Computational Forecasting Methodology for Acute Respiratory Infectious Disease Dynamics.

Area of Science:

• Epidemiology
• Computational Biology
• Public Health

Background:

• Early identification of infectious disease outbreaks is crucial for effective public health interventions.
• Acute Respiratory Infections (ARI) pose a significant public health challenge globally.

Purpose of the Study:

• To develop and validate a novel methodology for predicting infectious disease outbreaks.
• To improve the accuracy and timeliness of infectious disease forecasting using integrated computational models and real-world data.

Main Methods:

• A hybrid computational model integrating machine learning, projection modeling, and a smoothed endemic channel calculation was developed.
• Weekly acute respiratory infection (ARI) data from Mexico and Google search engine trends were utilized for predictions.
• The proposed methodology was benchmarked against state-of-the-art techniques.

Main Results:

• The methodology demonstrated reduced Root Mean Squared Percentage Error (RMPSE) and Maximum Absolute Percent Error (MAPE) compared to existing methods.
• A Maximum Absolute Percent Error (MAPE) of 21.7% was achieved, indicating improved prediction accuracy.
• The model successfully integrated epidemiological data with digital surveillance (Google Trends).

Conclusions:

• The proposed methodology offers a promising approach for early detection and alerting of acute respiratory infection (ARI) outbreaks.
• This integrated approach can be extended to monitor and predict other seasonal infectious diseases.
• Enhanced infectious disease surveillance through computational modeling can significantly aid public health preparedness.