Analysis of sensitivity and specificity of PerFRACTION system based on log files(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2021年第4期
- Page:
- 404-410
- Research Field:
- 医学放射物理
- Publishing date:
Info
- Title:
- Analysis of sensitivity and specificity of PerFRACTION system based on log files
- Author(s):
- YAN Huifang1; 2; LIU Hongyuan2; YANG Zhiyong2; LI Qin2; HAN Jun2; NIE Xin2; LIANG Zhiwen2; ZHANG Sheng2
- 1. Department of Radiation Oncology, the Second Affiliated Hospital of Zhengzhou University, Zhengzhou 450000, China 2. Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China
- Keywords:
- Keywords: PerFRACTION system log files gamma analysis receiver operating characteristic curve sensitivity specificity
- PACS:
- R815.6
- DOI:
- DOI:10.3969/j.issn.1005-202X.2021.04.002
- Abstract:
- Abstract: Objective To analyze the sensitivity and specificity of PerFRACTION system for volumetric modulated arc therapy (VMAT) plans executed on Trilogy accelerator, thereby obtaining the diagnostic efficacy and gamma passing criteria of PerFRACTION system for the dose verification of VMAT plans. Methods PerFRACTION is a real-time online dose verification system that could calculate the three-dimensional dose distribution through muti-leaf collimator log files, online field image data and convolution/superposition algorithm, and automatically generate dose volume histogram and gamma passing rate. Two groups of VMAT plans for different anatomical sites were randomly selected in the study, and each group of plans included original plans and modified plans with the introduction of random errors. The sensitivity and specificity of PerFRACTION system were detected on original plans and modified plans. In addition, two groups of the same plans were tested by ArcCHECK phantom, and the result was taken as a reference. Results The dosimetric analysis on original and modified plans were performed with gamma criteria of 3%/3 mm and 3%/2 mm. The gamma passing rates of original plans were above 98% and 95%, which met the requirements of AAPM-TG218 report, and some of modified plans were also above 98% and 95% under the same criteria. Therefore, the errors introduced in modified plans were not detected, indicating that the gamma criteria was not targeted. The study aimed to obtain gamma passing criteria with high diagnostic efficiency for specific diseases in our cancer center. Based on PerFRACTION system, the areas under the receiver operating characteristic curves (AUC) of head and neck VMAT plans were 0.86 and 0.88, respectively, and the corresponding AUC values of pelvic VMAT plans were 0.76 and 0.76, respectively. Similarly, based on ArcCHECK phantom, the AUC values of head and neck VMAT plans were 0.97 and 0.82 under the gamma criteria of 3%/3 mm and 3%/2 mm and the corresponding AUC values of pelvic VMAT plans were 0.98 and 0.86. Compared with the verification results of ArcCHECK phantom, ROC analysis showed that the gamma criteria (3%/3 mm, 3%/2 mm) of PerFRACTION system were suitable for the analysis of quality assurance (QA). The gamma passing rates of head and neck VMAT plans were 98.80% and 96.32%, respectively, and those of pelvic VMAT plans were 99.62% and 99.36%, respectively. Under the same criteria, PerFRACTION system had a higher sensitivity and specificity. Conclusion PerFRACTION system is reliable as ArcCHECK phantom and can be used to detect random errors. With the properties of convenience, real-time and highly sensitive, PerFRACTION system has greater advantages in monitoring daily plan transmission and plan implementation.
Last Update: 2021-04-29