Kyambogo University

Onsite sanitation coverage has expanded significantly since 2020, driven by global commitments to eliminate open defecation and the need for adaptable solutions in rapidly urbanising small- and medium-sized cities (<1 million inhabitants) in Low- and Middle-Income Countries. Currently, 53% of individuals with access to a toilet depend on onsite sanitation systems. Despite this prevalence, the contribution of greenhouse gas emissions from household-level excreta storage to climate change remains poorly quantified due to limited empirical evidence. We addressed this gap by conducting direct measurements of methane emissions from 146 onsite sanitation containment units locally referred as pit latrines, holding tanks and septic tanks, across Senegal, Ethiopia, Uganda and Nepal. Methane emission rates exhibited strong skewness with a geometric mean of 7.9 g CH₄ capita⁻¹ day⁻¹, indicating that onsite sanitation containment units alone may account for approximately 3.8% of global anthropogenic CH₄ emissions.

Uganda’s population growth, currently at 3.2%, has increased the demand for residential, commercial, and industrial buildings. Concrete constitutes nearly 60% of construction materials on building sites, and aggregates contribute 70–85% of its total weight. This places pressure on existing natural aggregate sources, creating the need for sustainable alternatives. Lightweight aggregates offer a potential solution, yet kukui seed shells (an organic and waste-derived option) have never been studied in the Ugandan context, despite their possible contribution to sustainable construction. This study adopted a combined experimental and numerical methodological framework to evaluate the suitability of kukui seed shells as lightweight aggregates. Mechanical characterisation of the shells was followed by an experimental program in which normal aggregates were partially replaced with kukui seed shells at 25%, 50%, 75%, and 100% for both Half Seed Shells and Quarter Seed Shells. The resulting concrete mixes were assessed through mechanical and water absorption, Fire resistance, and Thermal conductivity performance tests, while serviceability behaviour was analysed using ANSYS finite element simulations. The shells demonstrated satisfactory Aggregate Impact Value (AIV), Aggregate Crushing Value (ACV), Los Angeles Abrasion (LAA), and soundness performance. Increasing the replacement percentage led to reductions in density, compressive strength, flexural strength, splitting tensile strength, and thermal conductivity. Quarter Seed Shell mixes generally outperformed Half Seed Shell mixes. Based on overall performance, a 50% replacement using Quarter Seed Shells is recommended, yielding a compressive strength of 25.1 MPa, thermal conductivity of 1.6 W/mK, and reduced density of 2,112 kg/m³, making it a viable lightweight concrete option for sustainable construction in Uganda.

Agricultural Virtual Platforms (AVPs) enable diverse actors to exchange information that addresses multidimensional farming challenges. This study examined the diversity of actors engaged in M-Omulimisa AVP and how their roles collectively facilitated information exchange. A qualitative case study involving four key informant interviews, document reviews, 21 in-depth interviews, and three focus group discussions was conducted. Data analysis employed the Agricultural Innovation Systems (AIS) framework. Findings revealed variations in the diversity of actors per domain, their attributes, and their functional roles. Actors belonged to all five domains of AIS: education and research, intermediary, enterprise, demand, and support structures domains. However, diversity within each domain remained limited, with several key actors, especially from the livestock sector, absent. Actors’ attributes indicated that the majority enrolled in the AVP by invitation, had interests focused on crop enterprises, and varied capabilities. Joining the AVP by invitation partly explains the limited actor diversity within domains and the emphasis on crop enterprises. Three role categories that jointly shaped the functionality of the AVP emerged: those assigned and played, assigned but not played, and not assigned yet played. In conclusion, the platform achieved diversity across the AIS domains but not within each domain. Actors’ enrolment through invitation reduced actor diversity, which subsequently limited the information sources. Further, the unperformed roles led to the evolution of roles on the AVP and undermined information exchange. This study recommends enrolment to focus on intra-domain representation of actors, especially for the livestock sector, while promoting other enrolment methods to ensure diversity.

