Zero pronoun resolution aims at recognizing dropped pronouns and pointing out their anaphoric mentions, while non-zero coreference resolution targets at clustering mentions referring to the same entity. Existing efforts often deal with the two problems separately regardless of their close essential correlations. In this paper, we investigate the possibility of jointly solving zero pronoun resolution and coreference resolution via a novel end-to-end neural model. Specifically, we design a gapmasked self-attention model that encodes gaps and tokens in the same space, where gaps could capture valuable contextual information according to their surrounding tokens while tokens could maintain original sequential information without disturbance. Additionally, we also propose a two-stage interaction mechanism to make full use of the exclusive relationship between zero pronouns and mentions. Our empirical study conducted on the OntoNotes 5.0 Chinese dataset shows that our model could outperform corresponding stateof-the-art approaches on both tasks.

No abstract

No abstract

ObjectiveThough the prognostic value of the model for end-stage liver disease (MELD) score in tricuspid surgery was confirmed, the unstable international normalized ratio (INR) may affect the evaluation effectiveness of the MELD score for isolated tricuspid valve replacement (ITVR). The aim of the study was to assess the prognostic value of modified MELD for ITVR.Methods and ResultsA total of 152 patients who underwent ITVR were evaluated. The adverse outcome was defined as in-hospital mortality after surgery. The receiver operating characteristic (ROC) curve analysis demonstrated that a modified MELD score with albumin replacing INR (MELD-albumin) score presented well prognostic value [area under the curve (AUC) = 0.731, p = 0.006] for in-hospital mortality. Through Cox regression and further interval validation, the MELD-albumin score was identified as an independent predictor for in-hospital mortality. The optimal cutoff value of MELD-albumin was identified as 8.58 through maximally selected log-rank statistics. In addition, restricted cubic spline analysis demonstrated the linear inverse relationship between MELD-albumin and hazard ratio (HR) for in-hospital mortality. Kaplan–Meier analysis illustrated that in-hospital mortality was increased significantly in the high MELD-albumin (MELD-albumin ≥8.58) group than in the low MELD-albumin group (MELD-albumin <8.58; p < 0.001). Furthermore, high MELD-albumin was associated with lower body mass index (BMI), the incidence of lower extremities edema and moderate drinking history, and the MELD-albumin score was correlated with the value of aspartate transaminase (AST), alanine transaminase (ALT), and albumin. Furthermore, the incidence of renal failure (p = 0.003) and pulmonary infection (p = 0.042) was increased significantly in the high MELD-albumin group.ConclusionThe MELD-albumin score could provide prognostic value for ITVR. In addition, the MELD-albumin score was useful in risk stratification and patient selection for patients with tricuspid regurgitation (TR) prior to ITVR.