Machine learning models for predicting residual malaria infections using environmental factors: A case study of the Jazan region, Kingdom of Saudi Arabia
Background: Malaria is a global public health problem affecting more than 100 countries. Meteorological factors on the other hand represent a major driving force behind malaria transmission and other vector-borne diseases. This study aims to predict and forecast malaria incidence while exploring its...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2024-01-01
|
| Series: | Decoding Infection and Transmission |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2949924024000065 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Background: Malaria is a global public health problem affecting more than 100 countries. Meteorological factors on the other hand represent a major driving force behind malaria transmission and other vector-borne diseases. This study aims to predict and forecast malaria incidence while exploring its correlation with environmental factors. Method: Three Machine learning (ML) models, namely Artificial Neural Network (ANN), Random Forest Regression (RFR), and Regularized Linear Regression (RLR), were employed, along with a simple seasonal model, to predict and forecast malaria cases. Results: The ANN model outperformed the RFR and RLR models, with the lowest Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) of 0.313 and 0.146 respectively. A total of 10,778 malaria cases were reported from 2015 to 2020, with a monthly mean of 150 malaria infections. The study unveils no significant increase in malaria cases from 2020 to 2030. Furthermore, a strong negative correlation was found between monthly average malaria incidence and average temperature, minimum and maximum temperatures at p < 0.001. On the other hand, a strong positive correlation was observed between monthly average malaria incidence and relative humidity, which was statistically significant at p < 0.01. Conclusion: The Artificial Neural Network model is effective in predicting malaria cases compared to other models. The study revealed a negative correlation between malaria incidence and temperature, alongside a positive correlation with relative humidity. Although no significant increase in malaria cases is projected from 2020 to 2030, continuous monitoring of environmental factors and malaria prevalence remains crucial for effective disease control. |
|---|---|
| ISSN: | 2949-9240 |