Should All Pancreatic Cystic Lesions with Worrisome or High-Risk Features Be Resected? A Clinical and Radiological Machine Learning Model May Help to Answer.

RATIONALE AND OBJECTIVES: According to current guidelines, pancreatic cystic lesions (PCLs) with worrisome or high-risk features may have overtreatment. The purpose of this study was to build a clinical and radiological based machine-learning (ML) model to identify malignant PCLs for surgery among preoperative PCLs with worrisome or high-risk features. MATERIALS AND METHODS: Clinical and radiological details of 317 pathologically confirmed PCLs with worrisome or high-risk features were retrospectively analyzed and applied to ML models including Support Vector Machine, Logistic Regression (LR), Decision Tree, Bernoulli NB, Gaussian NB, K Nearest Neighbors and Linear Discriminant Analysis. The diagnostic ability for malignancy of the optimal model with the highest diagnostic AUC in the cross-validation procedure was further evaluated in internal (n = 77) and external (n = 50) testing cohorts, and was compared to two published guidelines in internal mucinous cyst cohort. RESULTS: Ten clinical and radiological feature-based LR model was the optimal model with the highest AUC (0.951) in the cross-validation procedure. In the internal testing cohort, LR model reached an AUC, accuracy, sensitivity, and specificity of 0.927, 0.909, 0.914, and 0.905; in the external testing cohort, LR model reached 0.948, 0.900, 0.963, and 0.826. When compared to the European guidelines and the ACG guidelines, LR model demonstrated significantly better accuracy and specificity in identifying malignancy, while maintaining the same high sensitivity. CONCLUSION: Clinical- and radiological-based LR model can accurately identify malignant PCLs in patients with worrisome or high-risk features, possessing diagnostic performance better than the European guidelines as well as ACG guidelines.

Transformer guided progressive fusion network for 3D pancreas and pancreatic mass segmentation.

Pancreatic masses are diverse in type, often making their clinical management challenging. This study aims to address the task of various types of pancreatic mass segmentation and detection while accurately segmenting the pancreas. Although convolution operation performs well at extracting local details, it experiences difficulty capturing global representations. To alleviate this limitation, we propose a transformer guided progressive fusion network (TGPFN) that utilizes the global representation captured by the transformer to supplement long-range dependencies lost by convolution operations at different resolutions. TGPFN is built on a branch-integrated network structure, where the convolutional neural network and transformer branches first perform separate feature extraction in the encoder, and then the local and global features are progressively fused in the decoder. To effectively integrate the information of the two branches, we design a transformer guidance flow to ensure feature consistency, and present a cross-network attention module to capture the channel dependencies. Extensive experiments with nnUNet (3D) show that TGPFN improves the mass segmentation (Dice: 73.93% vs. 69.40%) and detection accuracy (detection rate: 91.71% vs. 84.97%) on 416 private CTs, and also obtains performance improvements of mass segmentation (Dice: 43.86% vs. 42.07%) and detection (detection rate: 83.33% vs. 71.74%) on 419 public CTs.

[Guidelines for Diagnosis and Treatment of Pancreatic Cystic Neoplasms Based on Radiology].

As the detection rate of pancreatic cystic neoplasms (PCN) increases,recommendations or guidelines for the diagnosis and treatment of PCN have been released from professional organizations.From the perspective of radiology,we compared seven guidelines in terms of general introduction,preoperative monitoring methods and strategies,stratification of risk factors,surgical indications,and postoperative follow-ups,aiming to provide references for the evaluation of images and the formulation of individualized approach for the treatment of PCN.

Diffuse Involvement of Pancreas is not Always Autoimmune Pancreatitis.

RATIONALE AND OBJECTIVES: To determine the prevalence of diffuse involvement of pancreas and to identify the findings of malignancies using enhancement computed tomography (CT). MATERIALS AND METHODS: A total of 1,0249 patients performed enhancement CT in our hospital over 62 months were investigated and the final study cohort includes 245 patients (170 males, 75 females; mean age, 56.94 ± 12.17 years). The reference standard is the final clinical/pathological diagnosis. The lesion-to-aorta enhancement ratio (LAR) on the pancreatic arterial phase, portal phase and delayed phase (DP) and the traditional CT findings were evaluated. Intergroup comparisons between malignancies and non-malignancies lesions were performed. Univariate and multivariate analyses were conducted to identify findings predicting malignancies. RESULTS: The prevalence of malignancy was 45.3% (111/245) of diffuse enlargement of pancreas. All benign lesions were autoimmune pancreatitis 54.7% (n = 134). The most common malignant lesion was pancreatic ductal adenocarcinoma (n = 88, 35.9%). Other rare lesions with malignant potential included pancreatic neuroendocrine tumor (n = 11, 4.5%), lymphoma (n = 4, 1.6%), metastasis (n = 4, 1.6%), solid pseudopapillary neoplasm (n = 3, 1.2%) and acinar cell carcinoma (n = 1, 0.4%). Residual normal pancreas parenchyma, heterogeneity, short axis (cut-off value, 3.15 cm) and LARDP (cut-off value, 0.75) were independent predictors of malignancies. When the above predictors were combined, a sensitivity of 94.2%, a specificity of 90.8% were attained. CONCLUSION: Diffuse involvement of the pancreas is rare and is not a specific sign of autoimmune pancreatitis, and it is associated with a wide spectrum of malignant conditions. Dynamic enhancement CT is helpful to identifying malignancies.

