Mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN) account for a large portion of solitary pancreatic cystic neoplasms (PCN). In this study we implemented a convolutional neural network (CNN) model using ResNet50 to differentiate between MCN and SCN. The training data were collected retrospectively from 59 MCN and 49 SCN patients from two different hospitals. Data augmentation was used to enhance the size and quality of training datasets. Fine-tuning training approaches were utilized by adopting the pre-trained model from transfer learning while training selected layers. Testing of the network was conducted by varying the endoscopic ultrasonography (EUS) image sizes and positions to evaluate the network performance for differentiation. The proposed network model achieved up to 82.75% accuracy and a 0.88 (95% CI: 0.817–0.930) area under curve (AUC) score. The performance of the implemented deep learning networks in decision-making using only EUS images is comparable to that of traditional manual decision-making using EUS images along with supporting clinical information. Gradient-weighted class activation mapping (Grad-CAM) confirmed that the network model learned the features from the cyst region accurately. This study proves the feasibility of diagnosing MCN and SCN using a deep learning network model. Further improvement using more datasets is needed.

Endoscopic ultrasonography (EUS) plays an important role in diagnosing pancreatic cancer. Surgical therapy is critical to pancreatic cancer survival and can be planned properly, with the characteristics of the target cancer determined. The physical characteristics of the pancreatic cancer, such as size, location, and shape, can be determined by semantic segmentation of EUS images. This study proposes a deep learning approach for the segmentation of pancreatic cancer in EUS images. EUS images were acquired from 150 patients diagnosed with pancreatic cancer. A network with deep attention features (DAF-Net) is proposed for pancreatic cancer segmentation using EUS images. The performance of the deep learning models (U-Net, Attention U-Net, and DAF-Net) was evaluated by 5-fold cross-validation. For the evaluation metrics, the Dice similarity coefficient (DSC), intersection over union (IoU), receiver operating characteristic (ROC) curve, and area under the curve (AUC) were chosen. Statistical analysis was performed for different stages and locations of the cancer. DAF-Net demonstrated superior segmentation performance for the DSC, IoU, AUC, sensitivity, specificity, and precision with scores of 82.8%, 72.3%, 92.7%, 89.0%, 98.1%, and 85.1%, respectively. The proposed deep learning approach can provide accurate segmentation of pancreatic cancer in EUS images and can effectively assist in the planning of surgical therapies.

The gut microbiota are regarded as a functional organ that plays a substantial role in human health and disease. Proton pump inhibitors (PPIs) are widely used in medicine but can induce changes in the overall gut microbiome and cause disease-associated dysbiosis. The microbiome of the duodenum has not been sufficiently studied, and the effects of PPIs on the duodenal microbiome are poorly understood. In this study, we investigated the effect of PPI administration on duodenum microbiota in patients with a gastric ulcer. A total of 12 gastric ulcer patients were included, and PPI (Ilaprazole, Noltec®, 10 mg) was prescribed in all patients for 4 weeks. A total of 17 samples from the second portion of the duodenum were analyzed. Microbiome compositions were assessed by sequencing the V3–V4 region of the 16s rRNA gene (Miseq). Changes in microbiota compositions after 4 weeks of PPI treatment were analyzed. a-Diversity was higher after PPI treatment (p = 0.02, at Chao1 index), and β-diversity was significantly different after treatment (p = 0.007). Welch’s t-test was used to investigate changes in phyla, genus, and species level, and the abundance of Akkermansia muciniphila, belonging to the phylum Verrucomicrobia, and Porphyromonas endodontalis, belonging to the phylum Bacteroidetes, was significantly increased after treatment (p = 0.044 and 0.05). PPI administration appears to induce duodenal microbiome dysbiosis while healing gastric ulcers. Further large-scale studies on the effects of PPIs on the duodenal microbiome are required.

By far the most commo n cysti c lesion of th e pancreas is the inflammatory pseudocyst Amo ng th e tru e cysts , th e mu ci no us neopl asm is unique because of its proliferative nature and the pvtential for carcinomatous deg e neration. Hence , establishment of the diagnosis prior to surgical exploration is of g reat im portanceTh e imaging characteristics of ultrasonography and computed tomography are particularly useful in the preope rative differenti atio n of mu cinous cystic neoplasms from pse ud ocysts of th e pancreas Thi s report desc ribes the ultrasonograph ic and computed tom og raphic finding s of eight mucino us neoplasms (three cystade noma, fiv e cystadenocarcinoma) and twenty four pseud ocysts Mucinous cystade nomas are we ll -defin ed large multiloc ular cysti c masses co ntaining irreg u lar thick spe ta, solid components and local ly thicken ed wa ll . Sonographi c and computed tom ograph ic find ings of cystadenocarcino mas are si mi lar to those of cystadenoma, but cystade nocarcinomas fr equently show daughter cysts along the primary cystic wal!, locall y invasion and distant metastasis P se udocysts are unilocul ar cystic masses which wa lls are un iform in wid th, containing many debri s within the cysts and are often located at extrapancreatic areas

Background and Aims: Local ablative treatment is another option for improving outcomes and has been evaluated for locally advanced pancreatic cancer. We previously suggested endoscopic ultrasound (EUS)-guided interstitial laser ablation using a cylindrical laser diffuser (CILA) might be a feasible therapeutic option based on experiments performed on pancreatic cancer cell lines and porcine model with a short follow-up (3 days). The aim of this study was to investigate the safety of EUS-CILA performed using optimal settings in porcine pancreas with a long-term follow-up (2 weeks). Methods: EUS-CILA (laser energy of 450 J; 5 W for 90 s) was applied to normal pancreatic tissue in porcine (n = 5) under EUS guidance. Animals were observed clinically for 2 weeks after EUS-CILA to evaluate complications. Computed tomography and laboratory tests were carried out to evaluate safety. Two weeks after EUS-CILA, all pigs were sacrificed, and histopathological safety and efficacy evaluations were conducted. Results: EUS-CILA was technically successful in all five cases. No major complications occurred during the follow-up period. Body weight of porcine did not change during the study period without any significant change in feed intake. Animals remained in excellent condition throughout the experimental period, and laboratory tests and computed tomography (CT) scans provided no evidence of a major complication. Histopathological evaluation showed complete ablation in the ablated area with clear delineation of surrounding normal pancreatic tissue. Mean ablated volume was 55.5 mm2 × 29.0 mm and mean ablated areas in the pancreatic sections of the five pigs were not significantly different (p = 0.368). Conclusions: In conclusion, our experimental study suggests that EUS-CILA is safe and has the potential to be an effective local treatment modality. No major morbidity or mortality occurred during the study period. Further evaluations are warranted before clinical application.