Journal of the Korea Computer Graphics Society (한국컴퓨터그래픽스학회논문지)

Korea Computer Graphics Society (한국컴퓨터그래픽스학회)
권호 표지
DOI QR코드

DOI QR Code

Kidney Tumor Segmentation through Semi-supervised Learning Based on Mean Teacher Using Kidney Local Guided Map in Abdominal CT Images

복부 CT 영상에서 신장 로컬 가이드 맵을 활용한 평균-교사 모델 기반의 준지도학습을 통한 신장 종양 분할

  • Heeyoung Jeong (Department of Software Convergence, Seoul Women's University) ;
  • Hyeonjin Kim (Department of Software Convergence, Seoul Women's University) ;
  • Helen Hong (Department of Software Convergence, Seoul Women's University)
  • 정희영 (서울여자대학교 소프트웨어융합학과) ;
  • 김현진 (서울여자대학교 소프트웨어융합학과) ;
  • 홍헬렌 (서울여자대학교 소프트웨어융합학과)
  • Received : 2023.10.25
  • Accepted : 2023.11.24
  • Published : 2023.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

Accurate segmentation of the kidney tumor is necessary to identify shape, location and safety margin of tumor in abdominal CT images for surgical planning before renal partial nephrectomy. However, kidney tumor segmentation is challenging task due to the various sizes and locations of the tumor for each patient and signal intensity similarity to surrounding organs such as intestine and spleen. In this paper, we propose a semi-supervised learning-based mean teacher network that utilizes both labeled and unlabeled data using a kidney local guided map including kidney local information to segment small-sized kidney tumors occurring at various locations in the kidney, and analyze the performance according to the kidney tumor size. As a result of the study, the proposed method showed an F1-score of 75.24% by considering local information of the kidney using a kidney local guide map to locate the tumor existing around the kidney. In particular, under-segmentation of small-sized tumors which are difficult to segment was improved, and showed a 13.9%p higher F1-score even though it used a smaller amount of labeled data than nnU-Net.

부분신장절제술 전 수술 계획을 세우기 위해서는 신장 종양의 위치, 형태 및 수술 시 안전 마진 파악이 중요하므로 신장 종양을 정확히 분할하는 것이 필요하다. 그러나 신장 종양은 환자마다 위치 및 크기가 다양하며 소장과 비장 같은 주변 장기와 형태와 밝기값이 유사하여 신장 종양을 분할하는 것에 어려움이 있다. 본 논문에서는 레이블이 있는 데이터와 없는 데이터를 함께 사용하는 준지도학습 방법 중 하나인 평균-교사모델을 활용하여 신장의 여러 위치에서 발생하는 작은 크기의 신장 종양을 분할하기 위해 신장 위치 정보를 가지는 신장 로컬 가이드 맵을 이용해 신장 종양에 집중하는 평균-교사 네트워크를 제안하고, 신장 종양의 크기에 따른 성능을 분석한다. 실험 결과, 제안 방법은 신장 주변에 존재하는 종양의 위치를 찾기 위해 신장 로컬 가이드 맵을 사용하여 신장의 국소 정보를 고려함으로써 75.24%의 F1-score를 보였다. 특히 분할이 어려운 작은 크기의 종양에 대한 과소분할을 개선하였으며 nnU-Net보다 적은 양의 레이블 데이터를 사용하여도 13.9% 높은 F1-score를 보였다.

Keywords

Acknowledgement

본 연구는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원(No. RS-2023-00207947), 보건복지부의 재원으로 한국 보건산업진흥원의 보건의료기술연구개발사업 지원(HI22C1496) 및 서울여자대학교 학술연구비의 지원(2023-0112)을 받아 수행되었으며 이에 감사드립니다.

