Matching of Respiratory Wave Form and Ultrasound Image Measurement Using DTW (Dynamic Time Warping)

DOI Web Site 16 References
  • HIRASAWA Kousuke
    Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology Advanced Technology R&D Center, Mitsubishi Electric Corporation
  • MINOHARA Shinichi
    Research Center for Charged Particle Therapy, National Institute of Radiological Sciences
  • IMADA Hiroshi
    Research Center for Charged Particle Therapy, National Institute of Radiological Sciences
  • YANAGI Takeshi
    Research Center for Charged Particle Therapy, National Institute of Radiological Sciences
  • YASUDA Shigeo
    Research Center for Charged Particle Therapy, National Institute of Radiological Sciences
  • KATO Hirotoshi
    Research Center for Charged Particle Therapy, National Institute of Radiological Sciences
  • SAWADA Akira
    Graduate School of Medicine, Kyoto University
  • MIYAKE Yoshihiro
    Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology

Bibliographic Information

Other Title
  • DTW (Dynamic Time Warping) を用いた呼吸波形と超音波画像計測の対応付け
  • DTW Dynamic Time Warping オ モチイタ コキュウ ハケイ ト チョウオンパ ガゾウ ケイソク ノ タイオウ ズケ

Search this article

Description

In radiotherapy for lungs and livers, it is clinically important to concentrate its beam on the moving affected part in internal organs minimizing the irradiation to the healthy tissues. The method of irradiating the beam to the affected part synchronizing with the patient's respiration is widely recognized, where the indirect measurement of the marker displacement on the body surface is often used for its easy setup and measurement in real treatments. However, the irradiation accuracy is not enough when the respiratory status of the patients doesn't reflect the movement of internal organs accurately. In this paper, it has been studied how the movement of internal organs observed by 3D ultrasound image is related to the respiratory signal observed by the position sensitive camera. Firstly, it was evaluated based on the respiratory amplitude and respiratory phase, which made clear that there are some variances in the estimation of the internal organ's position from the respiratory signal in both cases. Secondly, the algorithm of DTW (Dynamic Time Warping) used by voice recognition has been applied to the respiratory wave form and the distance (amplitude difference) was calculated between the phase of each respiratory wave form and its average respiratory wave form. It was confirmed that the variance of the estimation of the internal organ's position from the respiratory signal was reduced from about 4.0 mm to 2.5 mm based on the phase matched by DTW.

Journal

References(16)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top