TeKo: Text-Rich Graph Neural Networks With External Knowledge.

Graph neural networks (GNNs) have gained great prevalence in tackling various analytical tasks on graph-structured data (i.e., networks). Typical GNNs and their variants adopt a message-passing principle that obtains network representations by the attribute propagates along network topology, which however ignores the rich textual semantics (e.g., local word-sequence) that exist in numerous real-world networks. Existing methods for text-rich networks integrate textual semantics by mainly using internal information such as topics or phrases/words, which often suffer from an inability to comprehensively mine the textual semantics, limiting the reciprocal guidance between network structure and textual semantics. To address these problems, we present a novel text-rich GNN with external knowledge (TeKo), in order to make full use of both structural and textual information within text-rich networks. Specifically, we first present a flexible heterogeneous semantic network that integrates high-quality entities as well as interactions among documents and entities. We then introduce two types of external knowledge, that is, structured triplets and unstructured entity descriptions, to gain a deeper insight into textual semantics. Furthermore, we devise a reciprocal convolutional mechanism for the constructed heterogeneous semantic network, enabling network structure and textual semantics to collaboratively enhance each other and learn high-level network representations. Extensive experiments illustrate that TeKo achieves state-of-the-art performance on a variety of text-rich networks as well as a large-scale e-commerce searching dataset.

Perinatal outcomes in dichorionic diamniotic twins with multifetal pregnancy reduction versus expectant management: A systematic review and meta-analysis.

BACKGROUND: Published findings on perinatal outcomes of multifetal pregnancy reduction (MPR) of dichorionic diamniotic (DCDA) twin pregnancy to singleton are controversial. We performed a meta-analysis to appraise the effects of MPR of DCDA twin pregnancy versus expectant management on perinatal outcomes. METHODS: Four electronic databases were searched from their inception to June 15, 2019, to identify publications that appraised MPR before 15 weeks of gestation. Studies reporting perinatal outcomes of both MPR of DCDA twin pregnancy to singleton and expectant management were considered. The relative risks (RRs) and mean differences with 95% confidence intervals (CIs) were pooled using a random-effects model. RESULTS: Six studies involving 7398 participants showed that MPR of DCDA twin pregnancy to singleton was associated with a lower risk of preterm birth (5 studies with 7297 participants; RR: 0.30, 95% CI: 0.22-0.40; P < .001) and higher birth weight (4 studies with 5763 participants; mean differences: 548.10 g, 95% CI: 424.04-672.15; P < .001) than expectant management; there was no difference in the occurrence of miscarriages (5 studies with 7355 participants; RR: 1.57, 95% CI: 0.90-2.75; P = .11). Sensitivity analysis showed that all the results were stable and reliable, with the omission of 2 studies with serious risk of bias. CONCLUSION: Compared to expectant management, MPR of DCDA twin pregnancy to singleton prevents preterm birth and low birth weight, without increasing the risk of miscarriages. Regarding perinatal morbidity related to preterm birth, MPR can be reserved as a remediation measure to improve the perinatal outcomes of DCDA twin pregnancies.

Oral Progestin Treatment for Early-Stage Endometrial Cancer: A Systematic Review and Meta-analysis.

OBJECTIVE: This study aimed to evaluate the efficacy and safety of oral progestin treatment for early-stage endometrial cancer. METHODS: We conducted a systematic review and meta-analysis of the proportions from observational studies. Original studies were selected if patients with early-stage endometrial cancer, especially those of reproductive age, were treated with oral progestin. We conducted searches on studies listed in MEDLINE, EMBASE, and Cochrane that were published through June 2014, and relevant articles were also searched. The methodological quality of the included studies was assessed using the Newcastle-Ottawa quality assessment scale. Funnel plots and metaregression analyses were used to assess bias. RESULTS: The final sample included 25 articles involving 445 patients. Based on a random-effects model, patients achieved a disease regression rate of 82.4% (95% confidence interval [CI], 75.3%-88.7%), a relapse rate of 25.0% (95% CI, 15.8%-35.2%), a pregnancy rate of 28.8% (95% CI, 22.5%-35.5%), and a live birth rate of 19.6% (95% CI, 12.8%-27.4%). Body weight gain, liver dysfunction, and abnormal blood coagulation test results were the most common treatment-related adverse effects. Only 2 disease-related deaths were reported during the follow-up duration. CONCLUSIONS: Based on the present systematic review and meta-analysis, oral progestin treatment is feasible and safe for patients of reproductive age.