16918 research outputs found
Sort by
Using multisource remotely sensed data and cloudcomputing approaches to map non-native species in thesemi-arid savannah rangelands of Mpumalanga, South Africa
No full textSemi-arid savannah rangelands are diverse environments (in terms of species) that play an important role in sustaining biodiversity and providing ecosystem services. However, the emergence of non-native species, as well as bush encroachment, are currently threatening these (semi-arid rangeland and grassland) ecosystems. The purpose of this study was therefore to map and quantify the spatial extents of non-native woody vegetation in the Kruger National Park and surrounding communal areas in Mpumalanga, South Africa. To achieve the study’s objectives, Sentinel-1 and Sentinel-2 remotely sensed data were combined and analysed using the random forest (RF) machine-learning algorithm in the Google Earth Engine (GEE) cloud computing platform. Specifically, spectral bands and selected spectral derivatives, e.g. enhanced vegetation index (EVI2), normalized difference moisture index (NDMI) and normalized difference phenology index (NDPI) were computed and used to map non-native woody vegetation. After optimizing the model combination, the classification outputs had an overall accuracy of 70%, with class accuracies such as producer’s accuracy (PA) and user’s accuracy (UA) ranging from 67% to 95%.It was shown in this study that using Sentinel-2 and Sentinel-1 data together led to better overall accuracy than using single sensor models when mapping semi-arid savannah rangelands. It was also found in this study that the overall classification accuracy of non-native (invasive) species using optical sensors was higher than in previous studies. On a free platform like GEE, it was possible to utilize advanced classification processes to fully exploit the informative content of Sentinel-1 and Sentinel-2 data
Automated detection of race condition vulnerabilities in binary programs using symbolic execution
No full textIdentifying race conditions in binary programs is challenging due to limited research on the subject and the lack of thoroughly evaluated methods, especially those applied to consumer-grade, off-the-shelf binary programs without requiring source code. Symbolic execution is a static analysis technique that significantly enhances vulnerability identi- fication, especially in fuzzing-driven security analysis. However, its scalability is often restricted by the computational demands of constraint solving, high memory consump-tion, and the state explosion problem—the rapid increase in the number of states caused by program structures. To overcome these challenges, many vulnerability analysis meth-ds incorporate techniques to manage state explosion, enabling the analysis of more complex programs. The approach presented by this thesis employs symbolic vulnerability analysis for bi-naries by introducing a novel approach called "Xegmap" which uses Directed Symbolic Execution to strategically guide the exploration of program states. Xegmap is designed to prioritise the detection of Global Memory Access Points, operating on the premise that identifying these points facilitates the detection of potentially race-prone threadinteractions. It accomplishes this through a two-phase process: a naive symbolic execu-tion phase, Negmap, followed by a directed phase, Degmap. Degmap directs symbolic execution toward global memory access points and evaluates memory interactions using a hybrid lock-set and happens-before analysis. Experimentation demonstrates Xegmap’s enhanced capacity to increase code coverage in consumer-grade, off-the-shelf binaries and to detect race conditions in binaries of varying complexity, including those with intricate input constraints and numerous threads, all without the need for source code
Triaxial nuclear shapes from simple ratios of electric-quadrupole matrix elements
No full textTheoretical models often invoke axially-asymmetric nuclear shapes to explain elusive collective phenomena, but such an assumption is not always easy to confirm experimentally. The only model-independent measurement of the nuclear axial asymmetry (or triaxiality) γ is based on rotational invariants of zero-coupled products of the electric-quadrupole (E2) operator — the Kumar-Cline sum rule analysis — which generally requires knowledge of a large number of E2 matrix elements connecting the state of interest. We propose an alternative method to determine γ using only two E2 matrix elements, which are among the easiest to measure. This approach is based on a standard rotational description of a nucleus with stable triaxial deformation, where all underlying assumptions are either empirically proven or unnecessary. It is applied to the 2+ states of the ground-state and the γ bands of even–even nuclei and is model-independent provided these 2+ states have rotational nature. This technique was applied to a number of deformed even–even nuclei for which the ratio of the energies of the yrast 4+ and 2+ states was R4/2> 2.4. Where sufficient experimental data were available for performing Kumar-Cline analysis, good agreement was observed between the γ values deduced in these two approaches. The agreement shows that (i) the 2+ states of the selected nuclei have indeed rotational nature, and (ii) the proposed method represents a simple and reliable deduction of γ. In the present work more than 60 even–even rotating nuclei were associated with axially-asymmetric nuclear shapes
