Abstract
Learning segmentation from noisy labels is an important task for medical image analysis due to the difficulty in acquiring high-quality annotations. Most existing methods neglect the pixel correlation and structural prior in segmentation, often producing noisy predictions around object boundaries. To address this, we adopt a superpixel representation and develop a robust iterative learning strategy that combines noise-aware training of segmentation network and noisy label refinement, both guided by the superpixels. This design enables us to exploit the structural constraints in segmentation labels and effectively mitigate the impact of label noise in learning. Experiments on two benchmarks show that our method outperforms recent state-of-the-art approaches, and achieves superior robustness in a wide range of label noises. Code is available at https://github.com/gaozhitong/SP_guided_Noisy_Label_Seg.
S. Li and Z. Gao—Equal contribution.
This work was supported by Shanghai Science and Technology Program 21010502700 and by the ShanghaiTech-UII Joint Lab.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
Clavicles are particularly small in chest x-ray images. To facilitate fine-grained segmentation and reduce consuming time, we crop their region of interest by statistics on the training set.
- 3.
We also observe that our method outperforms the baseline with 84.26% Dice on the original dataset, likely due to the noise in manual annotations [7].
References
Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P., Süsstrunk, S.: Slic superpixels compared to state-of-the-art superpixel methods. IEEE Trans. Pattern Anal. Mach. Intell. (2012)
Arpit, D., et al.: A closer look at memorization in deep networks. In: ICML (2017)
Ching, T., Himmelstein, D.S., et al.: Opportunities and obstacles for deep learning in biology and medicine. J. Royal Soc. Interface 15, 20170387 (2018)
Daoud, M.I., Atallah, A.A., Awwad, F., Al-Najjar, M., Alazrai, R.: Automatic superpixel-based segmentation method for breast ultrasound images. Expert Systems with Applications (2019)
Ginneken, B., Stegmann, M.B., Loog, M.: Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public database. Medical image analysis (2006)
Gurari, D., Theriault, D., et al.: How to collect segmentations for biomedical images? a benchmark evaluating the performance of experts, crowdsourced non-experts, and algorithms. In: 2015 IEEE Winter Conference on Applications of Computer Vision (2015)
Gutman, D., et al.: Skin lesion analysis toward melanoma detection: a challenge at the 2017 International Symposium on Biomedical Imaging (isbi), Hosted by the International Skin Imaging Collaboration (isic). ISBI (2018)
Han, B., et al.: Co-teaching: robust training of deep neural networks with extremely noisy labels. In: NeurIPS (2018)
Isensee, F., Jaeger, P.F., Kohl, S.A.A., Petersen, J., Maier-Hein, K.: nnu-net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods (2020)
Jiang, L., Zhou, Z., Leung, T., Li, L.J., Fei-Fei, L.: Mentornet: learning data-driven curriculum for very deep neural networks on corrupted labels. ArXiv (2018)
Kohli, M., Summers, R., Geis, J.: Medical image data and datasets in the era of machine learning–whitepaper from the 2016 c-mimi meeting dataset session. J. Digital Imaging 30, 392–399 (2017)
Li, Y., Jia, L., Wang, Z., Qian, Y., Qiao, H.: Un-supervised and semi-supervised hand segmentation in egocentric images with noisy label learning. Neurocomputing (2019)
Litjens, G., et al.: A survey on deep learning in medical image analysis. Medical Image Analysis (2017)
Malach, E., Shalev-Shwartz, S.: Decoupling “when to update” from “how to update". In: NIPS (2017)
Mirikharaji, Z., Yan, Y., Hamarneh, G.: Learning to segment skin lesions from noisy annotations. ArXiv (2019)
Northcutt, C.G., Jiang, L., Chuang, I.: Confident learning: Estimating uncertainty in dataset labels. ArXiv (2019)
Qin, W., et al.: Superpixel-based and boundary-sensitive convolutional neural network for automated liver segmentation. Physics in Medicine & Biology (2018)
Ren, M., Zeng, W., Yang, B., Urtasun, R.: Learning to reweight examples for robust deep learning. In: ICML (2018)
Shen, D., Wu, G., Suk, H.: Deep learning in medical image analysis. Annual review of biomedical engineering (2017)
Shiraishi, J., et al.: Development of a digital image database for chest radiographs with and without a lung nodule: receiver operating characteristic analysis of radiologists’ detection of pulmonary nodules. AJR, American journal of roentgenology (2000)
Tian, Z., Liu, L., Zhang, Z., Fei, B.: Superpixel-based segmentation for 3d prostate mr images. IEEE Trans. Med. Imaging 35, 791–801 (2015)
Wei, H., Feng, L., Chen, X., An, B.: Combating noisy labels by agreement: a joint training method with co-regularization. CVPR (2020)
Xue, C., Deng, Q., Li, X., Dou, Q., Heng, P.: Cascaded robust learning at imperfect labels for chest x-ray segmentation. In: MICCAI (2020)
Zhang, M., et al.: Characterizing label errors: confident learning for noisy-labeled image segmentation. In: MICCAI (2020)
Zhang, T., Yu, L., Hu, N., Lv, S., Gu, S.: Robust medical image segmentation from non-expert annotations with tri-network. In: MICCAI (2020)
Zhu, H., Shi, J., Wu, J.: Pick-and-learn: automatic quality evaluation for noisy-labeled image segmentation. In: MICCAI (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Li, S., Gao, Z., He, X. (2021). Superpixel-Guided Iterative Learning from Noisy Labels for Medical Image Segmentation. In: de Bruijne, M., et al. Medical Image Computing and Computer Assisted Intervention – MICCAI 2021. MICCAI 2021. Lecture Notes in Computer Science(), vol 12901. Springer, Cham. https://doi.org/10.1007/978-3-030-87193-2_50
Download citation
DOI: https://doi.org/10.1007/978-3-030-87193-2_50
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87192-5
Online ISBN: 978-3-030-87193-2
eBook Packages: Computer ScienceComputer Science (R0)
