Systemic autoimmune diseases represent a group of disorders characterized by loss of immune tolerance to self-antigens, leading to abnormal immune responses and subsequent tissue damage. Typical examples include systemic lupus erythematosus and systemic sclerosis. These conditions are marked by multi-system involvement, chronic progression, and recurrent flares. The pancreas, as a vital digestive and endocrine organ rich in glandular tissue and vascular supply, can also be affected by autoimmune processes. Pancreatic injury often indicates active and difficult-to-control disease, posing a serious threat to patient survival. Due to its relative rarity, diverse underlying mechanisms across different autoimmune diseases, and frequently nonspecific clinical presentations, pancreatic involvement is easily overlooked, resulting in delayed diagnosis and treatment.This article focuses on the clinical features and potential pathophysiological mechanisms of pancreatic injury associated with autoimmune diseases, such as systemic lupus erythematosus, systemic sclerosis, Sjögren's syndrome, and rheumatoid arthritis, aiming to enhance clinical awareness and facilitate early recognition and diagnosis of this condition.

To address the scarcity of annotated data for pediatric abdominal CT imaging and the insufficient generalization capability of existing models, we constructed a self-supervised pretraining architecture tailored for pediatric CT domain adaptation based on the visual foundation model DINOv3, and validated its performance in the task of pediatric abdominal multi-organ segmentation.

To develop a deep learning model based on arteriovenous dual-phase CT imaging features and evaluate its diagnostic value in differentiating benign from malignant pancreatic cystic lesions(PCLs).

To explore the clinical value of a multimodal predictive model based on multiparametric magnetic resonance imaging(MRI) radiomics combined with deep learning(DL) features for the preoperative noninvasive assessment of mismatch repair-deficient(MMRd) status in endometrial cancer(EC).

Objective:To investigate the value of nomogram model based on CT features in differentiating ectopic pancreas (EP) from gastrointestinal stromal tumors (GIST) with a long diameter less than 3 cm.Methods:This study was a case-control study. The clinical and imaging data of 43 patients with EP and 90 patients with GIST confirmed by pathology in the Affiliated Hospital of Guizhou Medical University from August 2013 to March 2024 were retrospectively analyzed. Preoperative CT images were analyzed to obtain qualitative features (number of lesions, location, morphology, growth pattern, borders, cystic degeneration, calcification, ulceration, catheter sign, central umbilication) and quantitative features (lesion long diameter, short diameter, long/short diameter, lesion and normal pancreas arterial-phase and venous-phase CT values, and enhancement ratio). Statistical analyses, including independent sample t-tests, Mann-Whitney U tests, χ2 tests, and Fisher exact tests, were performed to compare CT characteristics between the two groups. Binary logistic regression analysis was used to obtain independent predictors to identify the two groups, to establish a joint model, and to draw a nomogram. The discriminative performance of the independent predictors and the combined model was assessed using receiver operating characteristic (ROC) curves, while calibration curves were used to evaluate model fit. Results:The differences in age, location, morphology, border, catheter sign, central umbilication, short diameter, long/short diameter, arteriovenous phase enhancement CT value and arteriovenous phase enhancement ratio were statistically significant between the EP group and the GIST group (all P<0.05). The logistic analysis showed that the differences in age ( OR=0.920, 95% CI 0.885-0.956, P<0.001), border ( OR=5.994, 95% CI 2.111-17.022, P=0.001), long/short diameter ( OR=7.820, 95% CI 1.841-33.224, P=0.005), and venous phase enhancement ratio ( OR=8.847, 95% CI 1.103-70.972, P=0.040) were the independent predictors for distinguishing EP from GIST, and the area under the ROC curve (AUC) were 0.782 (95% CI 0.698-0.866), 0.684 (95% CI 0.600-0.767), 0.705 (95% CI 0.607-0.803), and 0.693 (95% CI 0.605-0.781), respectively. Combined age, border, long diameter/short diameter and venous phase enhancement ratio were plotted in a nomogram with an AUC of 0.881 (95% CI 0.817-0.945), sensitivity and specificity of 74.4% and 93.3%, respectively. The calibration curve demonstrated a strong agreement between predicted and actual probabilities (Hosmer-Lemeschow test, P=0.267). Conclusions:CT imaging reveals significant differences between EP and small GISTs (<3 cm). EP is more likely when patients are younger and lesions exhibit indistinct borders, a higher long-to-short diameter ratio, and greater venous-phase enhancement. The nomogram derived from CT features provides a valuable tool for differentiating EP from GIST.