Green-synthesis of MgO and ZrO2 nanocomposites: physicochemical properties and antiplasmodial activity in a mouse model
No full textMalaria remains a significant global health burden, particularly in Sub-Saharan Africa, where drug resistance necessitates novel therapeutic strategies. This study evaluates the antiplasmodial potential of green-synthesized magnesium oxide (MgO) and zirconium oxide (ZrO2) nanoparticles and their composite (Mg/ZrO2) using Eucalyptus camaldulensis leaf extract. MgO, ZrO2, and MgO/ZrO2 nanoparticles were synthesized and characterized using scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray diffraction (XRD), revealing crystalline structures with particle sizes ranging from 39 to 60 nm. Acute toxicity assessment in mice indicated an LD50 > 2000 mg/kg bodyweight, confirming their safety. In vivo antiplasmodial activity was assessed using Plasmodium berghei-infected mice, with treatment groups receiving 50, 100, and 200 mg/kg bodyweight each of the nanoparticles. In the suppressive test, MgO-NPs, ZrO2-NPs, and MgO/ZrO2-NPs exhibited dose-dependent parasite inhibition of 66.79%, 34.72%, and 41.02% respectively at 200 mg/kg bodyweight. The curative test further confirmed parasite clearance, with MgO-NPs demonstrating the highest efficacy. Nanoparticle treatment also improved survival time and maintained body weight compared to untreated controls. The observed antiplasmodial effects is attributed to enhanced cellular uptake, reactive oxygen species (ROS) generation, and disruption of parasite metabolic pathways. These findings highlight the potential of MgO, ZrO₂ and MgO/ZrO2 nanocomposites as promising candidates for antimalarial drug development, warranting further mechanistic studies and preclinical validation
Boycott
No full textA collection of digitised photographs taken at the University of the Western Cape in Bellville, South Africa c1989. The print photographs are housed at the Special Collections section of the Main University Library
Sex differences in adiposity and hemodynamic parameters as cardiovascular risk indicators among South African university staff: a descriptive cross-sectional study
No full textBackground: Cardiovascular diseases (CVDs) are the leading cause of death worldwide, with their prevalence continuing to rise each year. Adiposity indexes and hemodynamic parameters have been established as effective predictors of CVDs when analysed separately. However, the impact of sex differences on the distribution and combined use of these predictors remains largely unexplored, particularly in Sub-Saharan Africa. This study aimed to investigate the sex differences in the distribution of adiposity indexes (AI) and hemodynamic parameters (HP), as well as their associated indicators of cardiovascular diseases risks among staff members at Walter Sisulu University (WSU). Methods: This cross-sectional descriptive quantitative study was conducted on 100 healthy adults (50 males, 50 females) aged 18–65 years. AI were assessed using a stadiometer, body composition monitor, and tape measure, while HP were measured with a stethoscope and sphygmomanometer. Results: The study’s findings revealed that mean values for AI, including height, visceral adiposity index, and waist circumference, were higher in males compared to females, while weight, body mass index, and hip circumference were greater in females. Additionally, the study indicated that mean values for HP, such as systolic blood pressure, diastolic blood pressure, and mean arterial pressure, were elevated in males, whereas pulse pressure was higher in females. Notably, heart rate was consistent across both sexes. Conclusion: This study provides useful information about the sex-based patterns of adiposity indices and hemodynamic distribution among selected South African population
Use of artificial intelligence in healthcare in South Africa: a scoping review
No full textBackground: Artificial intelligence (AI) transformed healthcare worldwide and has the potential to address challenges faced in the South African healthcare sector, such as limited public institutional capacity, staff shortages, and variability in skills levels that exacerbate the demand on the healthcare system that can lead to compromised care and patient safety. Aim: This study aimed to describe how AI, especially machine learning is used in healthcare in South Africa over the last 5 years. Method: The Joanna Briggs Institute (JBI) methodology for scoping reviews was used. Peer-reviewed articles in English, which were published from 2020 to date were sourced and reviewed using the Population, Concept, Context (PCC) framework. Results: A total of 35 articles were selected. The results showed a focus on conventional machine learning, a health focus on HIV and/or tuberculosis (TB) and cancer, and a lack of big data in fields other than cancer. Conclusion: There has been an increase in the use of machine learning in the analysis of health data, but access to big data appears to be a challenge.
