DOI QR코드

DOI QR Code

A Study on the Implementation of Ultrasonic Guidance Algorithm for Improving Safety of Ultrasonic Varicose Vein Treatment

초음파 하지정맥류 치료의 안전성 개선을 위한 초음파 유도 알고리즘 구현에 관한 연구

  • Kim, Seong-Cheol (School of Biomedical Engineering, Inje University) ;
  • Kim, Ju-Young (School of Biomedical Engineering, Inje University) ;
  • Noh, Si-Cheol (Department of Radiological Science, International University of Korea) ;
  • Choi, Heung-Ho (School of Biomedical Engineering, Inje University)
  • Received : 2018.04.11
  • Accepted : 2018.06.30
  • Published : 2018.06.30

Abstract

In this study, we performed to design an image guiding algorithm to improve the efficiency and safety of treatment of varicose vein by focused ultrasound. The algorithm was suggested by different guiding images according to the location of varicose veins. In the case of deep-seated varicose veins, the target area was marked on the surface of the blood vessel in the obtained cross-sectional blood vessel ultrasound image. In the case of the superficial varicose vein, A guiding system based on image segmentation algorithm of the vascular region was suggested and designed two different algorithms according to varicose veins progression degree. as a results, the algorithm based on ultrasound image show a small error with $830{\mu}m$ at maximum. However, the algorithm based on charge coupled device image has a maximum error of 8.3 mm in some data. Therefore, it is expected that additional study is needed for superficial varicose vein image guiding algorithm, and it is expected that the accuracy of blood vessel tracking should be evaluated by constructing simple system.

본 연구에서는 집속초음파에 의한 하지정맥류 치료에서 치료의 효율과 안전성을 높이기 위한 영상유도 알고리즘을 제안하고자 하였다. 하지정맥류가 발생한 위치에 따라 영상 유도 기법을 달리하여 알고리즘을 수립 하였다. 심부성 하지정맥류의 경우는 획득된 초음파영상에서 혈관의 가로 단면 영상을 추출하고 혈관 중심부의 목표영역을 표시하도록 하였으며, 표재성 하지정맥류의 경우에는 인체의 표재정맥에서 획득한 CCD 카메라 영상에서 혈관 영역을 분리한 영상을 기반으로 한 영상 유도 시스템을 제작하고 하지정맥류 진행 정도에 따라 각기 다른 알고리즘을 설계 하였다. 실험결과 초음파 영상 기반의 알고리즘은 전체적으로 최대 $830{\mu}m$ 정도의 낮은 오차를 보였으나 CCD 영상 기반의 알고리즘은 일부 데이터에서 최대 8.3 mm 정도의 오차를 보였다. 이에 표재성 하지정맥류 영상 유도 알고리즘에 대해서는 추가적인 연구가 필요할 것으로 예상되며 이후 간단한 시스템을 구성함으로써 혈관 추적의 정확도를 평가할 필요성이 있을 것으로 예상된다.

Keywords

References

  1. Kovatcheva, et al, “Ultrasound-guided high- intensity focused ultrasound treatment of breast fibroadenoma multicenter experience,” Journal of therapeutic ultrasound, Vol. 3, No. 1, pp. 1, 2015. https://doi.org/10.1186/s40349-014-0022-3
  2. Li, Shen, WU, et al, "Magnetic resonance image-guided versus ultrasound-guided high- intensity focused ultrasound in the treatment of breast cancer," Chinese journal of cancer. Vol. 32, No. 8, pp. 441, 2013. https://doi.org/10.5732/cjc.012.10104
  3. Y. Zhan and D. Shen, “Deformable Segmentation of 3-D Ultrasound Prostate Images Using Statistical Texture Matching Method,” IEEE Trans. Med. Imag, Vol. 25, No. 3, pp. 245-255, 2006. https://doi.org/10.1109/TMI.2005.862743
  4. N. Terada et al, "Spectral radiative properties of a living human body," International Journal of Thermophys. Vol. 7, No. 5, pp. 1101-1113, 1986. https://doi.org/10.1007/BF00502381
  5. Soohwan Chae, Kyungkoo Jun, "HSV Color Model based Hand Contour Detctor Robust to Noise," Journal of Korea Multimedia Society, Vol. 1, No. 10, pp. 1149-1156, 2015
  6. Xiao, Chang-Yan, Zhang Su, and Ya-zhu Chen, “A diffusion stick method for speckle suppression in ultrasonic images,” Pattern Recognition Letters, Vol. 25, No. 16, pp. 1867-1877, 2004. https://doi.org/10.1016/j.patrec.2004.08.014
  7. Dahl, Anders Bjorholm, and Vedrana Andersen Dahl, "Dictionary snakes," Pattern Recognition(ICPR), 2014 22nd International Conference on. IEEE, 2014.
  8. Dahl, Anders Bjorholm, and Vedrana Andersen Dahl, "Dictionary based image segmentation," Scandinavian Conference on Image Analysis. Springer, Cham, pp 26-37, Cham, 2015.
  9. Lam. L., Seong-Whan Lee, Ching Y. Suen, “Thinning Methodologies-A Comprehensive Survey,” IEEE Trans actions on Patten Analysis and Machine Intelligence, Vol. 14, No. 9, pp. 879, 1992.