Automated Severity Classification for Patient Triage Using a Multilayer Perceptron Model in Emergency Departments

This paper presents a demonstration of the efficacy of neural networks in patient classification within emergency departments. Currently, patient triage in hospital emergency departments relies on the judgment of triage nurses, which is susceptible to human error and often leads to long waiting times. To improve this process, artificial intelligence (AI) has been proposed as a solution. However, the application of neural networks, a fundamental AI method for classification, has not been extensively tested in this context. Our study fills this gap by evaluating the performance of neural networks in automatic patient classification. Obtained results indicate that neural networks can achieve classification accuracy, or even better comparable to other AI methods, offering promising potential for enhancing patient triage efficiency in emergency departments.