A novel behavioral three-way decision model with application to the treatment of mild symptoms of COVID-19.

The Coronavirus Disease 2019 (COVID-19) has popularized since late December 2019. In present, it is still highly transmissible and has severe impact on the public health and global economy. Due to the lack of specific drug and the appearance of different variants, the selection of the antiviral therapy to treat the patients with mild symptom is of vital importance. Hence, in this paper, we propose a novel behavioral Three-Way Decision (3WD) model and apply it to the medicine selection decision. First, a new relative utility function is constructed by considering the risk-aversion behavior and regret-aversion behavior of human beings. Second, based on the relative utility function, some new rules are defined to calculate the thresholds and conditional probabilities in 3WD and some corresponding theorems are explored and proved. Next, a new information fusion mechanism in the framework of evidential reasoning algorithm is developed. Then, the decision results are obtained based on the Bayesian decision procedure and the principle of maximum utility. Finally, an example with large-scale data set and an example about medicine selection for COVID-19 are provided to show the implementation process and effectiveness of the proposed method. Comparative analysis and sensitivity analysis are also performed to illustrate the superiority and the robustness of the current proposal.

Delays in service for non-emergent patients due to arrival of emergent patients in the emergency department: a case study in Hong Kong.

BACKGROUND: In Hong Kong Emergency Departments (EDs), the timeliness of providing high-quality services has been compromised by the increasing attendance of non-emergent patients in addition to the unpredictable arrival of emergency patients. OBJECTIVES: We sought to quantify the impact of the presence of emergent patients and other related factors on the delay in service for non-emergent patients. METHODS: We conducted a retrospective study in patients who visited the ED of a large hospital in Hong Kong from July 1, 2009 to June 30, 2010. We estimated waiting and length of stay (LOS) for individual non-emergent patients registered during day and evening shifts. Using multiple linear regression, we estimated waiting time and LOS as a function of the presence of emergent patients and other related factors such as patient demographics and clinical factors. In particular, we evaluated the influence of the arrival or presence of emergent patients on the odds of violating the 120-min waiting time target for semi-urgent patients. RESULTS: The arrival of a new emergent patient prolonged the waiting time and LOS of a non-emergent patient by 14.9% (95% confidence interval [CI] 14.2-15.5) and 10.8% (95% CI 10.6-11.0), respectively. An additional patient-hour needed for an emergent patient increased the probability of violating the waiting time target for non-emergent patients (odds ratio 2.3, 95% CI 2.2-2.4). CONCLUSIONS: The arrival of an emergent patient significantly prolonged the waiting time and LOS for non-emergent patients. Discouraging non-urgent ED utilization and building a real-time decision-support system are critical methods needed to relieve staff pressure and guide contingent resource reallocation when emergent patients arrive.

Tracking traffic congestion and accidents using social media data: A case study of Shanghai.

Traffic congestion and accidents take a toll on commuters' daily experiences and society. Locating the venues prone to congestion and accidents and capturing their perception by public members is invaluable for transport policy-makers. However, few previous methods consider user perception toward the accidents and congestion in finding and profiling the accident- and congestion-prone areas, leaving decision-makers unaware of the subsequent behavior responses and priorities of retrofitting measures. This study develops a framework to identify and characterize the accident- and congestion-prone areas heatedly discussed on social media. First, we use natural language processing and deep learning to detect the accident- and congestion-relevant Chinese microblogs posted on Sina Weibo, a Chinese social media platform. Then a modified Kernel Density Estimation method considering the sentiment of microblogs is employed to find the accident- and congestion-prone regions. The results show that the 'congestion-prone areas' discussed on social media are mainly distributed throughout the historical urban core and the Northwest of Pudong New Area, in reasonably good agreements with actual congestion records. In contrast, the 'accident-prone areas' are primarily found in locations with severe accidents. Finally, the above venues are characterized in spatio-temporal and semantic aspects to understand the nature of the incidents and assess the priority level for mitigation measures. The outcomes can provide a reference for traffic authorities to inform resource allocation and prioritize mitigation measures in future traffic management.

A Pocket-Textured Surface for Improving the Tribological Properties of Point Contact under Starved Lubrication.

