뇌혈관 추출과 대화형 가시화를 위한 다중 GPU기반 영상정합

Multi GPU Based Image Registration for Cerebrovascular Extraction and Interactive Visualization

  • 박성진 (서울대학교 컴퓨터공학부) ;
  • 신영길 (서울대학교 컴퓨터공학부)
  • 발행 : 2009.06.15

초록

본 논문에서는 조영전 CT 와 조영후 CTA 영상 의 움직임을 보정하기 위하여 연산에 효율적인 다중 GPU 기반 영상정합 기법을 제안한다. 제안방법은 크게 다중 GPU 기반 정합과 뇌혈관 가시화의 두 단계로 구성된다. 우선, 복셀기반정합을 수행하기 위하여 GPU 내부의 병렬성뿐 아니라 GPU 간 병렬성도 고려함으로써 유사도값을 계산한다. 그리고 나서 CTA 영상데이터에서 최적변환행렬에 의하여 변환된 CT 영상데이터를 다중 GPU를 이용하여 차감하고, 차감된 결과를 GPU 기반 볼륨렌더링기법을 이용하여 가시화한다. 본 논문에서 제안한 방법을 화질과 수행시간측면에서 기존방법에 대한 우수성을 나타내기 위하여 5쌍의 조영전 뇌 CT 영상과 조영후 뇌 CTA 영상데이터를 사용하여 비교하였다. 실험결과 제안방법은 뇌혈관이 잘 가시화되어 혈관질환을 정확히 진단할 수 있었다. 다중 GPU 기반 방법은 CPU 기반 방법에 비하여 11.6배, 단일 GPU 기반 방법에 비하여 1.4배 빠른 결과를 보여주었다.

In this paper, we propose a computationally efficient multi GPU accelerated image registration technique to correct the motion difference between the pre-contrast CT image and post-contrast CTA image. Our method consists of two steps: multi GPU based image registration and a cerebrovascular visualization. At first, it computes a similarity measure considering the parallelism between both GPUs as well as the parallelism inside GPU for performing the voxel-based registration. Then, it subtracts a CT image transformed by optimal transformation matrix from CTA image, and visualizes the subtracted volume using GPU based volume rendering technique. In this paper, we compare our proposed method with existing methods using 5 pairs of pre-contrast brain CT image and post-contrast brain CTA image in order to prove the superiority of our method in regard to visual quality and computational time. Experimental results show that our method well visualizes a brain vessel, so it well diagnose a vessel disease. Our multi GPU based approach is 11.6 times faster than CPU based approach and 1.4 times faster than single GPU based approach for total processing.

키워드

참고문헌

  1. Jayaraman, M., Mayo-Smith, W., and Doberstein, C., “Intracanalicular Aneurysm of the Anterior Inferior Cerebellar Artery Revealed by Multi-Detector CT Angiography, AJNR Am J Neuro-radiol, Vol.24, pp. 1338-1340, 2003
  2. Yan, C., Hirano, S., and Hata, Y., “Extraction of blood vessel in CT angiography image aided by fuzzy logic,” Proc. of ICSP 2000, pp. 926-929, 2000 https://doi.org/10.1109/ICOSP.2000.891673
  3. Abrahams, J., Saha, P., Hurst, R., LeRoux, P., and Udupa, J., “Three-dimensional bone-free rendering of the cerebral circulation by use of computed tomographic angioraphy and fuzzy connected-ness,” Neurosurgery, Vol.51, pp. 264-269, 2002 https://doi.org/10.1097/00006123-200207000-00044
  4. Yeung, M., Yeo, B., Liou, S., and Banihashemi, A., “Three-dimensional image registration for spiral CT angiography,” Proc. of Computer Vision and Pattern Recognition, pp. 423-429, 1994 https://doi.org/10.1109/CVPR.1994.323861
  5. Venema, H., Hulsmans, F., and Heeten, G., “CT angiography of the circle of willis and intracranial internal carotid arteries: maximum intensity pro-jection with matched mask bone elimination-feasi-bility study,” Radiology, Vol.218, pp. 893-898, 2001 https://doi.org/10.1148/radiology.218.3.r01mr30893
  6. Kwon, S., Kim, Y., Kim, T., and Ra, J., “Digital subtraction CT angiography based on efficient 3D registration and refinement,” Computerized Medical Imaging and Graphics, Vol.28, pp. 391-400, 2004 https://doi.org/10.1016/j.compmedimag.2004.06.003
  7. Chisu, R., 'Techniques for Accelerating Intensity-based Rigid lmage Registration,' PhD thesis, Technische Universit at Munchen, Munchen, Germany, 2005
  8. Soza, G., Bauer, M., Hastreiter, P., Nimsky, Ch., and Greiner, G., 'Non-Rigid Registration with Use of Hardware-Based 3D B'ezier Functions,' MICC-Al 2002, pp. 549-556, 2002
  9. Kohn, A., Drexl, J., Ritter, F., Konig, M., and Peitgen, M., 'GPU accelerated image registration in two and three dimensions,' BiIdverarbeitung Fur die Medizin 2006, Vol.3, pp. 261-265, 2006 https://doi.org/10.1007/3-540-32137-3_53
  10. Maes, F., Collignon, A., Vandermeulen, D., Marchal, G., and Suetens, P., "Multimodality image registration by maximization of mutual information," IEEE Trans. on Medical Imaging, Vol.16, pp. 187-198, 1997 https://doi.org/10.1109/42.563664
  11. Kruger, J., and Westermann, R. 'Acceleration techniques for GPU-based volume rendering,' in Proceedings of IEEE Visualization '03, 2003 https://doi.org/10.1109/VISUAL.2003.1250384
  12. Studholme, C., Hill, D., and Hawkes, D., 'Auto-mated three-dimensional registration of magnetic resonance and positron emission tomography brain images by multiresolution optimization of voxel similarity measures,' Medical Physics, Vol.24, No.1., pp. 25-35, 1997 https://doi.org/10.1118/1.598130