Application of RAIC.OIS in automatic segmentation of the heart in patients with esophageal cancer(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2019年第12期
- Page:
- 1377-1382
- Research Field:
- 医学放射物理
- Publishing date:
Info
- Title:
- Application of RAIC.OIS in automatic segmentation of the heart in patients with esophageal cancer
- Author(s):
- SHI Feiyue1; 2; WANG Min1; QIN Wei1; JIN Xun3; ZHAO Huanyu1
- 1. Radiation Therapy Center, Nanjing First Hospital, Nanjing Medical University, Nanjing 210006, China; 2. Center of Medical Physics, Nanjing Medical University, Nanjing 210029, China; 3. Second Clinical Medical College, Nanjing University of Chinese Medicine, Nanjing 210023, China
- Keywords:
- esophageal cancer; automatic segmentation; heart; radiotherapy; cloud platform
- PACS:
- R318；R815.6
- DOI:
- DOI:10.3969/j.issn.1005-202X.2019.12.003
- Abstract:
- Abstract: Objective To test and evaluate the application of intelligent radiotherapy cloud platform (RAIC.OIS) for automatically segmenting the heart in patients with esophageal cancer. Methods Twenty patients with esophageal cancer admitted to hospital from February to November 2018 were enrolled in the study. The planning CT images of 20 patients were firstly transferred from Eclipse treatment planning system to RAIC.OIS of LINKING MED. Then the heart in CT images was automatically segmented by RAIC.OIS. Finally, the structure files of the heart were transferred to Eclipse after automatic segmentation. The feasibility of using the proposed software for automatic segmentation of the heart was evaluated by comparing the differences in volume, position, Dice similarity coefficient and segmentation time between automatic segmentation and manual segmentation. Results The measurement results showed that the shape and position of the heart in a certain patient were special. After eliminating the data of the patient, statistical analysis was conducted on the data of the other 19 patients. The differences in volume between automatic segmentation and manual segmentation were (-17.08±8.66)%, and Dice similarity coefficient value was 0.87±0.05. The position differences of x, y and z directions were (0.12±0.09), (0.11±0.08) and (0.22±0.16) cm, respectively, and the total position difference was (0.31±0.14) cm. The time for automatic segmentation and manual segmentation for the 19 patients were (83±12) and (284±58) s, respectively. Conclusion The proposed intelligent radiotherapy cloud platform can be used to obtain satisfactory automatic segmentation of the heart in most patients with esophageal cancer. The use of proposed automatic segmentation software can shorten the time for the heart segmentation and improve working efficiency.
Last Update: 2019-12-24