Abstract The development of the Wireless Power Transfer (WPT), also known as wireless charging, has revolutionized the charging process of Electric Vehicles (EVs), providing for greater safety, automation, and convenience compared to the traditional conductive charging process. However, the issue of misaligned coils between the transmitter and the receiver remains a major concern for the effective power transfer rate in the wireless charging process and the subsequent successful commercialization of wireless charging technology. In this regard, an effective approach to the above problem has been proposed by this paper. A theoretical study for the distribution of the magnetic field for different square coil geometries was carried out to analyze the effect of the turns, spacing, and phase of the current for flux uniformity. Using nonlinear programming, a non-uniform distance coil was designed to reduce the variance of the magnetic flux intensity. The results showed the optimized coil to be superior compared to the conventional uniform coils. In addition, the DNN model was trained for the prediction of the relationship between the phase differences of the coil currents and the magnetic flux within the designated regions of the coil array using 500 data sets. The trained network obtained an R 2 value of 0.9997, thus validating the high precision of the trained network. The results obtained validated the fact that the magnetic flux density within the planar coil could be greatly enhanced by the optimal phase arrangement. The proposed method effectively counteracts one of the biggest hindrances in WPT, coil misalignment, through the integration of advanced EM design and data-driven predictive modeling techniques. The results of the study provide a new insight for the development of intelligent and dynamic WPT techniques to retain high transfer efficiencies irrespective of the misalignment levels. Integration of non-uniform coil design and optimized techniques from the realm of the neural network provides a major breakthrough in the development of efficient wireless charging techniques for the envisioned EV framework of the coming years.

<title>Abstract</title> Gestational Diabetes Mellitus (GDM) is associated with increased risks of maternal and neonatal complications, and early identification of high-risk pregnancies is essential for improving clinical outcomes. This study develops an explainable stacking ensemble model for GDM prediction using routinely collected clinical and demographic variables. The ensemble integrates Logistic Regression, Random Forest, XGBoost, Support Vector Machine, and K-Nearest Neighbors as base learners, combined through a logistic meta-learner whose decision threshold is calibrated using a Maximum Expected Utility (MEU) framework to reflect the higher clinical cost of false-negative outcomes. The dataset consisted of 3,525 antenatal records and was processed using stratified splitting, fold-level K-Nearest Neighbors imputation, and class balancing applied only to the training subset to minimise the risk of information leakage. Model performance was evaluated on a hold-out test set using accuracy, precision, recall, F1-score, area under the ROC curve (AUC), and MEU-based utility metrics. The ensemble achieved an accuracy of 98.96%, precision of 99.30%, recall of 98.61%, F1-score of 98.95%, and an AUC of 0.9996, while the MEU-optimised threshold increased decision utility relative to default probability thresholds. SHAP-based explainability analysis showed that OGTT, BMI, PCOS, HDL cholesterol, and diastolic blood pressure were the most influential predictors. The results indicate that integrating ensemble learning, cost-sensitive threshold optimisation, and explainable AI can enhance predictive performance and interpretability for GDM risk stratification. However, as the findings are based on a single public dataset without external validation, the results should be interpreted cautiously. Future work will focus on temporal and cross-cohort validation, calibration analysis, and prospective evaluation in real clinical workflows to strengthen generalisability and translational value.

The study examined the moderating effect of reflective action team learning practices on the association between team psychological safety and knowledge acquisition and creation in teaching hospitals. The study population comprised trainees and health professional staff teams. Using a cross-sectional survey design and sampling frame of teaching hospitals, a sample of 132 respondents was selected. Statistical Package for Social Scientists (SPSS) with a PROCESS macro procedure by Hayes was used for data analysis. Results indicated that reflective action team learning moderates the relationship between team psychological safety and knowledge acquisition. There was a statistically significant regression of team psychological safety and knowledge creation. The interaction effect between team psychological safety and reflective action team learning was significant. Thus, nurturing reflection-in-action team learning practice has high ramifications for knowledge acquisition and creation, which are prerequisites for clinical evidence-based practices

