Cancer research and treatment · 2026
Radiomics-Based Prediction of Lymphedema after Radiotherapy in Breast Cancer: Integrating Clinical and Dosimetric Features.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
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.
Kim Jae Sik, Jeon Seung Hyuck, Jang Bum-Sup, Kim Jin Ho, Chang Ji Hyun, Kim Dowook, Shin Kyung Hwan
Аннотация
PURPOSE: Arm lymphedema is a common, debilitating complication in patients with breast cancer undergoing postoperative radiotherapy (PORT). Although clinical and dosimetric factors have been used for risk prediction, radiomics offers a novel approach for improving the predictive accuracy.
MATERIALS AND METHODS: We designed a predictive model for lymphedema using clinical, dosimetric, and radiomic features. We included 532 patients (399 training and 133 testing) who underwent breast cancer surgery followed by PORT. Radiomic features were extracted from axillary levels I, II, III, and supraclavicular regions, which were automatically contoured on PORT-planning computed tomography scans. Least absolute shrinkage and selection operator regression was used for feature selection. Model performance was evaluated using the area under the curve (AUC), accuracy, sensitivity, and specificity.
RESULTS: The Combined model integrating clinical, dosimetric, and radiomic features showed higher predictive performance (AUC: training 0.783, test 0.767, total 0.779) than the Clinical/Dosimetric (AUC: training 0.730, test 0.671, total 0.717) and Radiomics-only (AUC: training 0.721, test 0.668, total 0.708) models. The Combined model also achieved a higher accuracy (training 78.9%, test 78.2%, total 78.8%), sensitivity (training 74.6%, test 62.5%, total 72.0%), and specificity (training 79.7%, test 80.3%, total 79.9%) than the other models. DeLong's test confirmed that the Combined model significantly outperformed the Clinical/Dosimetric model (p=0.036 in training and p=0.010 in all datasets).
CONCLUSION: Integrating radiomic features with clinical and dosimetric factors showed potential to enhance lymphedema prediction in patients with breast cancer receiving PORT. This model can potentially guide personalized treatment strategies and improve patient outcomes.
MATERIALS AND METHODS: We designed a predictive model for lymphedema using clinical, dosimetric, and radiomic features. We included 532 patients (399 training and 133 testing) who underwent breast cancer surgery followed by PORT. Radiomic features were extracted from axillary levels I, II, III, and supraclavicular regions, which were automatically contoured on PORT-planning computed tomography scans. Least absolute shrinkage and selection operator regression was used for feature selection. Model performance was evaluated using the area under the curve (AUC), accuracy, sensitivity, and specificity.
RESULTS: The Combined model integrating clinical, dosimetric, and radiomic features showed higher predictive performance (AUC: training 0.783, test 0.767, total 0.779) than the Clinical/Dosimetric (AUC: training 0.730, test 0.671, total 0.717) and Radiomics-only (AUC: training 0.721, test 0.668, total 0.708) models. The Combined model also achieved a higher accuracy (training 78.9%, test 78.2%, total 78.8%), sensitivity (training 74.6%, test 62.5%, total 72.0%), and specificity (training 79.7%, test 80.3%, total 79.9%) than the other models. DeLong's test confirmed that the Combined model significantly outperformed the Clinical/Dosimetric model (p=0.036 in training and p=0.010 in all datasets).
CONCLUSION: Integrating radiomic features with clinical and dosimetric factors showed potential to enhance lymphedema prediction in patients with breast cancer receiving PORT. This model can potentially guide personalized treatment strategies and improve patient outcomes.