Early prediction of sepsis using double fusion of deep features and handcrafted features.

Sepsis is a life-threatening medical condition that is characterized by the dysregulated immune system response to infections, having both high morbidity and mortality rates. Early prediction of sepsis is critical to the decrease of mortality. This paper presents a novel early warning model called Double Fusion Sepsis Predictor (DFSP) for sepsis onset. DFSP is a double fusion framework that combines the benefits of early and late fusion strategies. First, a hybrid deep learning model that combines both the convolutional and recurrent neural networks to extract deep features is proposed. Second, deep features and handcrafted features, such as clinical scores, are concatenated to build the joint feature representation (early fusion). Third, several tree-based models based on joint feature representation are developed to generate the risk scores of sepsis onset that are combined with an End-to-End neural network for final sepsis detection (late fusion). To evaluate DFSP, a retrospective study was conducted, which included patients admitted to the ICUs of a hospital in Shanghai China. The results demonstrate that the DFSP outperforms state-of-the-art approaches in early sepsis prediction.

Maximal covering location problem of smart recycling infrastructure for recyclable waste in an uncertain environment.

To incentivise citizens to return recyclable waste (RW) through formal recycling channels, the Chinese government has introduced smart recycling facilities for RW. To maximise the satisfaction of citizens' demand for returning RW, the proper arrangement of recyclable facilities is a problem worth studying in cases in which there is a lack of data on the citizens' behaviour. Thus, to help government decision-makers rationally arrange smart recycling facilities, this paper discusses how to properly allocate the recycling infrastructure to cover the maximum demand with a limited number of smart recycling facilities in an uncertain environment. According to the uncertainty theory, the service cost is taken as an uncertain variable. Then, an uncertain maximal covering location problem model and an extended uncertain maximal covering location problem model are constructed, and solution procedures of the two models are provided. The effectiveness of the proposed mathematical formulations and solution procedures were validated through two application cases.

Impact of government subsidies on pricing strategies in reverse supply chains of waste electrical and electronic equipment.

Many governments use subsidies to encourage the recycling of waste electrical and electronic equipment. In order to assist policymakers in maximizing the benefit of these incentives, this paper investigates strategies for the allocation strategy of government subsidies among the parties in the reverse supply chain of e-waste consisting of one collector, one remanufacturer and two retailers. The optimal pricing decisions and effects of government subsidies with multiple subsidized parties are examined. Analytical results suggest that the remanufacturing utilization rate has great influence the allocation strategy of government subsidies. When the e-waste remanufacturing utilization rate is low, the marginal effect of the subsidy to the remanufacturer on economic benefit and the recycle quantity decreases as the subsidy increases. In this situation, the government should subsidize the collector and retailers. And when the e-waste remanufacturing utilization rate is relatively high, the marginal effect of the subsidy to the remanufacturer on economic benefit and the recycle quantity increase as the subsidy increases. In this case, the government should allocate as much support as possible to the remanufacturer.

A symmetric version of the generalized alternating direction method of multipliers for two-block separable convex programming.

This paper introduces a symmetric version of the generalized alternating direction method of multipliers for two-block separable convex programming with linear equality constraints, which inherits the superiorities of the classical alternating direction method of multipliers (ADMM), and which extends the feasible set of the relaxation factor α of the generalized ADMM to the infinite interval [Formula: see text]. Under the conditions that the objective function is convex and the solution set is nonempty, we establish the convergence results of the proposed method, including the global convergence, the worst-case [Formula: see text] convergence rate in both the ergodic and the non-ergodic senses, where k denotes the iteration counter. Numerical experiments to decode a sparse signal arising in compressed sensing are included to illustrate the efficiency of the new method.