Neural Network-based Time Series Forecasting of HIV Epidemics: The Impact of Antiretroviral Therapies in the Philippines

Abstract

In the past ten years, the HIV epidemic in the Philippines has grown and changed rapidly. The rate of detected HIV infections has sharply grown to 32 cases per day, going from a low and gradual to a fast and furious epidemic. Thus, some modeling and forecasting methods are necessary for the country to predict the spread pattern and enforce mitigation measures. In this study, the researcher uses Artificial Neural Network in time series forecasting to determine the impact of Antiretroviral Therapies (ART) in the Philippines. The datasets extracted from the HIV/AIDS and ART Registry of the Philippines for the period March 2009 – February 2022 (156 months) are carefully examined for forecasting and analysis. Findings revealed that the cumulative cases in the country by December 2030 will reach 256,983, showing an upward linear trend, with the highest peak in March 2025 of 4,225 cases. The observed and predicted values of HIV epidemics are somewhat close and similar, as supported by the lower values of its RMSE, MAE, and MAPE and higher coefficient of determination. Further, findings showed that as per the United Nations’ SDG-3 of Project 2030, the Philippines is still far from the goals for ending the HIV epidemic due to an increase in HIV incidence in the country. Thus, the Philippine government must continue to adopt the 90-90-90 UN targets and improve further its ART program.