Abstract
In diagnosing thyroid nodules, weakly supervised semantic segmentation methods alleviate the dependence on pixel-level segmentation labels. However, existing methods underutilize the image information of thyroid nodules under the supervision of image-level labels, which is reflected in intra-image and inter-image. Firstly, the imaging quality of ultrasound images is poor, making the model hardly mining semantic information, leading to misclassification in background. We propose an equivariant attention mechanism to enhance the nodules and background, enabling the model to extract more accurate semantic information intra-image. Secondly, thyroid nodules have fine-grained properties such as cystic, solid, and calcified, existing methods ignore the semantic information between different fine-grained nodules, making it difficult for the model to learn a comprehensive feature representation. We propose to collect the features of nodules and background in the dataset through a memory pool and provide the connections between these features through semantic sharing and contrast. Experiments on the TUI dataset show that our method significantly outperforms existing methods, with mIoU and Dice scores improving to 58.0% and 73.5%.
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
References
Ahn, J., Kwak, S.: Learning pixel-level semantic affinity with image-level supervision for weakly supervised semantic segmentation. In: CVPR, pp. 4981–4990. IEEE (2018)
Jo, S., Yu, I.J.: Puzzle-CAM: improved localization via matching partial and full features. In: ICIP, pp. 639–643. IEEE (2021)
Kolesnikov, A., Lampert, C.H.: Seed, expand and constrain: three principles for weakly-supervised image segmentation. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 695–711. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_42
Lee, J., Kim, E., Lee, S., Lee, J., Yoon, S.: FickleNet: weakly and semi-supervised semantic image segmentation using stochastic inference. In: CVPR, pp. 5267–5276 (2019)
Li, X., Zhou, T., Li, J., Zhou, Y., Zhang, Z.: Group-wise semantic mining for weakly supervised semantic segmentation. In: AAAI, vol. 35, pp. 1984–1992 (2021)
Stammes, E., Runia, T.F., Hofmann, M.: Find it if you can: end to-end adversarial erasing for weakly-supervised semantic segmentation. In: ICDIP, pp. 610–619. SPIE (2021)
Sun, G., Wang, W., Dai, J., Van Gool, L.: Mining cross-image semantics for weakly supervised semantic segmentation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12347, pp. 347–365. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58536-5_21
Wang, X., Liu, S., Ma, H., Yang, M.H.: Weakly-supervised semantic segmentation by iterative affinity learning. Int. J. Comput. Vis. 128, 1736–1749 (2020)
Wang, X., You, S., Li, X., Ma, H.: Weakly-supervised semantic segmentation by iteratively mining common object features. In: CVPR, pp. 1354–1362 (2018)
Wang, Y., Zhang, J., Kan, M.: Self-supervised equivariant attention mechanism for weakly supervised semantic segmentation. In: CVPR, pp. 12275–12284 (2020)
Wei, Y., Feng, J., Liang, X.: Object region mining with adversarial erasing: a simple classification to semantic segmentation approach. In: CVPR, pp. 1568–1576 (2017)
Yu, M., et al.: Adaptive soft erasure with edge self-attention for weakly supervised semantic segmentation: thyroid ultrasound image case study. Comput. Biol. Med. 144, 105347 (2022)
Yu, M., Han, M., Li, X.: SSE: scale-adaptive soft erase weakly supervised segmentation network for thyroid ultrasound images. In: BIBM, pp. 1615–1618. IEEE (2021)
Zhang, F., Gu, C., Zhang, C., Dai, Y.: Complementary patch for weakly supervised semantic segmentation. In: ICCV, pp. 7242–7251 (2021)
Zhang, M., Chen, Y.: PPO-CPQ: a privacy-preserving optimization of clinical pathway query for e-healthcare systems. IEEE Internet Things J. 7(10), 10660–10672 (2020)
Zhang, X., Peng, Z., Zhu, P.: Adaptive affinity loss and erroneous pseudo-label refinement for weakly supervised semantic segmentation. In: ACM Multimedia, pp. 5463–5472 (2021)
Zhang, X., Wei, Y., Feng, J., Yang, Y., Huang, T.S.: Adversarial complementary learning for weakly supervised object localization. In: CVPR, pp. 1325–1334 (2018)
Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: CVPR, pp. 2921–2929 (2016)
Zhou, T., Zhang, M., Zhao, F., Li, J.: Regional semantic contrast and aggregation for weakly supervised semantic segmentation. In: CVPR, pp. 4299–4309 (2022)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Gao, J., Yan, S., Fu, X., Liu, Z., Yu, R., Yu, M. (2023). IntrNet: Weakly Supervised Segmentation of Thyroid Nodules Based on Intra-image and Inter-image Semantic Information. In: Huang, DS., Premaratne, P., Jin, B., Qu, B., Jo, KH., Hussain, A. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2023. Lecture Notes in Computer Science, vol 14087. Springer, Singapore. https://doi.org/10.1007/978-981-99-4742-3_60
Download citation
DOI: https://doi.org/10.1007/978-981-99-4742-3_60
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4741-6
Online ISBN: 978-981-99-4742-3
eBook Packages: Computer ScienceComputer Science (R0)