Lung diaphragm tracking in CBCT images using spatio-temporal MRF
详细信息 查看全文
- 作者：Manivannan Sundarapandiana ; b ; manivannan.sundarapandiyan@siemens.com" class="auth_mail" title="E-mail the corresponding author ; Ramakrishnan Kalpathia ; krr@ee.iisc.ernet.in" class="auth_mail" title="E-mail the corresponding author ; R. Alfredo C. Siochic ; Amrut S. Kadamd
- 关键词：Markov Random Field ; Graph-cuts ; Cone-beam CT ; Diaphragm motion
- 刊名：Computerized Medical Imaging and Graphics
- 出版年：2016
- 出版时间：October 2016
- 年：2016
- 卷：53
- 期：Complete
- 页码：9-18
- 全文大小：1671 K
文摘
- •
We propose a method for tracking the lung diaphragm position on CBCT projection images.
- •
The diaphragm state is modeled as a spatio-temporal Markov Random Field.
- •
The associated energy minimization problem is solved using graph-cuts.
- •
On clinical datasets, our method outperforms the full search method in terms of accuracy.
- •
A GPU based implementation of our method achieves 16% acceleration over the benchmark.