Differentiation of intraductal papillary mucinous neoplasms and pancreatic ductal adenocarcinoma using arterial-phase CT radiomics combined with clinical features.
Аннотация
OBJECTIVE
To investigate the value of arterial-phase CT-based radiomics combined with clinical features in differentiating IPMN (Intraductal Papillary Mucinous Neoplasms) from PDAC (Pancreatic Ductal Adenocarcinoma).
METHODS
A total of 216 patients with pathologically confirmed IPMN or PDAC were retrospectively enrolled. Clinical data and contrast-enhanced CT images were collected. Patients were divided into a training cohort, a test cohort and the external validation cohort. Univariate and multivariate analyses were performed on clinical variables and CT features to identify independent predictors. Regions of interest (ROIs) were manually delineated using ITK-SNAP software, and radiomics features were extracted with the Pyradiomics package. Feature dimensionality reduction and selection were conducted using the least absolute shrinkage and selection operator (LASSO) method. A radiomics score was calculated, and radiomics and combined models were constructed using a random forest (RF) algorithm. The diagnostic performance and clinical utility of each model were evaluated.
RESULTS
Multivariate analysis identified single cystic lesion (OR = 2.33), mural nodules (OR = 2.69), and CA19-9 level (OR = 2.36) as independent factors for differentiating IPMN from PDAC (all P < 0.05). Radiomics features extracted from arterial-phase contrast-enhanced CT images were used to construct a radiomics model. A combined model incorporating both radiomics and clinical features demonstrated superior predictive performance. The area under the receiver operating characteristic curve (AUC) of the combined model was 0.935 (95% CI: 0.898-0.974) in the training cohort and 0.931 (95% CI: 0.846-1.000) in the test cohort, which was significantly higher than that of the radiomics model alone (AUC = 0.885 and 0.827, respectively). The AUC of the external validation cohort was 0.808 (95% CI:0.684-0.968), and calibration curves showed good agreement between predicted and observed outcomes. Decision curve analysis indicated that the combined model provided greater clinical benefit.
CONCLUSION
A predictive model based on arterial-phase CT radiomics combined with clinical features can effectively differentiate IPMN from PDAC.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
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.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.