The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지
DOI QR코드

DOI QR Code

A Method for Sinogram Interpolation for Reducing X-ray Dose

CT의 선량 감소를 위한 sinogram 보간 기법

  • 김재민 (건국대학교 전자공학과 디지털신호처리 연구실) ;
  • 이기승 (건국대학교 전자공학과 디지털신호처리 연구실)
  • Received : 2012.02.29
  • Accepted : 2012.06.25
  • Published : 2012.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, a limited-view CT image reconstruction method was studied to reduce the scan times and the X-ray dose for the patients. To reduce streak artifacts which is caused by insufficient number of views, we introduce a sinogram interpolation method based on image matching. Image matching is achieved using the characteristics of the neighboring views including intensity, gradient and distance between the pixels. Interpolation is performed using the image matching results.. A numerical phantom and Al-acryl phantom were used for evaluating the effectiveness of the proposed interpolation method. The results showed that streak artifacts were reduced in the reconstructed images while the details of the images were preserved. Moreover, maximum 5% improvements in terms of PSNR were observed.

본 논문에서는 전산화 단층촬영 시 스캔 시간의 단축 및 환자의 X-Ray 피폭량을 줄이기 위한 방법으로 샘플링 빈도를 줄이고 적은 수의 view를 이용하여 단면 영상을 재구성하는 방법에 대해 연구하였다. 부족한 view로 인해 단면 영상에 발생하는 방사형의 직선형 왜곡(streak artifact) 를 억제하기 위하여 본 논문에서는 sinogram 보간기법을 적용하였다. 인접 view간 영상 특성(밝기, 그라디언트, 픽셀거리)을 기반으로 sinogram 픽셀 간의 매칭을 수행하고 그 결과를 기반으로 변화패턴을 추적하여 sinogram을 보간하도록 하였다. Numerical 팬텀과 자체 제작된 팬텀을 이용하여 제안된 기법을 검증하였다. 선형 보간 기법에 비해 재구성된 단면영상의 디테일을 유지하면서 왜곡이 감소하는 것이 관찰되었으며, PSNR면에서 최대 5%의 향상이 있는 것을 확인하였다.

Keywords

References

  1. The Korean Society of Medical Imaging Technology, "Textbook of Computed Tomography 2nd edition", 2005년.
  2. Jianhua Ma, Jing Huang, Zhengrong Liang, Hua Zhang, Yi Fan, Qianjin Feng, Wofan Chen "Image Fusion for Low-Dose Computed Tomography Reconstruction," IEEE, Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC), pp.4239-4243, Oct, 2011.
  3. Tianfang Li, Xiang Li, Yuziang Xing, Holgbing Lu, Kiang Hsieh and Zhengrong Liang, "A strategy for reduction of streak artifacts in low-dose CT," IEEE, Nuclear Science Symposium Conference Record, Vol 4, pp.2743-2747, 2003.
  4. Nevzat Karabulut, Mustafa Toru, Veli Gelebek, Meltem Gulsun, O. Macit Ariyurek, "Comparison of low-dose and standard-dose helical CT in the evaluation of pulmonary nodules" ESTI, Vol.8, No.10, June, 2001.
  5. R. R. Galigekere, K. Wiesent, D. W. Holdsworth, "Techniques to alleviate the effects of view aliasing artifacts in computed tomography" Med. Phys., Vol.26, No.6, June, 2006.
  6. A. Goshtasby, David A. Turner, Laurens V. Ackerman "Matching of Tomographic Slices for Interpolation" IEEE, Transactions on Medical Imaging, Vol.11, No.4, Dec, 2005.
  7. Xiao Han, Erik Pearson, Jungou Bian, Seungryong Cho, Emil Y. Sidky, Charles A. Pelizzari, Xiaochuan Pan "Preliminary Investigation of Dose Allocation in Low-dose Cone-beam CT," IEEE, Nuclear Science Symposium Conference Record (NSS/MIC) pp.2051-2054, Nov, 2010.
  8. Junguo Bian, Xiao Han, Jiong Wang, Emil Y. Sidky, Lingxiong Shao, Xiaochuan Pan "Low-dose CT in SPECT/CT patient scan," IEEE, Nuclear Science Symposium Conference Record (NSS/MIC), pp.2042-2045, Nov, 2010.