| Gui Longgang,Li Jun,Shi Miao.Quantifying the effects of parameter interactions on treatment efficiency in Radixact helical tomotherapy for nasopharyngeal carcinoma[J].Chinese Journal of Radiological Medicine and Protection,2026,46(8):755-762 |
| Quantifying the effects of parameter interactions on treatment efficiency in Radixact helical tomotherapy for nasopharyngeal carcinoma |
| Received:February 06, 2026 |
| DOI：10.3760/cma.j.cn112271-20260206-00044 |
| KeyWords:Nasopharyngeal carcinoma|Radixact|Helical tomotherapy|Treatment efficiency|Response surface methodology|Parameter optimization |
| FundProject: |
|
| Hits: 399 |
| Download times: 8 |
| Abstract:: |
| Objective To explore the nonlinear effects of key machine parameters on treatment efficiency in helical tomotherapy (HT) for nasopharyngeal carcinoma (NPC), in order to develop and verify a response surface method ology (RSM)-based quantitative prediction model for treatment time. Methods A total of 10 NPC patients treated at the Department of Radiation Oncology, Northern Jiangsu People's Hospital, were enrolled in this study. On the premise that the absolute safety of organs at risk (OARs), including the brainstem and spinal cord, was preferentially guaranteed, 170 test plans were generated using the Radixact system by changing the combinations of field width, pitch, and modulation factor. Based on mechanical constraints and dosimetric criteria, 147 valid plans were finally selected. The characteristic variables of the 147 plans, including actual treatment time, conformity index (CI), and homogeneity index (HI), were recorded. Then, a regression model for predicting treatment time was established through fitting using a multivariable quadratic polynomial. Furthermore, the main effects and interactions of the parameters were assessed through analysis of variance (ANOVA). Results Under strict constraints on doses to OARs and the influence of adjacent anatomical structures, all valid plans showed rational dose compensation in the target volumes (CI: 1.03-1.18; HI: 1.04-1.15; mean PTV dose deviations: < 1%). Under comparable clinically acceptable quality, the plans with varying parameter combinations exhibited treatment times ranging from 245.0 to 1752.0 s. The regression model exhibited encouraging fitting effects, with the coefficient of determination (R2) exceeding 0.93. ANOVA result revealed that the main effects of field width, pitch, and modulation factor on treatment time were all statistically significant (F = 182.61, 64.80, 51.33, P < 0.01). Furthermore, the difference of interaction between pitch and modulation factor was also statistically significant (F = 12.71, P < 0.01). Conclusions The treatment delivery time in helical tomotherapy is not a simple linear superposition of individual machine parameters. The RSM-based quantitative model enables the effective characterization of the nonlinear interactions among multiple parameters. This model can improve parameter selection efficiency while maintaining doses to target volumes and provide a quantitative reference for subsequent parameter optimization. |
| HTML View Full Text View/Add Comment Download reader |
| Close |
|
|
|