Objective:To investigate the value of nomogram model based on CT features in differentiating ectopic pancreas (EP) from gastrointestinal stromal tumors (GIST) with a long diameter less than 3 cm.Methods:This study was a case-control study. The clinical and imaging data of 43 patients with EP and 90 patients with GIST confirmed by pathology in the Affiliated Hospital of Guizhou Medical University from August 2013 to March 2024 were retrospectively analyzed. Preoperative CT images were analyzed to obtain qualitative features (number of lesions, location, morphology, growth pattern, borders, cystic degeneration, calcification, ulceration, catheter sign, central umbilication) and quantitative features (lesion long diameter, short diameter, long/short diameter, lesion and normal pancreas arterial-phase and venous-phase CT values, and enhancement ratio). Statistical analyses, including independent sample t-tests, Mann-Whitney U tests, χ2 tests, and Fisher exact tests, were performed to compare CT characteristics between the two groups. Binary logistic regression analysis was used to obtain independent predictors to identify the two groups, to establish a joint model, and to draw a nomogram. The discriminative performance of the independent predictors and the combined model was assessed using receiver operating characteristic (ROC) curves, while calibration curves were used to evaluate model fit. Results:The differences in age, location, morphology, border, catheter sign, central umbilication, short diameter, long/short diameter, arteriovenous phase enhancement CT value and arteriovenous phase enhancement ratio were statistically significant between the EP group and the GIST group (all P<0.05). The logistic analysis showed that the differences in age ( OR=0.920, 95% CI 0.885-0.956, P<0.001), border ( OR=5.994, 95% CI 2.111-17.022, P=0.001), long/short diameter ( OR=7.820, 95% CI 1.841-33.224, P=0.005), and venous phase enhancement ratio ( OR=8.847, 95% CI 1.103-70.972, P=0.040) were the independent predictors for distinguishing EP from GIST, and the area under the ROC curve (AUC) were 0.782 (95% CI 0.698-0.866), 0.684 (95% CI 0.600-0.767), 0.705 (95% CI 0.607-0.803), and 0.693 (95% CI 0.605-0.781), respectively. Combined age, border, long diameter/short diameter and venous phase enhancement ratio were plotted in a nomogram with an AUC of 0.881 (95% CI 0.817-0.945), sensitivity and specificity of 74.4% and 93.3%, respectively. The calibration curve demonstrated a strong agreement between predicted and actual probabilities (Hosmer-Lemeschow test, P=0.267). Conclusions:CT imaging reveals significant differences between EP and small GISTs (<3 cm). EP is more likely when patients are younger and lesions exhibit indistinct borders, a higher long-to-short diameter ratio, and greater venous-phase enhancement. The nomogram derived from CT features provides a valuable tool for differentiating EP from GIST.

To establish a medical imaging teacher competency model and evaluate its application value in group teaching for undergraduates.

Pancreatic cancer, one of the most lethal cancers in the world, has been increasing in incidence and mortality year by year, and the overall prognosis of patients is poor. Early detection and effective treatment are crucial for improving the prognosis and survival rates of pancreatic cancer patients. Unlike traditional imaging, emerging molecular imaging can visualize the abnormalities at the molecular or cellular level in the process of tumor development. At present, multimodality molecular imaging that integrates multiple imaging methods to achieve complementary advantages and multifunctional nanoplatforms with integrated diagnosis and treatment functions have become research hotspots in the field of molecular imaging. Remarkable progress has been made in preclinical research concerning magnetic-optical multimodality molecular imaging probes and their derived multifunctional nanoplatforms, which provides new ideas for early detection, accurate treatment and efficacy evaluation of pancreatic cancer.

Objective:To construct a machine-learning model for predicting the progression of pancreatic cystic lesions (PCLs) based on clinical and CT features, and to evaluate its predictive performance in internal/external testing cohorts.Methods:Baseline clinical and radiological data of 200 PCLs in 177 patients undergoing abdominal thin slice enhanced CT examination at Peking Union Medical College Hospital from July 2014 to December 2022 were retrospectively collected. PCLs were divided into progressive and non-progressive groups according to whether the signs indicated for surgery by the guidelines of the European study group on PCLs were present during three-year follow-up. 200 PCLs were randomly divided into training (150 PCLs) and internal testing cohorts (50 PCLs) at the ratio of 1∶3. 15 PCLs in 14 patients at Jinling Affiliated Hospital of Medical School of Nanjing University from October 2011 to May 2020 were enrolled as external testing cohort. The clinical and CT radiological features were recorded. Multiple feature selection methods and machine-learning models were implemented and combined to identify the optimal machine-learning model based on the 10-fold cross-validation method. Receiver operating characteristics (ROC) curve was drawn and area under curve (AUC) was calculated. The model with the highest AUC was determined as the optimal model. The optimal model's predictive performance was evaluated on testing cohort by calculating AUC, sensitivity, specificity and accuracy. Permutation importance was used to assess the importance of optimal model features. Calibration curves of the optimal model were established to evaluate the model's clinical applicability by Hosmer-Lemeshow test.Results:In training and internal testing cohorts, the progressive and non-progressive groups were significantly different on history of pancreatitis, lesions size, main pancreatic duct diameter and dilation, thick cyst wall, presence of septation and thick septation (all P value <0.05) In internal testing cohort, the two groups were significantly different on gender, lesion calcification and pancreatic atrophy (all P value <0.05). In external testing cohort, the two groups were significantly different on lesions size and pancreatic duct dilation (both P<0.05). The support vector machine (SVM) model based on five features selected by F test (lesion size, thick cyst wall, history of pancreatitis, main pancreatic duct diameter and dilation) achieved the highest AUC of 0.899 during cross-validation. SVM model for predicting the progression of PCLs demonstrated an AUC of 0.909, sensitivity of 82.4%, specificity of 72.7%, and accuracy of 76.0% in the internal testing cohort, and 0.944, 100%, 77.8%, and 86.7% in the external testing cohort. Calibration curved showed that the predicted probability by the model was comparable to the real progression of PCLs. Hosmer-Lemeshow goodness-of-fit test affirmed the model's consistency with actual PCLs progression in testing cohorts. Conclusions:The SVM model based on clinical and CT features can help doctors predict the PCLs progression within three-year follow-up, thus achieving efficient patient management and rational allocation of medical resource.

As the detection rate of pancreatic cystic lesions(PCL)increases,artificial intelligence(AI)has made breakthroughs in the imaging workflow of PCL,including image post-processing,lesion detection,segmentation,diagnosis and differential diagnosis.AI-based image post-processing can optimize the quality of medical images and AI-assisted models for lesion detection,segmentation,diagnosis and differential diagnosis significantly enhance the work efficiency of radiologists.This article reviews the application progress of AI in PCL imaging and provides prospects for future research directions.