Surgical endoscopy · 2026
Radiomics analysis based on MRCP for predicting post-ERCP pancreatitis in patients with common bile duct stones.
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.
Chen Sheng, Jin Yiyi, Wu Yonghui, Tang Chaotao, Chen Youxiang
Аннотация
OBJECTIVES: Limited evidence exists on the predictive value of preoperative magnetic resonance cholangiopancreatography (MRCP) features of common bile duct stones (CBDS) for identifying post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP).
METHODS: Radiomics texture analysis was performed using 3D Slicer software (version 5.0.3) and the least absolute shrinkage and selection operator (LASSO) algorithm. Model performance was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and calibration plots.
RESULTS: A total of 1690 patients who underwent ERCP at our center were enrolled, among whom 134 (7.93%) developed PEP. In the training cohort, the complex model, combining radiomics and clinical factors, exhibited optimal predictive value. It achieved an area under the curve (AUC) of 0.863 (95% CI: 0.825-0.901), a sensitivity of 0.847 (95% CI: 0.757-0.909), and a specificity of 0.743 (95% CI: 0.688-0.792). The positive and negative likelihood ratios were 3.296 and 0.206, respectively. In contrast, the radiomics model showed moderate performance (AUC, 0.813, 95%CI: 0.767-0.859), while the clinical risk model performed the lowest performance (AUC,0.736, 95%CI: 0.678-0.794). Similarly, in the testing cohort, the complex model outperformed both the radiomics model and the clinical risk model. DeLong's test highlighted significant differences among the three models. Furthermore, decision curve analysis (DCA) and calibration curves indicated that both the complex model and the radiomics model surpassed the clinical risk model in terms of clinical applicability and value.
CONCLUSION: The radiomic features of CBDS are significant independent predictors of PEP. The model incorporating these radiomics features demonstrates strong performance for the clinical prediction of PEP.
METHODS: Radiomics texture analysis was performed using 3D Slicer software (version 5.0.3) and the least absolute shrinkage and selection operator (LASSO) algorithm. Model performance was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and calibration plots.
RESULTS: A total of 1690 patients who underwent ERCP at our center were enrolled, among whom 134 (7.93%) developed PEP. In the training cohort, the complex model, combining radiomics and clinical factors, exhibited optimal predictive value. It achieved an area under the curve (AUC) of 0.863 (95% CI: 0.825-0.901), a sensitivity of 0.847 (95% CI: 0.757-0.909), and a specificity of 0.743 (95% CI: 0.688-0.792). The positive and negative likelihood ratios were 3.296 and 0.206, respectively. In contrast, the radiomics model showed moderate performance (AUC, 0.813, 95%CI: 0.767-0.859), while the clinical risk model performed the lowest performance (AUC,0.736, 95%CI: 0.678-0.794). Similarly, in the testing cohort, the complex model outperformed both the radiomics model and the clinical risk model. DeLong's test highlighted significant differences among the three models. Furthermore, decision curve analysis (DCA) and calibration curves indicated that both the complex model and the radiomics model surpassed the clinical risk model in terms of clinical applicability and value.
CONCLUSION: The radiomic features of CBDS are significant independent predictors of PEP. The model incorporating these radiomics features demonstrates strong performance for the clinical prediction of PEP.