肺癌具有很高的死亡率，是全球主要的公共衛生問題。 它是癌症相關 死亡的主要原因，預計在未來會變得更加普遍。 肺癌篩查很重要，但這 個過程可能很耗時，而且容易出錯。 交互式分割是一種可以幫助醫生在 CT 圖像中快速、準確地標記結節的技術，是確定惡性概率和做出治療決 策的重要步驟。
隨著當前的發展趨勢，Transformer 已廣泛應用於醫學成像以提高分 割性能。 我們推薦的 3D 肺結節交互式分割方法使用改進的 iSegformer 進行 2D 手動分割和時空對應網絡來傳播 3D 圖像。 我們添加了幾個 組件來增強原始 iSegformer 模型的性能。 這些包括添加邊界損失函數 以增加網絡對結節邊界的關注，實施多尺度權重模塊以根據結節的大小 調整特徵的權重和塊的輸出，以及添加第一注意模塊以 改善全局信息並 提高首次點擊的準確性。
我們將該系統應用於國立成功大學醫院的肺癌數據庫，其中包含手術 患者和體檢患者兩個數據集。 我們使用修改後的 iSegformer 的結果在 肺結節數據集上取得了很好的結果，mNoC 分別為 85% 和 90% IoU， 分別為 9.1 和 12.4。

The lung cancer has a high mortality rate and is a major public health concern worldwide. It is the leading cause of cancer-related deaths and is expected to become even more prevalent in the future. Screening for lung cancer is important, but the process can be time-consuming and prone to errors. Interactive segmentation is a technique that can assist doctors in quickly and accurately labeling nodules in CT images, which is an important step in determining the probability of malignancy and making treatment decisions.
With the current development trend, transformers have been widely applied in medical imaging to increase segmentation performance. The method we recommend for interactive segmentation for 3D Lung Nodule uses Modified iSegformer for 2D manual segmentation and Space-time corresponding Network to propagate the 3D image. We have added several components to enhance the performance of the original iSegformer model. These include adding a boundary loss function to increase the network's focus on the nodule's boundary, implementing a Multi-scale weight module to adjust the features' weights and the blocks' output based on the size of the nodule, and adding a First Attention Module to improve global information and increase the accuracy of the first click.
We applied the system to the lung cancer database of the National Cheng Kung University Hospital, which contains two datasets of surgical patients and medical exam patients. Our results with Modified iSegformer have good results on Lung Nodule Dataset with mNoC at 85% and 90% IoU at 9.1 and 12.4 respectively.

[1] “Cancer Today.” https://gco.iarc.fr/today/home (accessed Jan. 13, 2023).
[2] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D. M. Parkin, D. Forman, and F. Bray, “Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012,” Int J Cancer, vol. 136, no. 5, pp. E359–E386, Mar. 2015, doi: 10.1002/ijc.29210.
[3] H. Ramadan, C. Lachqar, and H. Tairi, “A survey of recent interactive image segmentation methods,” Comput Vis Media (Beijing), vol. 6, no. 4, pp. 355–384, Dec. 2020, doi: 10.1007/S41095-020-0177-5.
[4] K. K. Maninis, S. Caelles, J. Pont-Tuset, and L. van Gool, “Deep Extreme Cut: From Extreme Points to Object Segmentation,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 616–625, Nov. 2017, doi: 10.48550/arxiv.1711.09081.
[5] Z. Lin, Z. Zhang, L.-Z. Chen, M.-M. Cheng, and S.-P. Lu, “Interactive Image Segmentation With First Click Attention,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 13339–13348, 2020, Accessed: Jan. 27, 2023. [Online]. Available: http://mmcheng.net/fclick/
[6] W. Y. Ma and B. S. Manjunath, “Edge flow: A framework of boundary detection and image segmentation,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 744–749, 1997, doi: 10.1109/CVPR.1997.609409.
[7] N. Xu, B. Price, S. Cohen, J. Yang, and T. Huang, “Deep GrabCut for Object Selection,” British Machine Vision Conference 2017 (BMVC 2017), Jul. 2017, doi: 10.48550/arxiv.1707.00243.
[8] A. Deshpande, P. Dahikar, and P. Agrawal, “An experiment with random walks and GrabCut in one cut interactive image segmentation techniques on MRI images,” Advances in Intelligent Systems and Computing, vol. 1108 AISC, pp. 993–1008, 2020, doi: 10.1007/978-3-030-37218-7_105.
