No abstract

Objective This study explored the feasibility of mesoplasty with end-to-side anastomosis in the treatment of different apple-peel mesenteric defects with high jejunal atresia. Methods A retrospective analysis was performed on 42 premature infants admitted to the hospital between 2014 and 2021. Prenatal ultrasound scans revealed bowel dilatation. The patients experienced vomiting after birth and produced white or no meconium. Plain radiography showed double or triple bubble signs and the patients underwent emergency laparotomy. High jejunal atresia with different apple-peel atresia appearance was discovered intraoperatively, involving mobilization of the ileocecal region. Patients received end-to-side anastomosis between the enlarged blind pouch and atretic bowel, as well as mesoplasty. A jejunal feeding tube was placed trans-nasally. Patients were discharged after achieving full enteral feeding. We also reviewed the literature on the subject. Results Three patients died and 39 survived. The discharged patients were followed up for 12 months, and none showed post-operative complications such as intestinal obstruction, malnutrition, or chronic diarrhea. All surviving patients reached the expected height and weight for children of the same age. Conclusion For cases of high jejunal atresia with apple-peel intestinal atresia, mesoplasty may be a good option to avoid postoperative volvulus.

Decision Support Systems (DSS) has become increasingly important due to its broad applications in various domains. Significant progresses have been made on ensuring more precise decision-making by leveraging appropriate data and knowledge from knowledge bases. However, the current DSSs related to antibiotics consider only therapy rather than diagnosis, and they were developed from a physician's perspective. Based on these two points, this study presents IDDAT, an ontology-driven decision support system for aiding Infectious Disease Diagnosis and Antibiotic Therapy. Based on patient-entered information, this freely accessible system aims to identify infectious disease, and provide an antibiotic therapy specifically adapted to the patient. We show the effectiveness of IDDAT by applying it to a diagnosis classification task. Experimental results reveal the system's advantages in term of the area under the curve (AUC) of receiver operating characteristic (ROC) (89.91%).

Extractive text summarization aims at extracting the most representative sentences from a given document as its summary. To extract a good summary from a long text document, sentence embedding plays an important role. Recent studies have leveraged graph neural networks to capture the inter-sentential relationship (e.g., the discourse graph) to learn contextual sentence embedding. However, those approaches neither consider multiple types of inter-sentential relationships (e.g., semantic similarity & natural connection), nor model intra-sentential relationships (e.g, semantic & syntactic relationship among words). To address these problems, we propose a novel Multiplex Graph Convolutional Network (Multi-GCN) to jointly model different types of relationships among sentences and words. Based on Multi-GCN, we propose a Multiplex Graph Summarization (Multi-GraS) model for extractive text summarization. Finally, we evaluate the proposed models on the CNN/DailyMail benchmark dataset to demonstrate the effectiveness and superiority of our method.

Human tackle reading comprehension not only based on the given context itself but often rely on the commonsense beyond. To empower the machine with commonsense reasoning, in this paper, we propose a Commonsense Evidence Generation and Injection framework in reading comprehension, named CEGI. The framework injects two kinds of auxiliary commonsense evidence into comprehensive reading to equip the machine with the ability of rational thinking. Specifically, we build two evidence generators: the first generator aims to generate textual evidence via a language model; the other generator aims to extract factual evidence (automatically aligned text-triples) from a commonsense knowledge graph after graph completion. Those evidences incorporate contextual commonsense and serve as the additional inputs to the model. Thereafter, we propose a deep contextual encoder to extract semantic relationships among the paragraph, question, option, and evidence. Finally, we employ a capsule network to extract different linguistic units (word and phrase) from the relations, and dynamically predict the optimal option based on the extracted units. Experiments on the CosmosQA dataset demonstrate that the proposed CEGI model outperforms the current state-of-the-art approaches and achieves the accuracy (83.6%) on the leaderboard.