.keyword | 醫療影像分析,領域自適應,不完整標註,跨模態,原型對比學習, | zh_TW |
| dc.subject.keyword | Medical Image Analysis,Domain Adaptation,Incomplete Annotations,Cross-Modality,Prototypical Contrastive Learning, | en |
| dc.relation.page | 51 | - |
| dc.identifier.doi | 10.6342/NTU202404141 | - |
| dc.rights.note | 同意授權(限校園內公開) | - |
| dc.date.accepted | 2024-08-14 | - |
| dc.contributor.author-college | 共同教育中心 | - |
| dc.contributor.author-dept | 統計碩士學位學程 | - |
| 顯示於系所單位： | 統計碩士學位學程 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-112-2.pdf 授權僅限NTU校內IP使用（校園外請利用VPN校外連線服務） | 9.92 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。