Medical physics · 2026
Feasibility study of surface deformation-based dosimetry for breast cancer radiotherapy.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Deng Yongjin, Qiu Minmin, Xiao Zhenhua, Huang Botian, Liu Yujian, Guan Qi, Zhong Jiajian, Bao Yong
Аннотация
BACKGROUND: With the increasing complexity of breast cancer radiotherapy techniques, accurate dose delivery has become more critical than ever. Real-time dosimetry now plays an essential role in ensuring both treatment efficacy and patient safety.
PURPOSE: To propose a novel dosimetry approach for breast cancer radiotherapy that leverages surface deformation to enhance dose delivery accuracy.
METHOD: This retrospective study analyzed 29 left breast cancer radiotherapy cases from an existing clinical database. The methodology incorporated nonrigid registration between planning CT and CBCT for dose recalculation, and surface images computed from planning CT and CBCT were analyzed using monocular vision algorithms to simulate clinical monocular camera-based surface deformation detection. A neural network architecture was subsequently developed to establish the predictive relationship between surface deformation and internal dose redistribution. Model performance was rigorously evaluated through RMSE and 3D gamma pass rate analysis.
RESULT: The analysis revealed a robust correlation between breast surface deformation features and internal dose redistribution, as evidenced by a first canonical correlation coefficient of 0.940. Quantitative evaluation demonstrated strong predictive performance, with RMSE of 0.005 pixels for the dose deformation field and 0.009 cGy for dose amplitude differences in the test set. Gamma analysis results indicated clinically acceptable accuracy, with all of cases achieving 100% pass rate under 2 mm/2% criteria.
CONCLUSIONS: This study presents a novel noninvasive dosimetry approach for breast cancer radiotherapy that enables real-time monitoring without requiring additional radiation exposure.
PURPOSE: To propose a novel dosimetry approach for breast cancer radiotherapy that leverages surface deformation to enhance dose delivery accuracy.
METHOD: This retrospective study analyzed 29 left breast cancer radiotherapy cases from an existing clinical database. The methodology incorporated nonrigid registration between planning CT and CBCT for dose recalculation, and surface images computed from planning CT and CBCT were analyzed using monocular vision algorithms to simulate clinical monocular camera-based surface deformation detection. A neural network architecture was subsequently developed to establish the predictive relationship between surface deformation and internal dose redistribution. Model performance was rigorously evaluated through RMSE and 3D gamma pass rate analysis.
RESULT: The analysis revealed a robust correlation between breast surface deformation features and internal dose redistribution, as evidenced by a first canonical correlation coefficient of 0.940. Quantitative evaluation demonstrated strong predictive performance, with RMSE of 0.005 pixels for the dose deformation field and 0.009 cGy for dose amplitude differences in the test set. Gamma analysis results indicated clinically acceptable accuracy, with all of cases achieving 100% pass rate under 2 mm/2% criteria.
CONCLUSIONS: This study presents a novel noninvasive dosimetry approach for breast cancer radiotherapy that enables real-time monitoring without requiring additional radiation exposure.