The Feasibility of Mapping Dose Distribution of 4DCT Images with Deformable Image Registration in Lung

Issue title: Frontiers in Biomedical Engineering and Biotechnology – Proceedings of the 2nd International Conference on Biomedical Engineering and Biotechnology, 11–13 October 2013, Wuhan, China

Article type: Research Article

Authors: Yu, Hui^; | Zhang, Shu-xu | Wang, Rui-hao | Zhang, Guo-qian | Tan, Jian-ming

Affiliations: Department of Radiotherapy, the Affiliated Tumor Hospital of Guangzhou Medical University, Guangzhou 510095, P.R. China

Note: [] Corresponding author. E-mail: [email protected].

Abstract: Calculating an accurate cumulative dose through individual phases for four-dimensional computed tomography (4DCT) images from the lung is time-consuming. Although the dose distribution of different phases is similar, copying the dose distribution of one phase directly to another phase would yield a dosimetric error of approximately 4% without further optimization. To reduce the dosimetric error, three-dimensional B-spline elastic deformable image registration (DIR) was used to quickly obtain a relatively accurate cumulative dose of 4DCT images acquired from ten lung cancer patients. The dose distribution of the end-expiration phase was mapped to the end-inspiration phase using DIR. The mapped dose in the end-inspiration phase was then compared with the directly copied dose by analysis (3cm/3%) and the t-test. The results showed that optimization using DIR was significantly better in the average pass rate (by 0.6–4.7%). Our results indicate it is feasible to map the dose distribution of 4DCT images in lung with DIR, and that the motion amplitude of individual respiratory and different DIR algorithms affect the differences between the mapped and actual dose.

Keywords: 4DCT, deformable image registration, dose distribution, γ analysis

DOI: 10.3233/BME-130794

Journal: Bio-Medical Materials and Engineering, vol. 24, no. 1, pp. 145-153, 2014