Two-stage stochastic programming approach for limited medical reserves allocation under uncertainties

Patient allocation method in major epidemics under the situation of hierarchical diagnosis and treatment

OBJECTIVES

Patients are classified according to the severity of their condition and graded according to the diagnosis and treatment capacity of medical institutions. This study aims to correctly assign patients to medical institutions for treatment and develop patient allocation and medical resource expansion schemes among hospitals in the medical network.

METHODS

Illness severity, hospital level, allocation matching benefit, distance traveled, and emergency medical resource fairness were considered. A multi-objective planning method was used to construct a patient allocation model during major epidemics. A simulation study was carried out in two scenarios to test the proposed method.

RESULTS

(1) The single-objective model obtains an unbalanced solution in contrast to the multi-objective model. The proposed model considers multi-objective problems and balances the degree of patient allocation matching, distance traveled, and fairness. (2) The non-hierarchical model has crowded resources, and the hierarchical model assigns patients to matched medical institutions. (3) In the "demand exceeds supply" situation, the patient allocation model identified additional resources needed by each hospital.

CONCLUSION

Results verify the maneuverability and effectiveness of the proposed model. It can generate schemes for specific patient allocation and medical resource amplification and can serve as a quantitative decision-making tool in the context of major epidemics.

Hybrid attention network with appraiser-guided loss for counterfeit luxury handbag detection

Recently, convolutional neural networks have shown good performance in many counterfeit detection tasks. However, accurate counterfeit detection is still challenging due to the following three issues: (1) fine-grained classification, (2) class imbalance, and (3) high imitation samples. To address these issues, we propose a hybrid attention network (HANet) for counterfeit luxury handbag detection. In HANet, a hybrid attention module is first designed. Compared with existing methods that directly use classic CNNs for counterfeit detection, the HA module jointly uses a channel attention unit and a spatial attention unit to learn important information on both the channel and spatial dimensions. The HA modules can be easily integrated into the ResNet architecture to enhance the discriminative representation ability of CNNs, so as to help the network find subtle differences between the real and counterfeit products. In addition, an appraiser-guided loss is proposed to train HANet. Considering the factor of class imbalance and high imitation samples, the proposed loss gives the counterfeit class a higher weighting, and meanwhile gives the high imitation samples a much higher weighting. The proposed loss introduces the knowledge of appraisers, which allows HANet to not only treat real and counterfeit samples relatively fairly, but also pay more attention to the learning of difficult samples. To evaluate the performance of our method, we have constructed a well-benchmarked luxury handbag dataset. On this dataset, the performance of HANet, ResNet50, and the state-of-the-art attention methods is compared. The results demonstrate that HANet achieve superior performance against all its competitors.