Malaria remains a major public health problem, particularly in sub-Saharan Africa, where microscopic diagnosis using blood smears remains the gold standard. However, this approach relies heavily on the expertise of laboratory technicians and can lead to variability in interpretation. In this study, we propose a hybrid artificial intelligence approach aimed at automatically calculating an image score based on images of blood smears. The methodology developed combines automatic segmentation of red blood cells, a convolutional neural network (CNN) for detecting infected cells, a multiscale texture analysis based on wavelets, and an anomaly detection module inspired by Denoising Diffusion Probabilistic Models (DDPM). The experiments were conducted using the public NIH Malaria Cell Images Dataset and BBBC041 datasets. The CNN classifier achieved an accuracy of 98.47%, an F1-score of 98.47%, an AUROC of 0.9988, and an AUPRC of 0.9989. Analysis of 1,328 blood smears yielded an average image score of 0.473, associated with an average parasitemia of 26.25%. The ablation study confirmed the complementary contribution of the wavelet and DDPM components, while machine learning of the weights showed that parasitic density is the most informative variable, with an importance of 72.09%. These results demonstrate the potential of the proposed approach for the development of intelligent systems to aid in the diagnosis and quantitative monitoring of malaria.

Accurate short-term forecasting of meteorological variables remains a major challenge in tropical regions characterized by strong diurnal variability, seasonal transitions, intermittent atmospheric fluctuations, and highly non-stationary temporal dynamics. This study proposes a hybrid deep-learning framework combining the Stationary Wavelet Transform (SWT), Multi-Head Attention (MHA) and Long Short-Term Memory (LSTM) networks’ forecasting layers for multivariate meteorological prediction over selected locations in Togo. The proposed architecture simultaneously forecasts temperature at 2 m, relative humidity, and wind speed using hourly ERA5 atmospheric variables enriched with lagged, rolling-statistical, and cyclic temporal features. SWT decomposition extracts multi-scale atmospheric structures while preserving temporal alignment, the attention mechanism dynamically emphasizes informative temporal sub-sequences, and the LSTM layers improve medium-range temporal dependency modeling. The forecasting task was reformulated under a strictly causal prediction protocol for 24 h and 72 h forecasting horizons. The proposed framework achieved competitive forecasting performance with MAE = 2.99, RMSE = 5.22, and ${\mathrm{R}}^2=0.77$ for the 24 h horizon, and MAE = 3.44, RMSE = 6.11, and ${\mathrm{R}}^2=0.73$ for the 72 h horizon. Temperature forecasting exhibited the highest stability, whereas wind-speed prediction remained more challenging because of stronger atmospheric intermittency. The results demonstrate the relevance of multi-scale temporal representation learning for ERA5-based tropical meteorological forecasting while highlighting important methodological and operational limitations.

Forecasting monthly extreme flows is a major challenge in hydrology due to their rarity and high intensity, particularly in tropical basins vulnerable to climate change. This study proposes an innovative hybrid approach combining STL decomposition, generalized extreme value (GEV) theory, and LSTM and GRU architectures to predict river flow: the case of the Mono River in Togo. The methodology is based on isolating the residual component, modeled by a GEV distribution, whose values are converted into probabilities using a cumulative distribution function. A unique feature of this approach is the incorporation of multivariate meteorological data. Unlike conventional approaches, the results show that the hybrid model particularly in its univariate sequential configuration reproduces extreme dynamics with a high degree of accuracy. The evaluation was conducted at various stations in Togo using the "Peak Over Threshold" approach, applied at the 75th percentile. At the Dotaicopé station, the model performed robustly, achieving an accuracy of 0.82, a recall of 0.74, an F1 score of 0.78, and a Kling-Gupta efficiency coefficient of 0.75. At the Tététou station, the multivariate model achieved an exceptional recall of 0.9, confirming its superior ability to detect critical thresholds in areas with high hydrological variability; the univariate model, on the other hand, performed less well in this regard, thereby demonstrating the significant contribution of climatic parameters. However, the study highlights a limitation related to data asymmetry, as climate forcings are only available starting in 1981, whereas discharge records date back to 1952. These results validate the potential of both univariate and multivariate probabilistic hybrid models for better characterization of hydrological regimes and early flood risk prevention.

TAccurate short-term electricity consumption forecasting is essential for operational planning, reserve allocation, and energy management in modern power systems. This study investigates the performance of recurrent and hybrid attention-based deep learning architectures for short-term electricity consumption forecasting, a sase study in Togo. The proposed framework integrates electricity consumption, meteorological, demographic, and temporal information in order to capture both intrinsic temporal dependencies and exogenous influences affecting electricity demand. Four forecasting architectures were evaluated using a weekly temporal window of 168 hours: LSTM-only, LSTM-decoder, LSTM-attention, and a hybrid LSTM--Multi-Head Attention--LSTM model. The experiments included multi-seed evaluation, ablation study, robustness analysis, and statistical comparison using the Wilcoxon signed-rank test. The results show that all models achieved extremely high forecasting accuracy, with coefficients of determination exceeding 0.9998 and MAPE values below 0.004%. The hybrid architectures slightly improved average forecasting performance, while the standalone LSTM model remained highly competitive. The robustness analysis revealed strong sensitivity to noisy inputs but moderate degradation under missing-data conditions. Statistical analysis indicated that the performance differences between architectures were not statistically significant at the 5% level. Overall, the study demonstrates that recurrent deep learning architectures provide reliable and operationally relevant solutions for electricity consumption forecasting in highly structured energy demand series.