deepMDDI: A deep graph convolutional network framework for multi-label prediction of drug-drug interactions.

It is crucial to identify DDIs and explore their underlying mechanism (e.g., DDIs types) for polypharmacy safety. However, the detection of DDIs in assays is still time-consuming and costly, due to the need for experimental search over a large space of drug combinations. Thus, many computational methods have been developed to predict DDIs, most of them focusing on whether a drug interacts with another or not. And a few deep learning-based methods address a more realistic screening task for identifying various DDI types, but they assume a DDI only triggers one pharmacological effect, while a DDI can trigger more types of pharmacological effects. Thus, here we proposed a novel end-to-end deep learning-based method (called deepMDDI) for the Multi-label prediction of Drug-Drug Interactions. deepMDDI contains an encoder derived from relational graph convolutional networks and a tensor-like decoder to uniformly model interactions. deepMDDI is not only efficient for DDI transductive prediction, but also inductive prediction. The experimental results show that our model is superior to other state-of-the-art deep learning-based methods. We also validated the power of deepMDDI in the DDIs multi-label prediction and found several new valid DDIs in the case study. In conclusion, deepMDDI is beneficial to uncover the mechanism and regularity of DDIs.

Deterministic full-scenario analysis for maximum credible earthquake hazards.

Great earthquakes are one of the major threats to modern society due to their great destructive power and unpredictability. The maximum credible earthquake (MCE) for a specific fault, i.e., the largest magnitude earthquake that may occur there, has numerous potential scenarios with different source processes, making the future seismic hazard highly uncertain. We propose a full-scenario analysis method to evaluate the MCE hazards with deterministic broadband simulations of numerous scenarios. The full-scenario analysis is achieved by considering all uncertainties of potential future earthquakes with sufficient scenarios. Here we show an application of this method in the seismic hazard analysis for the Xiluodu dam in China by simulating 22,000,000 MCE scenarios in 0-10 Hz. The proposed method can provide arbitrary intensity measures, ground-motion time series, and spatial ground-motion fields for all hazard levels, which enables more realistic and accurate MCE hazard evaluations, and thus has great application potential in earthquake engineering.

5S management improves the service quality in the outpatient-emergency pharmacy: from management process optimisation to staff capacity enhancement.

OBJECTIVE: As a high-efficiency demanding department in a hospital, the outpatient pharmacy has a great need for quality improvement to provide superior medical service for patients. Little is known about the application of 5S management in a hospital pharmacy department. The aim of this study was to evaluate the impacts of 5S management on pharmaceutical service quality and staff capacity in the outpatient-emergency pharmacy. METHODS: We carried out a 5S project in the outpatient-emergency pharmacy at a local hospital that involved processes including waste elimination, workplace standardisation, and optimisation of workflow and staff quality, and then evaluated the effects of the project. RESULTS: The equipment and items in the outpatient-emergency pharmacy were sorted. All the drugs were categorised and put in order. The redesigned workspace and standardised workflow during the project improved the accuracy and efficiency of drug dispensing. The satisfaction rate of patients regarding the pharmaceutical service quality in the outpatient-emergency pharmacy was elevated, as well as the satisfaction rate of pharmacists about their work experiences. The optimisation of objective conditions also stimulated a positive working attitude and professional ability promotion of pharmacists in the outpatient-emergency pharmacy. CONCLUSIONS: In this study, the 5S management method has proven useful for quality and efficiency improvement in the outpatient-emergency pharmacy, and could be generalised to other departments in a hospital, which provides further evidence of the advantages of the Lean tool in healthcare system management.

Mechanism for large-scale canyon deformations due to filling of large reservoir of hydropower project.

Large storage dam projects may modify geo-environmental conditions in many ways. The reservoir impoundment of the 285.5 m high Xiluodu arch dam located on the Jinsha River (China) caused large-scale canyon deformations, including significant canyon contraction and uplift movements from reservoir to downstream valley. The dam experienced subsequent tilting towards upstream and raised a safety concern of the project. A Thermo-Hydro-Mechanical (THM) mechanism is proposed for this extraordinary behavior. Due to reservoir impounding and seepage, significant temperature drops and fluid pressure increase within the underlying geothermal limestone aquifer in a synclinal basin are primary root causes. Finite element THM simulations successfully reproduce these unique deformations. Recent observations of large quantities of thermalized discharge water downstream and high pore pressure in the limestone layer provide further support for the proposed mechanism. Furthermore, refined numerical modeling is adopted to evaluate the safety of Xiluodu dam subjected to potential larger canyon contractions. We conclude that these unprecedented phenomena are dominantly the consequence of THM response to regional hydrogeological evolution following the build-up of a large reservoir. The accumulated canyon contractions at the current stage would not pose a direct threat to the dam safety, but a tripled situation may cause severe safety issues.