The effect of tissue inhomogeneity correction on the accuracy of dose calculation in brachytherapy of cervical carcinoma
10.3760/cma.j.cn113030-20231230-00220
- VernacularTitle:宫颈癌后装计划组织不均匀性校正对剂量计算的影响
- Author:
Junjie HAN
1
;
Huikuan GU
;
Zhenyu QI
Author Information
1. 南方医科大学附属广东省人民医院（广东省医学科学院）放疗科，广州 510080
- Keywords:
Brachytherapy;
Monte Carlo;
Egs_brachy;
Tissue inhomogeneity correction;
Treatment planning system
- From:
Chinese Journal of Radiation Oncology
2024;33(11):1049-1055
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the impact of tissue inhomogeneity correction on the accuracy of dose calculations in brachytherapy of cervical carcinoma by comparing the results of Monte Carlo (MC) dose simulation with those (TG43 algorithm) of treatment planning system (TPS).Methods:Firstly, the 192Ir source was modeled by using a MC code specially designed for brachytherapy, called egs_brachy. The accuracy of this model was verified by comparing it with the published data. Then, 8 brachytherapy plans of cervical carcinoma were selected which completed treatment at Sun Yat-sen University Cancer Center from January 2022 to May 2023, and their CT image data and treatment parameters were exported. Relevant plan information such as the source dwell positions and their corresponding dwell times were reconstructed on the patient's individualized CT images using a self-developed program. The MC dose distributions were calculated for each case and compared with the TPS calculations. When the anterior wall of the rectum was filled with gas, the differences between MC simulation and TPS calculation were compared. Additionally, 5 different calculation ranges were set for MC simulation, and the MC simulation results of different calculation ranges were compared with the TPS's, combining with the time of MC simulation, a reasonable MC calculation range was comfirmed. Then, the scipy.stats library of Python was utilized to perform independent sample t-test on dosimetric comparison results, including D 90% of high risk clinical target volume (HR-CTV) and D max, D mean, and D 2 cm3 of organs at risk (OAR). Results:The comparison between MC calculations and TPS results showed that the CTV's D 90%, the bladder's D mean and the small intestine's D mean were all within ± 1%, except for the D max difference of approximately 3% on the anterior wall of the rectum. The 2%/2 mm gamma pass rates were all>98%. When anterior wall of the rectum filled with gas, compared with MC, TPS overestimated the anterior wall of the rectum's D 2 cm3 and D mean by approximately 6.06% ( t=-6.80, P=0.002) and 5.35% ( t=-6.57, P=0.003), respectively. When the dose calculation range of MC was consistent with that of TPS, the MC calculation result underestimated the dose distribution in water by approximately 4%. When extending the MC dose calculation range by 2 cm beyond the TPS calculation range, the dose difference between MC and TPS in homogeneous water was approximately 1%, and the calculation time was saved by at least 8 h compared to MC dose calculation based on the whole CT. Conclusion:The existing TPS TG43 algorithm can ensure that the dose calculation of cervical carcinoma meets the basic accuracy requirements of clinical practice, but tissue inhomogeneity correction is recommended to improve the accuracy of dose calculation whenever possible.
