Journal of the Korea Computer Graphics Society (한국컴퓨터그래픽스학회논문지)

Korea Computer Graphics Society (한국컴퓨터그래픽스학회)
권호 표지
DOI QR코드

DOI QR Code

Enhanced Lung Cancer Segmentation with Deep Supervision and Hybrid Lesion Focal Loss in Chest CT Images

흉부 CT 영상에서 심층 감독 및 하이브리드 병변 초점 손실 함수를 활용한 폐암 분할 개선

  • Min Jin Lee (Department of Software Convergence, Seoul Women's University) ;
  • Yoon-Seon Oh (Department of Software Convergence, Seoul Women's University) ;
  • Helen Hong (Department of Software Convergence, Seoul Women's University)
  • 이민진 (서울여자대학교 소프트웨어융합학과) ;
  • 오윤선 (서울여자대학교 소프트웨어융합학과) ;
  • 홍헬렌 (서울여자대학교 소프트웨어융합학과)
  • Received : 2023.10.19
  • Accepted : 2023.12.27
  • Published : 2024.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

Lung cancer segmentation in chest CT images is challenging due to the varying sizes of tumors and the presence of surrounding structures with similar intensity values. To address these issues, we propose a lung cancer segmentation network that incorporates deep supervision and utilizes UNet3+ as the backbone. Additionally, we propose a hybrid lesion focal loss function comprising three components: pixel-based, region-based, and shape-based, which allows us to focus on the smaller tumor regions relative to the background and consider shape information for handling ambiguous boundaries. We validate our proposed method through comparative experiments with UNet and UNet3+ and demonstrate that our proposed method achieves superior performance in terms of Dice Similarity Coefficient (DSC) for tumors of all sizes.

폐암은 크기가 다양하고 유사한 밝기값을 갖는 주변 구조물이 존재하기 때문에 흉부 CT 영상에서 폐암을 정확하게 분할하는 것이 어렵다. 이러한 문제를 해결하기 위해 본 논문에서는 심층 감독을 포함하고 UNet3+를 백본으로 사용하는 폐암 분할 네트워크를 제안한다. 또한, 픽셀 기반, 영역 기반 및 형태 기반의 3가지 구성 요소로 이루어진 하이브리드 병변 초점 손실함수를 제안한다. 이를 통해 배경에 비해 작은 영역을 차지하는 폐암 부분에 집중하고, 불명확한 경계를 처리하는데 도움이 되는 형태 정보를 고려할 수 있다. 제안 방법을 UNet 및 UNet3+와 비교 실험을 통해 검증하였고, 제안 방법은 모든 폐암 크기에서 DSC 측면에서 가장 우수한 성능을 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(2024)을 받아 수행되었으며 이에 감사드립니다.

References

  1. H. Sung, J. Ferlay, RL. Siegel, M. Laversanne, I. Soerjomataram, A. Jemal and F. Bray, Global cancer statistics 2020: LOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 71(3), pp. 209-249, 2021  https://doi.org/10.3322/caac.21660
  2. V. Greenberg, I. Lazarev, Y. Frank, J. Dudnik, S. Ariad and I. Shelef, "Semi-automatic volumetric measurement of response to chemotherapy in lung cancer patients: How wrong are we using RECIST?," Lung Cancer,108, pp. 90-95, 2017  https://doi.org/10.1016/j.lungcan.2017.02.017
  3. S. Hossain, S. Najeeb, A. Shahriyar, Z.R. Abdullah and M.A. Haque, "A pipeline for lung tumor detection and segmentation from CT scans using dilated convolutional neural networks," ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal, pp.1348-1352, 2019. 
  4. H.J. Tuba, M.A. Ullah and F.K. Badhon," Lung Tumor Segmentation and Detection using U-Net with Dilated Convolutions," 5th International Conference on Electrical Information and Communication Technology (EICT), pp. 1-5, 2021. 
  5. F. Zhang, Q. Wang and H. Li, "Automatic segmentation of the grosstarget volume in non-small cell lung cancer using a modified version of ResNet," Technology in Cancer Research & Treatment, 19, 2020. 
  6. J. Lee, J. Jung, H. Hong and B Kim, "Automatic Segmentation of Lung Cancer in Chest CT Images through Capsule Network-based Dual-Window Ensemble Learning," Journal of the Korea Institute of Information Scientists and Engineers, 48(8), pp.905-912, 2021  https://doi.org/10.5626/JOK.2021.48.8.905
  7. J.G. Jin, Y.J. Kim and KG Kim, "A Study on Lung Cancer Segmentation Algorithm using Weighted Integration Loss on Volumetric Chest CT Image," Journal of Korea Multimedia Society, 23(5), pp. 625-632, 2020.  https://doi.org/10.9717/KMMS.2020.23.5.625
  8. H. Huang, L. Lin, R. Tong, H. Hu, Q. Zhang, Y. Iwamoto, X. Han, Y.W. Chen, and J. Wu, "UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation," IEEE international conference on acoustics, speech and signal processing (ICASSP), pp. 1055-1059, 2020. 
  9. O. Ronneberger, P. Fischer, and T. Brox," U-net: Convolutional networks for biomedical image segmentation," Medical Image Computing and Computer-Assisted Intervention (MICCAI ), pp. 234-241, 2015