Contribution: There is a need to have access to high-quality big data, inclusive policies that promote access to the benefits of using machine learning in healthcare, and AI literacy in the health sector to understand and address ethical implication
Interference of a phytoconstituent from Nymphaea lotus-derived ligand N-acetyl glucosamine with signaling receptors in diabetes mellitus development: a targeted computational analysis
No full textDiabetes mellitus is a world-wide health concern with several millions affected in all ages. Computer-aided drug design (CADD) is a powerful tool that has revolutionized the process of discovering and developing new drugs. It provides innovative methods that can speed up drug discovery and lower costs thereby results to increase enthusiasm at developing instinctive antidiabetic agents as alternatives for managing diabetes. Nymphaea lotus, a plant with medicinal properties known for its anti-diabetic effects, contains bioactive components like N-acetyl glucosamine. An in silico study was conducted to investigate its potential in targeting proteins related to diabetes. Molecular docking studies, toxicity prediction, examination of drug depiction, and Molecular Dynamics Simulation (MDs) of the ligands with the identified receptor target were conducted using the Schrödinger platform. The receptor-ligand complex of Nymphaea lotus was compared with known inhibitors. Molecular dynamics simulation, principal component analysis, and free energy landscape analysis showed that the binding affinity of the Nymphaea lotus complex was higher than that of reference ligands. This suggests that Nymphaea lotus and its bioactive compounds have promising medicinal value for managing type 2 diabetes, warranting further research into their therapeutic potentia
The geology, geochemistry, and geochronology of proterozoic gneisses in the Lüderitz area, southern Namibia: insights into the evolution of the NW Namaqua Metamorphic Province
No full textThe basement rocks of the Lüderitz area, SW Namibia are dominated by migmatitic metavolcanic and intrusive gneisses that resemble those of the Paleoproterozoic Richtersveld Magmatic Arc (~1890 Ma; RMA; of the Namaqua Natal Metamorphic Province (NNMP)). Detailed geological mapping, geochemistry, isotopic and geochronological analysis was undertaken to establish a new, modern lithostratigraphy for the area and to determine whether the Lüderitz gneisses indeed form part of the RMA. The mafic, andesitic and dacitic metavolcanic gneisses (1876 ± 9 Ma) are grouped in the Albatrosskop Formation and occur as rafts within the coeval orthogneisses (1918-1855 Ma). Six main types of orthogneiss were mapped, namely: the Kolmanskop Migmatite, Adventure Bay metagabbro-metadiorite, the Radford Bay, Kowisberg and Albatross Bay granodiorite-granite gneisses and the Elizabeth Point leucogranite gneiss. The volcanic and plutonic rocks are overlain by the Dagger Rocks Group that yielded a youngest detrital grain age of 1731 ± 46 Ma. The supracrustal and plutonic rocks were strongly deformed and metamorphosed during the polyphase Mesoproterozoic Namaqua Orogeny
Leveraging TikTok to enhance understanding and engagement in university chemistry students: A South African case study
No full textThis study explores TikTok’s potential as an educational tool in South African undergraduate chemistry education. TikTok is a social media platform that enables users to create, share, and explore brief videos. TikTok’s user-friendly interface and creative features offer an engaging platform for learning. While learning technologies have traditionally supported knowledge transfer, research highlights their cognitive, behavioural, and affective benefits. Through a teaching intervention and qualitative feedback, students reported improved understanding of complex concepts. The TikTok assignment aligned with both inquiry-based learning (IBL) and decolonial pedagogies, encouraging independent inquiry, collaboration, and cultural relevance by valuing students’ voices and contexts. Though the study reflects a short-term, context-specific case and is not generalisable, it provides useful insights into students’ positive perceptions of using social media in science education. The findings indicate that integrating social media tools into the curriculum within the South African higher education landscape can introduce innovative learning methods. Future research should encompass a wider range of establishments and extend over longer durations to enhance its generalizability and address the current study’s limitations