Abdominal radiology (New York) · 2026
Development and validation of a preoperative CT-based radiomics nomogram for predicting postpancreatectomy acute pancreatitis after pancreatoduodenectomy.
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 Hebing, Sun Limeng, Liu Jingyi, Cong Fuze, Li Xiang, Xu Qiang, Xue Huadan
Аннотация
PURPOSE: To investigate the value of radiomics based on preoperative contrast-enhanced computed tomography (CECT) for predicting postpancreatectomy acute pancreatitis (PPAP) after pancreatoduodenectomy (PD).
METHODS: A retrospective analysis included 104 patients who underwent PD and preoperative CECT between May 2019 and March 2025. Patients were randomly divided into training (n = 73) and test (n = 31) sets at a 7:3 ratio. PPAP was defined according to the biochemical and radiologic components of the International Study Group for Pancreatic Surgery (ISGPS) criteria. After pancreatic parenchyma delineation, radiomics feature extraction and selection, an optimal model was developed on the training set to calculate the radiomics score (rad-score). Independent clinical and morphological features were selected by univariate and multivariate logistic regression analyses. Conventional quantitative CT attenuation ratios of the pancreatic parenchyma were also evaluated. Predictive performance was assessed by receiver operating characteristic (ROC) analysis.
RESULTS: PPAP occurred in 41 patients. Four models were compared: a clinical-morphologic model (age and main pancreatic duct [MPD] dilation), a conventional model (age, MPD dilation and arterial-to-portal ratio [A/P]), a radiomics model (rad-score alone), and a radiomics-integrated model (age, MPD dilation and rad-score). The radiomics-integrated model achieved the highest area under the curve (AUC) among the four models, with the AUC of 0.854 (95% CI 0.758-0.951) in the training set and 0.862 (95% CI 0.732-0.993) in the test set.
CONCLUSION: The radiomics-integrated model demonstrated promising performance for preoperative prediction of PPAP after PD, which may facilitate individualized risk stratification and perioperative management.
METHODS: A retrospective analysis included 104 patients who underwent PD and preoperative CECT between May 2019 and March 2025. Patients were randomly divided into training (n = 73) and test (n = 31) sets at a 7:3 ratio. PPAP was defined according to the biochemical and radiologic components of the International Study Group for Pancreatic Surgery (ISGPS) criteria. After pancreatic parenchyma delineation, radiomics feature extraction and selection, an optimal model was developed on the training set to calculate the radiomics score (rad-score). Independent clinical and morphological features were selected by univariate and multivariate logistic regression analyses. Conventional quantitative CT attenuation ratios of the pancreatic parenchyma were also evaluated. Predictive performance was assessed by receiver operating characteristic (ROC) analysis.
RESULTS: PPAP occurred in 41 patients. Four models were compared: a clinical-morphologic model (age and main pancreatic duct [MPD] dilation), a conventional model (age, MPD dilation and arterial-to-portal ratio [A/P]), a radiomics model (rad-score alone), and a radiomics-integrated model (age, MPD dilation and rad-score). The radiomics-integrated model achieved the highest area under the curve (AUC) among the four models, with the AUC of 0.854 (95% CI 0.758-0.951) in the training set and 0.862 (95% CI 0.732-0.993) in the test set.
CONCLUSION: The radiomics-integrated model demonstrated promising performance for preoperative prediction of PPAP after PD, which may facilitate individualized risk stratification and perioperative management.