Radiology and oncology · 2026
Dual-channel ultrasonic images empowered deep learning: significantly improving prediction of occult central lymph node metastases in solitary papillary thyroid microcarcinoma.
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.
Li Meihua, Jia Chao, Li Gang, Shi Qiusheng, Liu Long
Аннотация
BACKGROUND: Central lymph node metastasis (CLNM) significantly elevates the risk of postoperative recurrence and contributes to ongoing debates regarding the necessity of prophylactic dissection in clinically node-negative papillary thyroid microcarcinoma (PTMC). Therefore, accurate preoperative prediction of occult CLNM is crucial for tailoring individualized treatment strategies.
PATIENTS AND METHODS: This retrospective study included 461 patients with PTMC from two hospitals who underwent preoperative ultrasound. A dual-channel deep learning (DL) model was developed by combining longitudinal and transverse ultrasound images. The model's performance was compared with single-direction DL models and a clinical model using machine learning classifiers. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) and calibration curves.
RESULTS: The dual-channel DL model outperformed the single-direction models, with AUC values of 0.765 in the training set and 0.726 in the external test set. The combined model, which integrated DL features and clinical indicators, achieved the highest AUC of 0.900 in the training set and 0.873 in the external test set, surpassing both the deep learning model using fused DL model (DL_F) and clinical models.
CONCLUSIONS: The dual-channel DL model demonstrated superior performance in predicting occult CLNM in PTMC patients. When combined with clinical features, it offers a robust tool for personalized risk stratification and treatment decision-making, providing a non-invasive method for predicting occult CLNM and supporting individualized treatment strategies.
PATIENTS AND METHODS: This retrospective study included 461 patients with PTMC from two hospitals who underwent preoperative ultrasound. A dual-channel deep learning (DL) model was developed by combining longitudinal and transverse ultrasound images. The model's performance was compared with single-direction DL models and a clinical model using machine learning classifiers. Performance was evaluated using the area under the receiver operating characteristic curve (AUC) and calibration curves.
RESULTS: The dual-channel DL model outperformed the single-direction models, with AUC values of 0.765 in the training set and 0.726 in the external test set. The combined model, which integrated DL features and clinical indicators, achieved the highest AUC of 0.900 in the training set and 0.873 in the external test set, surpassing both the deep learning model using fused DL model (DL_F) and clinical models.
CONCLUSIONS: The dual-channel DL model demonstrated superior performance in predicting occult CLNM in PTMC patients. When combined with clinical features, it offers a robust tool for personalized risk stratification and treatment decision-making, providing a non-invasive method for predicting occult CLNM and supporting individualized treatment strategies.