Journal of KIISE (정보과학회 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Automatic Segmentation of Femoral Cartilage in Knee MR Images using Multi-atlas-based Locally-weighted Voting

무릎 MR 영상에서 다중 아틀라스 기반 지역적 가중투표를 이용한 대퇴부 연골 자동 분할

  • 김현아 (서울여자대학교 소프트웨어융합학과) ;
  • 김현진 (서울여자대학교 소프트웨어융합학과) ;
  • 이한상 (한국과학기술원 전기 및 전자공학과) ;
  • 홍헬렌 (서울여자대학교 소프트웨어융합학과)
  • Received : 2015.12.23
  • Accepted : 2016.05.10
  • Published : 2016.08.15
⟨ Previous Next ⟩

Abstract

In this paper, we propose an automated segmentation method of femoral cartilage in knee MR images using multi-atlas-based locally-weighted voting. The proposed method involves two steps. First, to utilize the shape information to show that the femoral cartilage is attached to a femur, the femur is segmented via volume and object-based locally-weighted voting and narrow-band region growing. Second, the object-based affine transformation of the femur is applied to the registration of femoral cartilage, and the femoral cartilage is segmented via multi-atlas shape-based locally-weighted voting. To evaluate the performance of the proposed method, we compared the segmentation results of majority voting method, intensity-based locally-weighted voting method, and the proposed method with manual segmentation results defined by expert. In our experimental results, the newly proposed method avoids a leakage into the neighboring regions having similar intensity of femoral cartilage, and shows improved segmentation accuracy.

본 논문에서는 무릎 MR 영상에서 다중 아틀라스 기반 지역적 가중투표를 이용한 대퇴부 연골 자동 분할 방법을 제안한다. 제안하는 방법은 다음의 두 단계로 구성된다. 첫째, 대퇴부 연골이 대퇴골에 붙어 있다는 형상정보를 이용하기 위해 볼륨 및 객체 정합 기반의 지역적 가중투표와 협대역 영역확장을 통해 대퇴골을 분할한다. 둘째, 대퇴골의 객체 기반 어파인 변환을 대퇴부 연골 정합에 적용한 후, 다중 아틀라스 형상 기반의 지역적 가중투표를 통해 대퇴부 연골을 분할한다. 제안 방법의 성능을 평가하기 위해 다수투표 기법, 밝기값 기반 지역적 가중투표 기법과 제안 방법의 분할 결과를 전문가에 의한 수동 분할 결과와 비교한다. 실험 결과 제안 방법이 주변 유사 밝기값 영역으로의 누출을 방지하여 분할 정확도가 향상되었음을 보여준다.

Keywords

Acknowledgement

Supported by : 서울여자대학교 컴퓨터과학연구소

References

  1. H. Shim, S. C. Chang, C. Tao, J.H. Wang, C. K. Kwoh and K. T. Bae, "Knee Cartilage: Efficient and Reproducible Segmentation on High-Spatial-Resolution MR Images with the Semiautomated Graph-Cut Algorithm Method," Radiology, Vol. 251, No. 2, pp. 548-556, May. 2009. https://doi.org/10.1148/radiol.2512081332
  2. G. Vincent, C. Wolstenholme, I. Scott and M. Bowes, "Fully Automatic Segmentation of the Knee Joint using Active Appearance Models," Proc. of MICCAI Workshop on Medical Image Analysis for the Clinic: A Grand Challenge, pp. 224-230, 2010.
  3. H. Seim, D. Kainmueller, H. Lamecker, M. Bindernagel, J. Malinowski and S. Zachow, "Model-based Auto-Segmentation of Knee Bones and Cartilage in MRI Data," Proc. of MICCAI Workshop on Medical Image Analysis for the Clinic: A Grand Challenge, pp. 215-223, 2010.
  4. Y. Yin, R. Williams, D. D. Anderson and M. Sonka, "Hierarchical Decision Framework with a Prior Shape Models for Knee Joint Cartilage Segmentation," Proc. of MICCAI Workshop on Medical Image Analysis for the Clinic: A Grand Challenge, pp. 241-250, 2010.
  5. S. Lee, S. H. Park, H. Shim, I. D. Yun and S. U. Lee, "Optimization of Local Shape and Appearance Probabilities for Segmentation of Knee Cartilage in 3D MR Images," Computer Vision and Image Understanding, pp. 1710-1720, 2011. https://doi.org/10.1016/j.cviu.2011.05.014
  6. Z. Wang, C. Donoghue and D. Rueckert, "Patchbased Segmentation without Registration: Application to Knee MRI," Proc. of MICCAI Workshop on Machine Learning in Medical Imaging, pp. 98-105, 2013.
  7. Q. Wang, D. Wu, L. Lu, M. Liu, K. L. Boyer and S. K. Zhou, "Semantic Context Forests for Learning-Based Knee Cartilage Segmentation in 3D MR Images," Proc. of MICCAI Workshop on Medical Computer Vision, Large Data in Medical Imaging, Vol. 8331, pp. 105-115, 2014.
  8. M. Amberg, M. Luthi and T. Vetter, "Fully Automated Segmentation of the Knee using Local Deformation-Model Fitting," Proc. of MICCAI Workshop on Medical Image Analysis for the Clinic 2010, pp. 251-260, 2010.
  9. J. G. Lee, S. Gumus, C. H. Moon, C. K. Kwoh and K. T. Bae, "Fully automated segmentation of cartilage from the MR images of knee using a multiatlas and local structural analysis method," Medical Physics, Vol. 41, No. 092303, pp. 1-10, 2014. https://doi.org/10.1118/1.4893533
  10. L. Shan, C. Zach, C. Charles and M. Niethammer, "Automatic atlas-based three-label cartilage segmentation from MR knee images," Medical Image Analysis, Vol. 18, pp. 1233-1246, 2014. https://doi.org/10.1016/j.media.2014.05.008
  11. Z. Wang, C. Donoghue and D. Rueckert, "Patch-Based Segmentation without Registration: Application to Knee MRI," Machine Learning in Medical Imaging, Springer International Publishing, pp. 98-105, 2013.
  12. H. A. Kim, H. J. Kim, H. S. Lee and H. Hong, "Multi-Atlas Segmentation of Femoral Cartilage in Knee MR Images," The 2015 Fall Conference of Korea Multimedia Society, Vol. 18, No. 2, pp. 2, 2015. (in Korean)
  13. H. Lee, H. Kim, H. Kim, H. Hong, Y. Yoon and J. Kim, "Automated Multi-Atlas Segmentation of Cartilage from Knee MR Images Using Locally-Weighted Voting and Graph Cuts," Radiology Society of North America(RSNA) 2015.
  14. J. G. Lee, J. H. Kim, S. H. Kim, H. S. Park and B. I. Choi, "A straightforward approach to computeraided polyp detection using a polyp-specific volumetric feature in CT colonography," Computers in biology and medicine, Vol. 41, No. 9, pp. 790-801, 2011. https://doi.org/10.1016/j.compbiomed.2011.06.015
  15. R, A, Heckemann, J. V. Hajnal, P. Aljabar, D. Rueckert and A. Hammers, "Automatic anatomical brain MRI segmentation combining label propagation and decision fusion," Neuroimage, Vol. 33, No. 1, pp. 115-126, 2006. https://doi.org/10.1016/j.neuroimage.2006.05.061