[9] H. Yu, Y. Zhou, H. Qian, M. Xian, and S. Wang, “Loosecut: Interactive image segmentation with loosely bounded boxes,” Proceedings - International Conference on Image Processing (ICIP), vol. 2017-September, pp. 3335–3339, Feb. 2018, doi: 10.1109/ICIP.2017.8296900.
[10] O. Juan and Y. Boykov, “Active graph cuts,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 1023–1029, 2006, doi: 10.1109/CVPR.2006.47.
[11] L. Grady, “Random walks for image segmentation,” IEEE Trans Pattern Anal Mach Intell, vol. 28, no. 11, pp. 1768–1783, 2006, doi: 10.1109/TPAMI.2006.233.
[12] J. Ning, L. Zhang, D. Zhang, and C. Wu, “Interactive image segmentation by maximal similarity based region merging,” Pattern Recognition, vol. 43, no. 2, pp. 445–456, Feb. 2010, doi: 10.1016/J.PATCOG.2009.03.004.
[13] K. Sofiiuk, I. Petrov, O. Barinova, and A. Konushin, “F-BRS: Rethinking Backpropagating Refinement for Interactive Segmentation,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 8623–8632, 2020, Accessed: Jan. 27, 2023. [Online]. Available: https://github.com/saic-vul/
[14] X. Chen, Z. Zhao, Y. Zhang, M. Duan, D. Qi, and H. Zhao, “FocalClick: Towards Practical Interactive Image Segmentation,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1300–1309, 2022.
[15] L. C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,” IEEE Trans Pattern Anal Mach Intell, vol. 40, no. 4, pp. 834–848, Jun. 2016, doi: 10.48550/arxiv.1606.00915.
[16] D. C. Liu and J. Nocedal, “On the limited memory BFGS method for large scale optimization,” Math Program, vol. 45, no. 1–3, pp. 503–528, Aug. 1989, doi: 10.1007/BF01589116/METRICS.
[17] Y. Wang, B. Liang, M. Ding, and J. Li, “Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery,” Remote Sensing 2019, Vol. 11, Page 20, vol. 11, no. 1, p. 20, Dec. 2018, doi: 10.3390/RS11010020.
[18] Y. Yuan, X. Chen, and J. Wang, “Object-Contextual Representations for Semantic Segmentation,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), vol. 12351 LNCS, pp. 173–190, 2020, doi: 10.1007/978-3-030-58539-6_11.
[19] Q. Liu, Z. Xu, Y. Jiao, and M. Niethammer, “iSegFormer: Interactive Segmentation via Transformers with Application to 3D Knee MR Images,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), vol. 13435 LNCS, pp. 464–474, 2022, doi: 10.1007/978-3-031-16443-9_45.
[20] Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, and B. Gou, “Swin Transformer: Hierarchical Vision Transformer using Shifted Windows,” Proceedings of the IEEE International Conference on Computer Vision, pp. 9992–10002, Mar. 2021, doi: 10.48550/arxiv.2103.14030.
[21] H. K. Cheng, Y.-W. Tai, and C.-K. Tang, “Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation,” Adv Neural Inf Process Syst, vol. 34, pp. 11781–11794, Dec. 2021, Accessed: Jan. 27, 2023. [Online]. Available: https://github.com/hkchengrex/STCN
[22] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollar, “Focal Loss for Dense Object Detection,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2980–2988, Oct. 2017.
[23] R. Yip, C. I. Henschke, D. F. Yankelevitz, and J. P. Smith, “Screening for lung cancer: Alternative definitions of positive test result based on the national lung screening trial and international early lung cancer action program databases,” Radiology, vol. 273, no. 2, pp. 591–596, Nov. 2014, doi: 10.1148/RADIOL.14132950.
[24] D. Karimi and S. E. Salcudean, “Reducing the Hausdorff Distance in Medical Image Segmentation with Convolutional Neural Networks,” IEEE Trans Med Imaging, vol. 39, no. 2, pp. 499–513, Feb. 2020, doi: 10.1109/TMI.2019.2930068.
[25] J. Hu, L. Shen, and G. Sun, “Squeeze-and-Excitation Networks,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 7132–7141, Jun. 2018, Accessed: Jan. 27, 2023. [Online]. Available: http://image-net.org/challenges/LSVRC/2017/results
[26] A. A. Taha and A. Hanbury, “Metrics for evaluating 3D medical image segmentation: Analysis, selection, and tool,” BMC Med Imaging, vol. 15, no. 1, pp. 1–28, Aug. 2015, doi: 10.1186/S12880-015-0068-X/TABLES/5.