Prediction and Mathematical Modeling of Diseases due to Climate Change with Artificial Intelligence:
the Sitma Example
Egemen Yılmaz (2008), Erdem Özdemir (2008)
Mamak Bilim ve Sanat Merkezi, Ankara, Turkey
Climate change significantly affects temperature and precipitation, directly impacting daily life, the economy, ecosystems, and public health. One of its critical health consequences is its influence on vector-borne diseases such as malaria, an acute febrile illness caused by the Plasmodium parasite and transmitted by Anopheles mosquitoes. Without treatment, malaria can be fatal, affecting approximately 500 million people annually. The spread of Anopheles mosquitoes is closely tied to variations in temperature and precipitation.
This study analyzed monthly temperature averages from the past 100 years and hourly data from the past 10 years for Ankara. To predict future climate trends, various artificial neural network models were employed, including Baseline, Linear, Prophet, and LSTM (Long-Short Term Memory). ANN modeling, a non-linear statistical technique, was chosen for its effectiveness in addressing complex problems that traditional methods struggle to solve. After testing different models, the most accurate was selected as the foundation for the project’s climate change model.
The predicted climate data was then used in conjunction with malaria spread equations derived from literature research to develop a malaria prediction model for Ankara. The findings aim to improve malaria control strategies and contribute to recommendations for future research.
