Hidden Markov model-based extraction of target objects in X-ray image sequence for lung radiation therapy

Masahiro Shindo, Kei Ichiji, Noriyasu Homma, Xiaoyong Zhang, Shungo Okuda, Norihiro Sugita, Shunsuke Yamaki, Yoshihiro Takai, Makoto Yoshizawa

研究成果: Article査読

1 被引用数 (Scopus)


It is an important task to accurately track the target tumor with respiratory movement during radiation therapy. X-ray imaging technique is capable of observing the internal organ motion. However, superimposed tissues and structures in X-ray images decrease tumor localization accuracy. This paper presents a target extraction method based on hidden Markov model (HMM) to enhance the target tumor in X-ray images for improving the tumor tracking accuracy. We first simulate possible combinations of image intensities of target objects as hidden states and observable X-ray image intensities as output symbol in HMM by using digitally reconstructed radiographs generated from four-dimensional X-ray computed tomography. Subsequently, the transition dynamics of the hidden states and output symbols is estimated by applying Baum-Welch algorithm to a training dataset. The transition sequence of the hidden states is inversely estimated from the observed X-ray image sequence by using Viterbi algorithm, and then the transition sequence is finally decomposed into the subset image sequences. Experimental results demonstrated that tracking performance of the proposed method is superior to that of conventional tumor enhancement method and raw images. Therefore, the proposed method has potential for contributing to effectively observe internal organ motion.

ジャーナルIEEJ Transactions on Electronics, Information and Systems
出版ステータスPublished - 2020

ASJC Scopus subject areas

  • 電子工学および電気工学


「Hidden Markov model-based extraction of target objects in X-ray image sequence for lung radiation therapy」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。