There is a paucity of data regarding initial bacterial colonization on admission to Intensive Care Units (ICUs) in low and middle-income countries (LMICs). Patients admitted to ICUs in LMICs are at high-risk of subsequent infection with antimicrobial-resistant organisms (AROs). We conducted a prospective, observational study at the Hospital for Tropical Diseases in Ho Chi Minh City, Vietnam from November 2014 to January 2016 to assess the colonization and antimicrobial susceptibility of Staphylococcus aureus, Escherichia coli, Klebsiella spp., Pseudomonas spp. and Acinetobacter spp. among adult patients within 48 hours of ICU admission. We found the admission colonization prevalence (with at least one of the identified organisms) was 93.7% (785/838) and that of AROs was 63.1% (529/838). The colonization frequency with AROs among patients admitted from the community was comparable to those transferred from other hospitals (62.2% vs 63.8%). Staphylococcus aureus was the most commonly isolated bacteria from nasal swabs (13.1%, 110/838) and the methicillin-resistant Staphylococcus aureus nasal colonization prevalence was 8.6% (72/838). We isolated Escherichia coli from rectal swabs from almost all enrolled patients (88.3%, 740/838) and 52.1% (437/838) of patients were colonized by extended spectrum β-lactamase producing Escherichia coli. Notably, Klebsiella pneumoniae was the most frequently isolated bacteria from the tracheal swabs (11.8%, 18/153). Vietnamese ICU patients have a high rate of colonization with AROs and are thus at risk of subsequent infections with these organisms if good infection control practices are not in place.

No abstract

Anomaly detection in the sound from machines is an important task in machine monitoring. An autoencoder architecture based on the reconstruction error using a log-Mel spectrogram feature is a conventional approach for this domain. However, because of the non-stationary nature of some sounds from the target machine, such a conventional approach does not perform well in those circumstances. In this paper, we propose a novel approach regarding the choice of used features and a new auto-encoder architecture. We created the Mixed Feature, which is a mixture of different sound representations, and a new deep learning method called Fully-Connected U-Net, a form of autoencoder architecture. With experiments on the same dataset as the baseline system, using the same architecture for all types of machines, the experimental results showed that our methods outperformed the baseline system in terms of the AUC and pAUC evaluation metrics. The optimized model achieved 83.38% AUC and 64.51% pAUC on average overall machine types on the developed dataset and outperformed the published baseline by 13.43% AUC and 8.13% pAUC.

Background: Fusarium root rot disease in Indian mulberry (Morinda officinalis How.) (FRRBK), caused by Fusarium proliferatum (FP), is widespread and responsible for serious economic losses in Viet Nam. The efficacy of a new bioproduct named MICROTECH-1(NL) is compared with other commercial products for suppression of FP under in vitro, pot, nursery as well as in the field conditions. Results: In in vitro antagonistic assay, MICROTECH-1(NL) significantly inhibited the mycelial growth of FP (72.38%). Under pot conditions, the efficacy of all the bio-products was significantly higher when applied prior to pathogen inoculation. The disease severity of treatments with double application of MICROTECH-1(NL) (applied both in the nursery and in the pot soil) was only 15.56%, significantly lower than control (80%). Thus, the application of MICRO-TECH-1(NL) significantly reduced the incidence of FP and markedly increased the number of plant beneficial bacteria and actinobacteria in rhizoplane of M. officinalis compared to untreated control. In the field conditions, double application of MICROTECH-1(NL) (both in the nursery and in the field soils) significantly decreased disease severity in comparison to single application in nursery or field. Conclusion: The most effective treatment was double application of MICROTECH-1(NL), which significantly reduced the disease severity and FP population in roots of M. officinalis and increased the population of plant beneficial microbes.

Tóm tắt. Rút trích cụm danh từ song ngữ là một trong những bài toán quan trọng trong xử lý ngôn ngữ tự nhiên (NLP). Bài toán này càng trở nên khó khăn hơn với cặp song ngữ Anh-Việt do thiếu vắng nguồn tài nguyên tiếng Việt bao gồm các công cụ xử lý ngôn ngữ tự nhiên như treebanks, part-of-speech taggers, parsers và dữ liệu huấn luyện có chú giải. Trong bài báo này, chúng tôi đề xuất một mô hình tổ hợp sử dụng đặc tính ngôn ngữ đích để rút trích cụm danh từ song ngữ qua phương pháp chiếu trên kết quả đối sánh từ bằng phương pháp thống kê. Đặc tính ngôn ngữ đích được sử dụng trong mô hình này là phân đoạn từ, trật tự từ và phân lớp từ [1]. Mô hình của chúng tôi không những khắc phục được sự thiếu vắng nguồn tài nguyên cho xử lý ngôn ngữ tự nhiên tiếng Việt mà còn cải thiện được kết quả do đối sánh rỗng, đối sánh lỗi, vấn đề chồng chéo và xung đột của phương pháp chiếu. Mô hình đề xuất có thể được áp dụng cho các cặp ngôn ngữ khác. Thực nghiệm trên 66.646 cặp câu song ngữ Anh-Việt, mô hình đề xuất cho kết quả rất khả quan.Từ khóa. Npbase, từ phân lớp, trật tự từ, NLP Abstract. Bilingual Base Noun Phrase (BaseNP) extraction is one of the key tasks of Natural Language Processing (NLP). This task is more challenging for the pair of English-Vietnamese due to the lack of available Vietnamese language resources such as treebanks, part-of-speech taggers, and parsers. In this paper, we propose a combination model that uses language characteristics based on statistics and projection method to extract BaseNP correspondences from a bilingual corpus. The language characteristics used in this model include the word segmentation, word order and word classification [1]. Our model not only overcomes the lack of resources of Vietnamese but also improves the performance of miss-alignment, null-alignment, overlap and conflict projection of the existing methods. The proposed model can be easily applied to another language pairs. Experiment on 66,646 pairs of sentences in the English-Vietnamese bilingual corpus shows that our proposed model is very satisfactory.