No abstract

Background Conventionally, disconnected pancreatic duct syndrome is treated surgically. Endoscopic management is associated with lesser morbidity and mortality than that observed with surgery and shows similar success rates. However, limited data are available in this context. We evaluated the efficacy of endotherapeutic management for this syndrome. Methods We prospectively obtained data of patients with disconnected pancreatic duct syndrome between September 2008 and January 2016. Demographic and clinical data were assessed, and factors affecting clinical outcomes were statistically analyzed. Results Thirty-one patients underwent 40 endoscopic transpapillary procedures, and 1 patient developed an infection after prosthesis insertion. Etiological contributors to disconnected pancreatic duct syndrome were abdominal trauma (52%) and acute necrotizing pancreatitis (48%). The median interval between the appearance of pancreatic leaks and disconnected pancreatic duct syndrome was 6.6 months (range 0.5–84 months). The median follow-up after the last treatment procedure was 38 months (range 17–99 months). Patients with complete main pancreatic duct disruption in the body/tail showed a low risk of pancreatic atrophy ( P = 0.009). This study highlighted the significant correlation between endoscopic transpapillary drainage and clinical success ( P = 0.014). Conclusions Disconnected pancreatic duct syndrome is not an uncommon sequel of pancreatic injury, and much of the delayed diagnosis is attributable to a lack of knowledge regarding this disease. Endoscopic transpapillary intervention with ductal stenting is an effective and safe treatment for this condition.

<div>Retrosynthesis is the process of recursively decomposing target molecules into available building blocks. It plays an important role in solving problems in organic synthesis planning. To automate the retrosynthesis analysis, many retrosynthesis prediction methods have been proposed.</div><div>However, most of them are cumbersome and lack interpretability about their predictions.</div><div>In this paper, we devise a novel template-free algorithm, RetroXpert, for automatic retrosynthetic expansion by automating the procedure that chemists used to do.</div><div>Our method disassembles retrosynthesis into two steps: i) we identify the potential reaction center within the target molecule through a graph neural network and generate intermediate synthons; and ii) we predict the associated reactants based on the obtained synthons via a reactant generation model. </div><div>While outperforming the state-of-the-art baselines by a significant margin, our model also provides chemically reasonable interpretation.</div>