References

  1. H. Sung, J. Ferlay, R. L. Siegel, M. Laversanne, I. Soerjo mataram, I. Soerjomataram, A. Jemal, and F. Bray, "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries," A Cancer Journal for Clinicians, vol. 71, no. 3, pp. 209-249, 2021.  https://doi.org/10.3322/caac.21660
  2. Z. Du, W. Chen, Q. Xia, O. Shi, and Q. Chen, "Trends and projections of kidney cancer incidence at the global and national levels, 1990-2030: A Bayesian age-period-cohort modeling study," Biomarker Research, vol. 16, pp. 8, 2020. 
  3. S. Tangal, K. Onal, M. Yigman and A.H. Haliloglu, "Relation of neutrophil lymphocyte ratio with tumor characteristics in localized kidney tumors." The New Journal of Urology, vol. 13, no. 1, pp. 12-15, 2018. 
  4. M. Sun, F. Abdollah, M. Bianchi, Q. D. Trinh, C. Jeldres, R. Thuret, Z. Tian, S.F. Shariat, F. Montorsi, P. Perrotte, and P.I. Karakiewicz, "Treatment management of small renal masses in the 21st century: A paradigm shift." Annals of Surgial. Oncology, vol. 19, pp. 2380-2387, 2012.  https://doi.org/10.1245/s10434-012-2247-0
  5. G. Yang, J. Gu, Y. Chen, W. Liu, L. Tang, H. Shu, and C. Toumoulin, "Automatic kidney segmentation in ct images based on multi-atlas image registration." in Proc, of the 36th Annual Int. Conf, of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, pp. 5538-5541, 2014. 
  6. D.K. Kim, Y. Jang, J. Lee, H. Hong, K.H. Kim, T.Y. Shin, D.C. Jung, Y.D. Choi, and K.H. Rha, "Two-year Analys is for Predicting Renal Function and Contralateral Hypertrophy after Robot-assisted Partial Nephrectomy: A Three-dimensional Segmentation Technology Study." International Journal of Urology, Vol. 22, pp. 1105-1111, 2015.  https://doi.org/10.1111/iju.12913
  7. J. Lee, H. Hong, K.H. Rha, "Automatic Segmentation of Renal Parenchyma using Graph-cuts with Shape Constraint based on Multi-probabilistic Atlas in Abdominal CT Images" Journal of the Korea Computer Graphics Society, 22(4) 11-19. doi: 10.15701/kcgs.2016.22.4.11, 2016. 
  8. H. Wei, Q. Wang, W. Zhao, M. Zhang, K. Yuan, and Z. Li, "Two-phase Framework for Automatic Kidney and Kidney Tumor Segmentation.'' 2019 Kidney Tumor Segmentation Challenge: KiTS19, 2019. 
  9. X. Xie, L. Li, S. Lian, S. Chen, and Z. Luo, "SERU: A cascaded SE-ResNeXT U-Net for kidney and tumor segmentation." Concurrency and Computation: Practice and Experience Volume 32, Issue 14, 2020 
  10. J. Wen, Z. Li, Z. Shen, Y. Zheng, and S. Zheng, "Squeeze-and-Excitation Encoder-Decoder Network for Kidney and Kidney Tumor Segmentation in CT images." Kidney and Kidney Tumor Segmentation. KiTS 2021. Lecture Notes in Computer Science, vol 13168. Springer, Cham. 2022. 
  11. F. Isensee, and K.H. Maier-Hein, "An attempt at beating the 3D U-Net." ArXiv abs/1908.02182, 2019. 
  12. X. Hou, C. Xie, F. Li, and Y. Nan, "Cascaded Semantic Segmentation for Kidney and Tumor." 10.24926/548719.002., 2019. 
  13. J. MA, "Solution to the Kidney Tumor Segmentation Challenge 2019." 10.24926/548719.005, 2019. 
  14. M.J. Willemink, W.A. Koszek, C. Hardell, J. Wu, D. Fleischmann, H. Harvey, L.R. Folio, R.M. Summers, D.L. Rub in, and M.P. Lungren, "Preparing medical imaging data for machine learning." Radiology 295, 2020. 
  15. S.Y. Huang, W.L. Hsu, R.J. Hsu, and D.W. Liu, "Fully Convolutional Network for the Semantic Segmentation of Medical Images: A Survey." Diagnostics 12, 2022. 
  16. Z. Yang, P. Xu, Y. Yang, and B.K. Bao, "A Densely Connected Network Based on U-Net for Medical Image Segmentation." ACM Trans Multimed Comput Commun Appl 17, 2021. 
  17. L. Liu, J.M. Wolterink, C. Brune, and R.N.J. Veldhuis, "Anatomy-aided deep learning for medical image segmentation: A review." Phys Med Biol 66, 2021. 
  18. D. Karimi, S.D. Vasylechko, and A. Gholipour. "Convolution-Free Medical Image Segmentation Using Transformers." Leet Notes Comput Sci (Including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 12901, 2021. 
  19. Y. Wang, Y. Zhang, J. Tian, C. Zhong, Z. Shi, Y. Zhang, and Z. He, "Double-Uncertainty Weighted Method for Semi-supervised Learning." International Conference on Medical Image Computing and Computer-Assisted Intervention, 2020. 
  20. J. Wang, and T. Lukasiewicz, "Rethinking Bayesian Deep Learning Methods for Semi-Supervised Volumetric Medical Image Segmentation," in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA pp. 182-190, 2022. 
  21. Y. Tang, S. Wang, Y. Qu, Z. Cui, and W. Zhang, "Consistency and adversarial semi-supervised learning for medical image segmentation." Computers in Biology and Medicine, Volume 161, 2023. 
  22. L. Hu, J. Li, X. Peng, J. Xiao, B. Zhan, C. Zu, X. Wu, J. Zhou, and Y. Wang, "Semi-supervised NPC segmentation with uncertainty and attention guided consistency." Knowledge-Based Systems, Volume 239, 2022. 
  23. A. Tarvainen, and H. Valpola, "Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results." Adv Neural Inf Process Syst 2017-December, 2017. 
  24. O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," International Conference on Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015, Vol.9351, pp. 234-241, 2015. 
  25. F. Isensee, J. Petersen, A. Klein, D. Zimmerer, P. F.Jaeger, S. Kohl, J. Wasserthal, G. Koehler, T. Norajitra, S. Wirkert, and K.H. Maier-Hein, "nnU-Net: Self-adapting framework for u-net-based medical image segmentation." arXiv preprint arXiv:1809.10486, 2018. 
  26. E. Yang, C. K. Kim, Y. Guan, B. Koo, and J. Kim, "3D multi-scale residual fully convolutional neural network for segmentation of extremely large-sized kidney tumor." Computer Methods and Programs in Biomedicine, Volume 215, 2022. 
  27. R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, "Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization," 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, pp. 618-626, 2017.