A deep learning algorithm to improve readers' interpretation and speed of pancreatic cystic lesions on dual-phase enhanced CT.

PURPOSE: To develop a deep learning model (DLM) to improve readers' interpretation and speed in the differentiation of pancreatic cystic lesions (PCLs) on dual-phase enhanced CT, and a low contrast media dose, external testing set validated the model. MATERIALS AND METHODS: Dual-phase enhanced CT images of 363 patients with 368 PCLs obtained from two centers were retrospectively assessed. Based on the examination date, a training and validation set of 266 PCLs, an internal testing set of 52 PCLs were designated from center 1. An external testing set included 50 PCLs from center 2. Clinical and radiological characteristics were compared. The DLM was developed using 3D specially designed densely connected convolutional networks for PCL differentiation. Radiomic features were extracted to build a traditional radiomics model (RM). Performance of the DLM, traditional RM, and three readers was compared. RESULTS: The accuracy for differential diagnosis was 0.904 with DLM, which was the highest in the internal testing set. Accuracy differences between the DLM and senior radiologist were not significant both in the internal and external testing set (both p > 0.05). With the help of the DLM, the accuracy and specificity of the junior radiologist were significantly improved (all p < 0.05), and all readers' diagnostic time was shortened (all p < 0.05). CONCLUSION: The DLM achieved senior radiologist-level performance in differentiating benign and malignant PCLs which could improve the junior radiologist's interpretation and speed of PCLs on CT.

M3Net: A multi-scale multi-view framework for multi-phase pancreas segmentation based on cross-phase non-local attention.

The complementation of arterial and venous phases visual information of CTs can help better distinguish the pancreas from its surrounding structures. However, the exploration of cross-phase contextual information is still under research in computer-aided pancreas segmentation. This paper presents M3Net, a framework that integrates multi-scale multi-view information for multi-phase pancreas segmentation. The core of M3Net is built upon a dual-path network in which individual branches are set up for two phases. Cross-phase interactive connections bridging the two branches are introduced to interleave and integrate dual-phase complementary visual information. Besides, we further devise two types of non-local attention modules to enhance the high-level feature representation across phases. First, we design a location attention module to generate cross-phase reliable feature correlations to suppress the misalignment regions. Second, the depth-wise attention module is used to capture the channel dependencies and then strengthen feature representations. The experiment data consists of 224 internal CTs (106 normal and 118 abnormal) with 1 mm slice thickness, and 66 external CTs (29 normal and 37 abnormal) with 5 mm slice thickness. We achieve new state-of-the-art performance with average DSC of 91.19% on internal data, and promising result with average DSC of 86.34% on external data.

Application of Compressed Sensing 3D MR cholangiopancreatography (CS-MRCP) with Contact-Free Physiological Monitoring (CFPM) for Pancreaticobiliary Disorders.

RATIONAL AND OBJECTIVES: To prospectively evaluate the clinical feasibility of the magnetic resonance cholangiopancreatography (MRCP) protocol using both contact-free physiological monitoring (CFPM) and compressed sensing (CS) (CS-CFPM-MRCP) and to compare its performance with that of the standard navigator-triggered (NT) CS-NT-MRCP and NT-MRCP. MATERIALS AND METHODS: A total of 63 patients (36 males, 27 females, age range: 18-83 years, mean age: 52.30 ± 15.70 years) suspected with duct-related pathologies were prospectively enrolled and performed the three MRCP protocols randomly. The acquisition time was compared. The pancreaticobiliary system was divided into 12 segments and evaluated based on a five-point Likert scale and compared by the Friedman test with a post hoc test. The diagnostic performance of the 3 MRCP was evaluated by the AUC value and compared by Delong's test. The interobserver agreement was evaluated by Kendall's W test. RESULTS: Compared to NT-MRCP, the acquisition time of CS-NT-MRCP and CS-CFPM-MRCP was significantly decreased (both p < 0.001). There is no significant difference in the overall imaging quality (p > 0.05) between the NT-MRCP and CS-CFPM-MRCP protocols. CS-CFPM-MRCP depicted pancreatic duct and intrahepatic ducts better than CS-NT-MRCP (all p < 0.05) and was comparable with that of the NT-MRCP (all p > 0.05). For identification of abnormalities and diseases associated with MPD anatomy, the mean AUC value for NT-MRCP and CS-CFPM-MRCP were 0.896 (95%CI: 0.834, 0.958) and 0.905 (95%CI: 0.846, 0.964), which were significantly higher when compared to that for CS-NT-MRCP (0.713 [95%CI:0.622, 0.805]) (p = 0.001 and < 0.001). All evaluations showed good to excellent agreement (0.619-0.897). CONCLUSION: The combination of CS and CFPM is considered feasible for shortening the scan time of 3D free breath MRCP without impairing the imaging quality and CS-CFPM-MRCP is considered feasible for patients suspected with pancreaticobiliary diseases.