Unsupervised 3D Lung Segmentation by Leveraging 2D Segment Anything Model

Page view(s)
5
Checked on Apr 13, 2025
Unsupervised 3D Lung Segmentation by Leveraging 2D Segment Anything Model
Title:
Unsupervised 3D Lung Segmentation by Leveraging 2D Segment Anything Model
Journal Title:
2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Publication Date:
17 December 2024
Citation:
Liu, T., Wei, Q., Chen, J., Liu, W., Huang, W., Srivastava, R., Cheng, Z., Zeng, Z., Veeravalli, B., & Yang, X. (2024). Unsupervised 3D Lung Segmentation by Leveraging 2D Segment Anything Model. 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 1–4. https://doi.org/10.1109/embc53108.2024.10782129
Abstract:
Lung segmentation is the first important step for lung nodule detection and lung cancer analysis. Deep neural networks have achieved state-of-the-art for most tasks in medical image analysis, including lung segmentation. However, training a deep learning model requires a large amount of annotated samples, which is not practical in medical imaging. In this study, we make efforts to perform unsupervised lung segmentation on 3D lung CT data by leveraging foundational 2D segment anything model (SAM). The approach utilizes SAM to segment 2D slides and generate 2D masks, then reconstruct multiple 2D masks from the same subject into one 3D mask. In such a way, we can train a 3D lung segmentation model by using the reconstructed 3D masks without the requirement of any ground truth annotations, namely, in an unsupervised manner. The evaluation on LUNA16 dataset shows our proposed unsupervised 3D model achieves comparable results with enhanced stability compared to the supervised one trained with ground truth annotations.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the Agency for Science, Technology and Research - Manufacturing, Trade, and Connectivity Programmatic Funds
Grant Reference no. : M23L7b0021
Description:
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
2694-0604
Files uploaded:

File Size Format Action
embc2024-unsupervised-3d-lung-segmentation.pdf 953.97 KB PDF Request a copy