<title>Abstract</title> Accurate and efficient object detection is essential for intelligent road-scene monitoring systems operating in visually complex and resource-constrained environments. While one-stage detectors achieve high inference speed, they often struggle with precise localization of small or low-contrast objects, whereas two-stage detectors provide higher accuracy at the cost of increased latency. To address this trade-off, this paper proposes a hybrid object-detection architecture that integrates You Only Look Once version 5-Small (YOLOv5s) as a fast proposal generator with Faster Region-Based Convolutional Neural Network (Faster R-CNN) as a region-wise refinement module. The proposed framework replaces the Region Proposal Network of Faster R-CNN with high-confidence YOLOv5s detections and employs confidence-weighted fusion to produce spatially consistent final predictions. The hybrid model was evaluated on complex road-scene data using standard object-detection metrics, including mean Average Precision at IoU 0.50 (mAP@50), precision, recall, and inference speed. Experimental results show that the proposed approach achieves mAP@50 of 0.89, improving upon the YOLOv5s baseline by 4.7 percentage points, while maintaining near–real-time performance at 45 frames per second, which is approximately three times faster than a standalone Faster R-CNN. The hybrid detector also attained a precision of 0.93 and a recall of 0.90, demonstrating improved localization accuracy and reduced false detections, particularly for small and visually ambiguous road-scene objects. Repeated experiments confirmed the robustness of the approach, with consistent accuracy gains and low variance across runs. These results demonstrate that strategically combining one-stage and two-stage detection paradigms can yield a favorable accuracy–efficiency balance, making the proposed hybrid architecture suitable for practical deployment in intelligent road-infrastructure and smart-city applications.

Phytochemicals or secondary metabolites are key constituents of plants that play a major role in treating and alleviating human ailments. They possess numerous therapeutic and physiological properties, including disease prevention and treatment, symptom management, and the promotion of physical and mental well-being. Although 80% of the global population relies on plants for medicinal purposes, and analysing the phytochemical components of traditionally used plants is crucial for understanding their potential to treat specific diseases, many of these plants remain unresearched. This study assessed the phytochemical composition of Ficus saussureana DC., Clerodendrum rotundifolium Oliv., and Microglossa pyrifolia (Lam.) O. Ktze qualitatively and quantitatively via standard methods. The gas chromatography-mass spectrometry (GC-MS) method was used to detect general compounds in the methanol extracts. The results indicated the presence of most of the analysed compounds, such as alkaloids, steroids, phenols, tannins, flavonoids, coumarins, and terpenoids. The highest amounts of polyphenols (217±25.05 mg/g) and tannins (179.75±3.44 mg/g) were detected in the Ficus saussureana methanolic extract, whereas the highest amounts of flavonoids (28.75±0.98 mg/g), saponins (225.07±4.11 mg/g), and alkaloids (116.15±3.73 mg/g) were detected in the Clerodendrum rotundifolium methanolic extract. GC-MS profiles revealed 14 compounds in Ficus saussureana and 30 compounds in Clerodendrum rotundifolium and Microglossa pyrifolia. Several of the identified compounds demonstrate pharmacological activities relevant to the treatment of diabetes, hypertension, and various other human ailments. This research validates the assertions of traditional herbalists concerning the three selected plants, uncovering compounds with potential antidiabetic and antihypertensive properties. These findings not only support their traditional use but also highlight their potential for future drug discovery

This study examined the relationship between Teacher professional development (TPD) and the quality of teaching methods used in science education. The major forms of professional development considered the major contributors to enhancing teaching methods included formal, collaborative and self-directed/ online TPD. A cross-sectional research design was used to collect data from 150 Science Teachers and 10 Head teachers from 10 secondary schools of Fort Portal City. Quantitative data was analysed using the Spearman rank correlation coefficient and multiple regression analysis, while qualitative data was grouped into themes and analysed descriptively according to the study objectives. The study revealed significantly significant and positive correlation across all three forms of teacher professional development with teaching methods improvement. However, results from multiple regression show that Formal TPD Participation is the strongest contributor to teaching methods improvement. Formalise and institutionalise teacher professional development as a core function within Uganda’s education system