DOI QR코드

DOI QR Code

Automatic Meniscus Segmentation from Knee MR Images using Multi-atlas-based Locally-weighted Voting and Patch-based Edge Feature Classification

무릎 MR 영상에서 다중 아틀라스 기반 지역적 가중 투표 및 패치 기반 윤곽선 특징 분류를 통한 반월상 연골 자동 분할

  • Kim, SoonBeen (Department of Software Convergence, Seoul Women's University) ;
  • Kim, Hyeonjin (Department of Software Convergence, Seoul Women's University) ;
  • Hong, Helen (Department of Software Convergence, Seoul Women's University) ;
  • Wang, Joon Ho (Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine)
  • 김순빈 (서울여자대학교 소프트웨어융합학과) ;
  • 김현진 (서울여자대학교 소프트웨어융합학과) ;
  • 홍헬렌 (서울여자대학교 소프트웨어융합학과) ;
  • 왕준호 (성균관대학교 의과대학 삼성서울병원 정형외과)
  • Received : 2018.08.13
  • Accepted : 2018.08.27
  • Published : 2018.09.01

Abstract

In this paper, we propose an automatic segmentation method of meniscus in knee MR images by automatic meniscus localization, multi-atlas-based locally-weighted voting, and patch-based edge feature classification. First, after segmenting the bone and knee articular cartilage, the volume of interest of the meniscus is automatically localized. Second, the meniscus is segmented by multi-atlas-based locally-weighted voting taking into account the weights of shape and intensity distribution in the volume of interest of the meniscus. Finally, to remove leakage to the collateral ligaments with similar intensity, meniscus is refined using patch-based edge feature classification considering shape and distance weights. Dice similarity coefficient between proposed method and manual segmentation were 80.13% of medial meniscus and 80.81 % for lateral meniscus, and showed better results of 7.25% for medial meniscus and 1.31% for lateral meniscus compared to the multi-atlas-based locally-weighted voting.

본 논문에서는 무릎 MR 영상에서 반월상 연골의 자동 위치화, 다중 아틀라스 기반 지역적 가중 투표를 통한 반월상 연골 분할 및 패치 기반 윤곽선 특징 분류를 통한 반월상 연골 자동 분할 방법을 제안한다. 첫째, 뼈와 무릎 관절 연골을 분할한 후 이를 이용하여 반월상 연골의 관심볼륨영역을 자동 위치화한다. 둘째, 반월상 연골의 관심볼륨영역에서 형상 및 밝기값 분포 가중치를 고려한 다중 아틀라스 기반 지역적 가중 투표를 통해 반월상 연골을 분할한다. 셋째, 밝기값이 유사한 측부 인대로의 누출을 제거하기 위해 형상 및 거리 가중치를 고려한 패치 기반 윤곽선 특징 분류를 통해 반월상 연골 분함을 개선한다. 제안 방법을 통한 분할 결과와 수동 분할 결과 간 다이스 유사계수는 내측 반월상 연골은 80.13%, 외측 반월상 연골은 80.81%를 보였으며 다중 아틀라스 기반 지역적 가중투표를 통한 분할 방법과 비교하여 내 측 및 외측 반월상 연 골 각각 7.25%, 1.31% 향상되었다.

Keywords

References

  1. D. J. Hunter, Y. Q. Zhang, J. B. Niu, X. Tu, S. Amin, M. Clancy, D. T. Felson, "The association of meniscal pathologic changes with cartilage loss in symptomatic knee osteoarthritis," Arthritis & Rheumatism, Vol. 54, No. 3, pp. 795-801, 2006. https://doi.org/10.1002/art.21724
  2. M. Roth, W. Wirth, K. Emmanuel, A. G. Culvenor, F. Eckstein, "The contribution of 3D quantitative meniscal and cartilage measures to variation in normal radiographic joint space width- Data from the Osteoarthritis Initiative healthy reference cohort," European journal of radiology, Vol. 87, pp. 90-98, 2017. https://doi.org/10.1016/j.ejrad.2016.12.009
  3. M. S. Swanson, J. W. Prescott, T. M. Best, K. Powell, R. D. Jackson, F. Haq, and M. N. Gurcan, "Semi-automated segmentation to assess the lateral meniscus in nonnal and osteoarthritic knees," Osteoarthritis and cartilage, Vol. 18, No. 3, pp. 344-353, 2010. https://doi.org/10.1016/j.joca.2009.10.004
  4. M. S. M. Swamy, and M. S. Holi, "Knee Joint Menisci Visualization and Detection of Tears by Image Processing," Computing, Communication and Applications, 2012 International Conference on. IEEE, pp. 1-5, 2012.
  5. J. Fripp, P. Bourgeat, C. Engstrom, S. Ourselin, S. Crozier, and O. Salvado, "Automated Segmentation of the Menisci from MR Images, " Biomedical Imaging, pp. 210-513, 2009.
  6. M. Kim, J. Yoo, and H. Hong, "Automatic Segmentation of the meniscus based on Active Shape Model in MR Images through Interpolated Shape Information," Journal of Korean Institute of Information Scientists and Engineers, Vol. 16, No. 11 , pp. 1096-1100, 2010.
  7. A. Paproki, C. Engstrom, S. Chandra, A. Neubert, J. Fripp, and S. Crozier, "Automated segmentation and analysis of normal and osteoarthritic knee menisci from magnetic resonance unages - data from the Osteoarthritis Initiative," Osteoarthritis Cartilage, Vol. 22, No. 9, pp. 1259-1270, 2014. https://doi.org/10.1016/j.joca.2014.06.029
  8. S. Kim, H. Kim, H. Hong, J. Wang, "Automatic Segmentation of Meniscus using Position Estimation and Multi-atlas based Locally-weighted Voting in Knee MR Images," KllSE, Proceedings of the Korea Computer Congress, pp. 1412-1414, 2018.
  9. E. B. Dam, M. Lillholm, J. Marques, and M. Nielsen, "Automatic segmentation of high-and low-field knee MRls using knee image quantification with data from the osteoarthritis initiative," Journal of Medical Imaging, 2015.
  10. K. Zhang, W. Lu, and P. Marziliano, "The unified extreme learning machines and discriminative random fields for automatic knee cartilage and meniscus segmentation from multi-contrast MR images," Machine vision and applications, Vol. 24, No. 7, pp. 1459-1472, 2013. https://doi.org/10.1007/s00138-012-0466-9
  11. B. Nonnan, V. Pedoia, and S. Majumdar. "Use of 2D U-Net Convolutional Neural Networks for Automated Cartilage and Meniscus Segmentation of Knee MR Imaging Data to Determine Relaxometry and Morphometry," Radiology, Ahead of Print, 2018.
  12. A. Raj, S. Vishwanathan, B. Ajani, K. Krishnan, and H. Agarwal, "Automatic knee cartilage segmentation using fully volumetric convolutional neural networks for evaluation of osteoarthritis," Biomedical Imaging, pp. 851-854, 2018.
  13. H. Kim, H. Kim, H. Lee, H. Hong, "Automatic Segmentation of Femoral Cartilage in Knee MR Images using Multi-atlas-based Locally-weighted Voting," Journal of Korean Institute of Information Scientists and Engineers, Vol. 43, No. 8, pp. 889-877, 2016.