This paper reports a novel pocket-textured surface for improving the tribological properties of point contact under starved lubrication by possibly storing and releasing oil, and homogenizing the surface contact pressure. The ball-on-disk experimental results confirmed the coefficient of friction (COF) and wear reduction effect of such pocket-texturing. The maximum reduction rate was 40% compared with a flat surface under the same operating conditions. Analyses on experimental results attributed the oil storage effect and enhanced the secondary lubrication effect within the starved lubrication state, to become the main mechanism. In addition, the plate elasticity and the Hertzian contact principles were employed to estimate the pressure and the load acting on the surface. The experimental results and numerical analysis substantiated the design of pocket-textured surface, making it likely to enlarge about 50% of contact surface and to reduce 90% of equivalent stress in comparison to those of conventional surfaces.

Evaluation and Selection of HazMat Transportation Alternatives: A PHFLTS- and TOPSIS-Integrated Multi-Perspective Approach.

Outsourcing the hazardous materials (HazMat) transportation is an effective way for manufacturing enterprises to avoid risks and accidents as well as to retain sustainable development in economic growth and social inclusion while not bringing negative impacts on the public and the environment. It is imperative to develop viable and effective approaches to selecting the most appropriate HazMat transportation alternatives. This paper aims at proposing an integrated multi-criteria group decision making approach that combines proportional hesitant fuzzy linguistic term set (PHFLTS) and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to address the problem of HazMat transportation alternative evaluation and selection. PHFLTSs are adopted to represent the congregated individual evaluations in a bid to avoid information loss and increase the reliability of results. Two weight assignment models are then proposed to determine the comprehensive weights of experts and criteria. Furthermore, several novel manipulations of PHFLTS are also defined to enrich its applicability. The TOPSIS method is subsequently extended to the context of PHFLTSs to rank alternatives and choose the best one. Eventually, the feasibility and validity of the proposed approach are verified by a practical case study of a HazMat transportation alternative evaluation and selection decision and further comparison analyses.

A universal deep learning approach for modeling the flow of patients under different severities.

BACKGROUND AND OBJECTIVE: The Accident and Emergency Department (A&ED) is the frontline for providing emergency care in hospitals. Unfortunately, relative A&ED resources have failed to keep up with continuously increasing demand in recent years, which leads to overcrowding in A&ED. Knowing the fluctuation of patient arrival volume in advance is a significant premise to relieve this pressure. Based on this motivation, the objective of this study is to explore an integrated framework with high accuracy for predicting A&ED patient flow under different triage levels, by combining a novel feature selection process with deep neural networks. METHODS: Administrative data is collected from an actual A&ED and categorized into five groups based on different triage levels. A genetic algorithm (GA)-based feature selection algorithm is improved and implemented as a pre-processing step for this time-series prediction problem, in order to explore key features affecting patient flow. In our improved GA, a fitness-based crossover is proposed to maintain the joint information of multiple features during iterative process, instead of traditional point-based crossover. Deep neural networks (DNN) is employed as the prediction model to utilize their universal adaptability and high flexibility. In the model-training process, the learning algorithm is well-configured based on a parallel stochastic gradient descent algorithm. Two effective regularization strategies are integrated in one DNN framework to avoid overfitting. All introduced hyper-parameters are optimized efficiently by grid-search in one pass. RESULTS: As for feature selection, our improved GA-based feature selection algorithm has outperformed a typical GA and four state-of-the-art feature selection algorithms (mRMR, SAFS, VIFR, and CFR). As for the prediction accuracy of proposed integrated framework, compared with other frequently used statistical models (GLM, seasonal-ARIMA, ARIMAX, and ANN) and modern machine models (SVM-RBF, SVM-linear, RF, and R-LASSO), the proposed integrated "DNN-I-GA" framework achieves higher prediction accuracy on both MAPE and RMSE metrics in pairwise comparisons. CONCLUSIONS: The contribution of our study is two-fold. Theoretically, the traditional GA-based feature selection process is improved to have less hyper-parameters and higher efficiency, and the joint information of multiple features is maintained by fitness-based crossover operator. The universal property of DNN is further enhanced by merging different regularization strategies. Practically, features selected by our improved GA can be used to acquire an underlying relationship between patient flows and input features. Predictive values are significant indicators of patients' demand and can be used by A&ED managers to make resource planning and allocation. High accuracy achieved by the present framework in different cases enhances the reliability of downstream decision makings.