RESUMO
In Uganda, the absence of a unified dataset for constructing machine learning models to predict Foot and Mouth Disease outbreaks hinders preparedness. Although machine learning models exhibit excellent predictive performance for Foot and Mouth Disease outbreaks under stationary conditions, they are susceptible to performance degradation in non-stationary environments. Rainfall and temperature are key factors influencing these outbreaks, and their variability due to climate change can significantly impact predictive performance. This study created a unified Foot and Mouth Disease dataset by integrating disparate sources and pre-processing data using mean imputation, duplicate removal, visualization, and merging techniques. To evaluate performance degradation, seven machine learning models were trained and assessed using metrics including accuracy, area under the receiver operating characteristic curve, recall, precision and F1-score. The dataset showed a significant class imbalance with more non-outbreaks than outbreaks, requiring data augmentation methods. Variability in rainfall and temperature impacted predictive performance, causing notable degradation. Random Forest with borderline SMOTE was the top-performing model in a stationary environment, achieving 92% accuracy, 0.97 area under the receiver operating characteristic curve, 0.94 recall, 0.90 precision, and 0.92 F1-score. However, under varying distributions, all models exhibited significant performance degradation, with random forest accuracy dropping to 46%, area under the receiver operating characteristic curve to 0.58, recall to 0.03, precision to 0.24, and F1-score to 0.06. This study underscores the creation of a unified Foot and Mouth Disease dataset for Uganda and reveals significant performance degradation in seven machine learning models under varying distributions. These findings highlight the need for new methods to address the impact of distribution variability on predictive performance.
RESUMO
We explored the viability of using air quality as an alternative to aggregated location data from mobile phones in the two most populated cities in Uganda. We accessed air quality and Google mobility data collected from 15th February 2020 to 10th June 2021 and augmented them with mobility restrictions implemented during the COVID-19 lockdown. We determined whether air quality data depicted similar patterns to mobility data before, during, and after the lockdown and determined associations between air quality and mobility by computing Pearson correlation coefficients ([Formula: see text]), conducting multivariable regression with associated confidence intervals (CIs), and visualized the relationships using scatter plots. Residential mobility increased with the stringency of restrictions while both non-residential mobility and air pollution decreased with the stringency of restrictions. In Kampala, PM2.5 was positively correlated with non-residential mobility and negatively correlated with residential mobility. Only correlations between PM2.5 and movement in work and residential places were statistically significant in Wakiso. After controlling for stringency in restrictions, air quality in Kampala was independently correlated with movement in retail and recreation (- 0.55; 95% CI = - 1.01- - 0.10), parks (0.29; 95% CI = 0.03-0.54), transit stations (0.29; 95% CI = 0.16-0.42), work (- 0.25; 95% CI = - 0.43- - 0.08), and residential places (- 1.02; 95% CI = - 1.4- - 0.64). For Wakiso, only the correlation between air quality and residential mobility was statistically significant (- 0.99; 95% CI = - 1.34- - 0.65). These findings suggest that air quality is linked to mobility and thus could be used by public health programs in monitoring movement patterns and the spread of infectious diseases without compromising on individuals' privacy.