Application of automatic segmentation software in head and neck cone beam CT(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2019年第3期
- Page:
- 254-258
- Research Field:
- 医学放射物理
- Publishing date:
Info
- Title:
- Application of automatic segmentation software in head and neck cone beam CT
- Author(s):
- HU Xiaowei1; SUN Xinchen2; LI Jinkai2; CHANG Zhigang2; CHENG Hongyan2
- 1. Department of Special Medicine, Nanjing Medical University, Nanjing 210009, China; 2. Department of Radiation Oncology, the First Affiliated Hospital of Nanjing Medical University, Nanjing 210009, China
- Keywords:
- ; Keywords: nasopharyngeal carcinoma; cone beam CT; deformable registration; rigid registration; automatic segmentation software
- PACS:
- R312；R811.1
- DOI:
- DOI:10.3969/j.issn.1005-202X.2019.03.002
- Abstract:
- Abstract: Objective To evaluate the feasibility of using automatic segmentation software (MIM) to automatically segment the target areas and organs-at-risk on head and neck cone beam CT (CBCT) image. Methods The clinical data of 10 patients with nasopharyngeal carcinoma (NPC) who were treated with radiotherapy positioning twice were analyzed retrospectively. In MIM software system, the contours of 1CT and 1CBCT scanned before the first treatment fraction were propagated to 2CBCT scanned before the second treatment fraction by deformable image registration. The segmented structures of 2CT were used as the reference to compare the precision of the two methods, and the evaluation index includes Dice similarity coefficient (DSC) and Hausdorff distance (HD). The structures obtained by the two methods were duplicated to 2CT. With 2CT plan as the reference plan, the dose deviations in target areas and organs-at-risk were compared. Results The structural contours segmented by 1CBCT-2CBCT were better than those obtained by 1CT-2CBCT. In 1CBCT-2CBCT, except for PTVnd, spinal cord, lens, optic nerve, optic chiasma, eyeball and pituitary, all the structures had DSC greater than 0.7 and HD less than 2 cm. The dose deviation between the reference plan and 1CBCT-2CBCT were less than ±3%. Conclusion With the use of MIM software, the target area and organs-at-risk, excluding the structures with volume less than 10 cm3 (except for bone structures), spinal cord and PTVnd, can be delineated by 1CBCT-2CBCT, achieving satisfactory segmentation results and providing clinical guidance for adaptive radiotherapy.
Last Update: 2019-03-25