FGraDA: A Dataset and Benchmark for Fine-Grained Domain Adaptation in Machine Translation

Previous research for adapting a general neural machine translation (NMT) model into a specific domain usually neglects the diversity in translation within the same domain, which is a core problem for domain adaptation in real-world scenarios. One representative of such challenging scenarios is to deploy a translation system for a conference with a specific topic, e.g., global warming or coronavirus, where there are usually extremely less resources due to the limited schedule. To motivate wider investigation in such a scenario, we present a real-world fine-grained domain adaptation task in machine translation (FGraDA). The FGraDA dataset consists of Chinese-English translation task for four sub-domains of information technology: autonomous vehicles, AI education, real-time networks, and smart phone. Each sub-domain is equipped with a development set and test set for evaluation purposes. To be closer to reality, FGraDA does not employ any in-domain bilingual training data but provides bilingual dictionaries and wiki knowledge base, which can be easier obtained within a short time. We benchmark the fine-grained domain adaptation task and present in-depth analyses showing that there are still challenging problems to further improve the performance with heterogeneous resources. © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.

[COVID-19 treated with oral Nirmatrelvir-Ritonavir in 3 children].

Objective: To summarize the application experience and the therapeutic effect of Nirmatrelvir-Ritonavir (trade name: Paxlovid) for COVID-19 in children. Methods: A retrospective analysis was performed on the clinical data, including collecting the clinical manifestations and clinical outcomes, dynamically monitoring the blood routine, hepatic and renal function and SARS-CoV-2 nucleic acid results, and observing the related side effects during the treatment, etc, of 3 cases with COVID-19 treated with Paxlovid admitted to Shanghai Children's Hospital (designated referral hospital for SARS-CoV-2 infection in Shanghai) from May 1st to June 1st, 2022. Results: The 3 cases were 12, 14, 17 years of age, among which 2 cases were males, 1 case was female. All 3 cases were mild cases with underlying diseases and risk of developing into severe COVID-19, with symptoms of high fever, sore throat and dry cough. The treatment of Paxlovid at 3rd day of symptom onset contributed to the symptom-free after 1-2 days and negative results of SARS-CoV-2 nucleic acid after 2-4 days. All patients had no adverse manifestations of gastrointestinal tract and nervous system but a case had little skin rashes, which recovered after the withdrawal of Paxlovid. Three cases had normal hepatic and renal function during the Paxlovid treatment. At 3 months after discharge, no clinical manifestations of post-COVID syndrome were found in all 3 cases. Conclusion: Paxlovid was effective and relatively safe in the treatment of 3 children with COVID-19.

TRENDS IN AVOIDABLE AND UNAVOIDABLE U.S. EMERGENCY DEPARTMENT VISITS DURING THE COVID-19 PANDEMIC, 2019-2021

BACKGROUND: Following early reductions in U.S. emergency department (ED) care after the COVID-19 pandemic's onset, it is unclear whether avoidable and unavoidable ED visits returned to expected rates, particularly for socioeconomically disadvantaged Medicaid and dual eligible MedicareMedicaid patients. Further, the degree to which avoidable and unavoidable ED visits are associated with hospitalizations during the pandemic is unknown. METHODS: In a retrospective cohort study of ED care patterns from Jan 1, 2019-Feb 28, 2021, we analyzed claims data from multiple U.S. payers from MedInsight's research database. Using difference-in-differences methods, we assessed the degree to which ED use during the pandemic differed from expected rates had the pandemic not occurred. We compared rate changes between Jan-Feb 2020 and each subsequent 2-month timeframe during the pandemic vs changes in the corresponding months in the year prior. We stratified visits by avoidable vs unavoidable using MediCal avoidable ED visit criteria, which flags visits that generally could have been managed by a primary care physician (e.g., rhinitis, cystitis, administrative exams) had patients had access to one. We tested the relationship between avoidable vs unavoidable ED visits and hospitalization, and estimated age-sex adjusted Poisson regressions of monthly use counts, offsetting for total patient-months and stratifying by insurance. RESULTS: We studied 14.5 million U.S. adults (mean age 53;55% female) using 11 million ED visits (6% avoidable, 94% unavoidable) in 2019-21. Unavoidable visits had higher odds of hospitalization than avoidable visits in each insurance group (for all groups: OR 2.8 [2.7-2.8]). In Mar-Apr 2020, avoidable and unavoidable ED visits respectively declined similarly, to 68.1% [67-69%] and 68.9% [68-69%] of expected rates. By Nov-Dec 2020, avoidable visits declined further to 52.0% [51-53%], while unavoidable visits returned to 93.7% [93-94%] of expected rates. During pandemic wave 2 in Jan-Feb 2021, avoidable and unavoidable visits declined to 45.1% [44-46%] and 82.2% [82-83%] respectively and varied by insurance (Table). CONCLUSIONS: Following declines early in the pandemic, unavoidable ED visits nearly returned to expected rates by Nov-Dec 2020, only to decline again during pandemic wave 2 in Jan-Feb 2021. In contrast, avoidable visits consistently declined without rebound. While reductions in avoidable visits lower unnecessary costs, persistent declines in unavoidable visits raise concern that patients with more serious conditions may be delaying needed ED care, particularly among socioeconomically disadvantaged groups.

An air quality prediction model based on deep learning and wavelet analysis considering the COVID-19 pandemic factors

Based on the air pollution data in China from January 1,2014 to December 31,2020, the characteristics of extreme value and period of air quality in different regions on different time scales were studied by using wavelet analysis. Wavelet coherence analysis was used to evaluate the relationship between air quality and meteorological factors in the period of COVID-19. We found that the spatial characteristics of air quality changed significantly in summer. Generally, air pollution is more severe in spring and winter. During the lockdown period, the overall air quality in the study area improved significantly. In general, except for 03, the concentration of all other pollutants has dropped considerably. The improvement in air quality is a direct result of emission reductions due to the implementation of the COVID-19 blockade, which is unsustainable in the long term. Eventually, a prediction model attention_CNN_LSTM based on deep learning method is proposed in this paper. The experimental results show that the attention proposed in this study the model has a good prediction effect in the long-term prediction of air quality, but the attention mechanism's impact is lower. After shortening the prediction period, the attention_CNN_LSTM model has good prediction performance on most data sets, with average MAPE = 2.67% and RMSE = 2.29.