Future oncology (London, England) · 2026
Machine learning prediction of axillary lymph node metastasis using multimodal ultrasound in breast cancer.
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.
Shen Yan, Yu Weiquan, Chen Qingqing, Hu Jiaojiao, Yu Xiaoqing, He Jie, Dong Jiangnan, Chu Xuchun, Yang Jingdong, Fu Xiaohong
Аннотация
OBJECTIVES: Accurate prediction of axillary lymph node metastasis (ALNM) is crucial for tailoring breast cancer treatments, this study aimed to develop a machine-learning model for predicting ALNM in patients with breast cancer based on multimodal ultrasound (MU) features.
MATERIALS AND METHODS: A total of 696 breast cancer patients were included From January 2016 to December 2022. Data and MU images were collected and incorporated using 9 machine learning models to predict ALNM. Model performance was evaluated by area under the receiver operating characteristics curve (AUC-ROC).
RESULTS: A total of 606 patients (238 with ALNM) were included in the training set. In addition, 90 patients (34 with ALNM) were included in the validation set. Among the nine algorithms, the predictive ability of the XGBoost model was the highest, with ROC-AUC, sensitivity, and specificity of 0.936, 0.951, and 0.949, respectively. External validation showed that the ROC-AUC, sensitivity, and specificity of the XGBoost model were 0.944, 1.000, and 0.850, respectively, indicated that these features should be paid special attention in a clinical scenario.
CONCLUSION: This study developed and validated a nomogram for predicting ALNM in patients with breast cancer based on MU features. This provides a promising tool for noninvasive, preoperative ALNM prediction.
MATERIALS AND METHODS: A total of 696 breast cancer patients were included From January 2016 to December 2022. Data and MU images were collected and incorporated using 9 machine learning models to predict ALNM. Model performance was evaluated by area under the receiver operating characteristics curve (AUC-ROC).
RESULTS: A total of 606 patients (238 with ALNM) were included in the training set. In addition, 90 patients (34 with ALNM) were included in the validation set. Among the nine algorithms, the predictive ability of the XGBoost model was the highest, with ROC-AUC, sensitivity, and specificity of 0.936, 0.951, and 0.949, respectively. External validation showed that the ROC-AUC, sensitivity, and specificity of the XGBoost model were 0.944, 1.000, and 0.850, respectively, indicated that these features should be paid special attention in a clinical scenario.
CONCLUSION: This study developed and validated a nomogram for predicting ALNM in patients with breast cancer based on MU features. This provides a promising tool for noninvasive, preoperative ALNM prediction.