A comprehensive analysis of the artificial neural networks model for predicting monkeypox outbreaks

Area of Science:

• Epidemiology
• Computational Biology
• Public Health

Background:

• Monkeypox outbreaks pose a significant global public health challenge.
• Accurate outbreak severity prediction is vital for effective preparedness and response.
• Limited research exists on monkeypox in specific nations like Chile and Mexico.

Purpose of the Study:

• To develop and validate an artificial neural network (ANN) model for predicting monkeypox outbreak severity.
• To analyze time series data of monkeypox cases from multiple countries, including Chile and Mexico.
• To compare ANN model performance against Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models.

Main Methods:

• Utilized a time series dataset of monkeypox cases from Argentina, Brazil, France, Germany, Chile, and Mexico.
• Developed single and two-hidden-layer ANNs using the Levenberg-Marquardt learning technique.
• Employed K-fold cross-validation with early stopping for model training and validation.
• Compared ANN performance with LSTM and GRU models.

Main Results:

• Successfully developed and validated ANN models for monkeypox outbreak prediction.
• Demonstrated the efficacy of ANNs in analyzing time series epidemiological data.
• Established a baseline for future predictive modeling of infectious disease outbreaks.

Conclusions:

• Artificial neural networks show promise for predicting monkeypox outbreak severity.
• The developed models can aid public health officials in preparedness and resource allocation.
• Further research is warranted to refine models and expand their application to other regions.