Classification of COVID-19 Mortality Risk: A Neural Network-Based Approach Using Mexican Healthcare Sector Data

This study focuses on the use of neural networks and clinical data collected by the Mexican Ministry of Health to classify the risk of death from COVID-19. A multi-layer perceptron neural network model was designed, achieving remarkable results with an accuracy of 96.28%, sensitivity of 99.23%, and an F1 score of 0.9773. The model was optimized through meticulous exploration of various network configurations and performance enhancement techniques. The results showcase the efficacy of neural networks in predicting the risk of death, allowing healthcare professionals to prioritize treatment and allocate resources more efficiently. The value of artificial intelligence in the fight against the pandemic is emphasized, along with its potential application in diverse geographical and healthcare contexts. This work contributes to the advancement of predictive models and encourages further research in the fields of epidemiology and artificial intelligence to combat COVID-19.