Clinical Severity Level Prediction Based Optimal Medical Resource Allocation at Mass Casualty Incident

Controlling the mortality rate at the Mass Casualty Incident presents an increasingly important challenge for Emergency service organizations. Preparation is required by any healthcare system to minimize the loss of life and maximize casualty recovery. The literature conjectures that casualty rates can be decreased by integrating technology with the emergency service management system. To develop a superior mortality rate-controlling plan it is critical to anticipate the clinical condition deterioration of the casualties and effectively serve a large number of casualties with an optimal number of medical resources. Thus, this research proposes a multi methodology approach that integrates a prediction model to forecast the worsening of the casualties’ clinical conditions with an optimization model to detect the number of medical resources needed to treat the casualties. The method incorporates the Gravitational Search based Back Propagation Neural Network-based prediction model together with the qSOFA medical score to produce an accurate casualty’ clinical condition prediction. The findings of the prediction model are then combined with methods to identify in advance the optimum number of medical resources needed to control the rate of incoming casualties. The proposed multi-methodological approach experimented on the MIMIC-II dataset results show that clinical condition prediction and allocation of the optimal number of medical resources supports reduction in mortality length. The prediction model has the Accuracy, Sensitivity, and Specificity values 91.9%, 94.7%, and 83.9% respectively. Further, the optimization model is compared with the literature work and the result shows the better performance with regard to mortality length and Queue length performance parameters. The proposed work involving the prediction model of clinical conditions and the optimization model of medical resources facilitates clinicians to serve the casualties with better health treatment, thus reducing the mortality length.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.