4667 research outputs found
Sort by
Early detection of breast cancer using machine learning
Full text linkThesis (Ph. D. (Computer Science)) -- University of Limpopo, 2025Recent developments in artificial intelligence have allowed significant gains in computer vision, cybersecurity, e-Commerce, and healthcare through the application of machine learning and deep learning models. A plethora of applications have offered effective ways to support physicians and radiologists in their medical imaging analysis, which continues to be the fundamental component of the visual representation used to formulate the final observation and diagnostic.
Expert knowledge in data science and computer engineering has recently been combined with medical studies in oncology. In this regard, automatic assistance, sometimes referred to as computer-aided diagnosis (CAD) systems, has gained popularity as a field for study and development in recent decades. As a result, experience and multidisciplinary knowledge were used in the creation of CAD systems, which are now used to analyse patient data and assist physicians and physicians in making decisions. Every day, radiologists and oncologists tackle the vital task of diagnosing, evaluating, and treating cancer, in addition to preventing it. To provide decision support for many applications in cancer patient care operations, such as lesion detection, cancer staging, characterisation, recurrence, tumour evaluation, and prognosis forecasting, a computer-aided design (CAD) system may be developed.
Throughout the world, breast cancer is believed to be one of the most prevalent cancers that affect women. It was also thought to be the main cause of death for women, and the number of deaths from it has been rising yearly. Early detection and diagnosis of breast cancer through regular screenings is the most effective way to significantly improve the chances of successful treatment by identifying breast cancer at its early stages. Cancer arises from uncontrolled cell division that allows these cells to invade surrounding tissues. This uncontrolled growth is often triggered by mutations within genes that are the blueprints for cellular function. Breast cancer is a highly heterogeneous disease that exhibits variations in tumour biology, aggressiveness, and response to treatment. In South Africa, breast cancer is the most common cancer among women, posing a significant health threat. Early detection is crucial, as it greatly improves treatment outcomes. While experienced physicians play a vital role in diagnosis, machine learning systems are emerging as promising tools with high accuracy in cancer identification. The primary goal of this research was to create a model that can assist radiologists in identifying and categorising breast cancer. The Mankweng Hospital repository provided the mammogram images that were used to create the deep learning algorithm. In transfer learning, a pre-trained model, VGG19, InceptionV3 and MobilenetV2, was utilised for fine-tuning the model. A convolutional neural network (CNN) model was developed and optimised using techniques to determine the best batch size, learning rate, epoch, and optimise parameters. During training, the InceptionV3 model achieved the best accuracy of 88%. The models generated are capable of classifying breast cancer cases. However, for some classes, there was not enough data available. This study applied augmentation to address the over-fitting of data. Thus, the next steps of this dissertation involve collecting a lot of data for every class and creating a more reliable categorisation model.
This dissertation therefore suggests a new model for breast cancer diagnostics that is based on the latest advances in computer vision and machine learning technologies. The best method to identify breast lesions early on and to reduce the risk of death is mammography screening. It helps to expose breast anomalies such as microcalcification, architectural distortion, and mass lesions. Having an additional reading tool or help system could improve the breast cancer diagnostic procedure, as the number of patients examined every day is always increasing. Several modalities, including an X-ray scanner and a full-field digital mammography (FFDM) system, can be used to obtain mammograms. The ultimate diagnosis may depend on the quality of the mammograms, the attributes (such as density and size) or the attributes (such as location, size, and form). Consequently, radiologists run the risk of overlooking the lesions, which could lead to incorrect diagnosis and detection. Therefore, the objective of this effort was to improve mammography reading to increase the accuracy of difficult assignments. Incoming research may incorporate novel approaches to techniques by merging several mammography data sets and enhancing the extended training of deep learning models. Motives could also improve the model by including more breast cancer lesions, such as calcification and architectural distortion, using annotated information. In this dissertation, a model was presented to help medical professionals and specialists determine the probability of the presence of breast cancer. All things considered, the suggested model approach combines the latest developments in deep learning, image processing, and picture-to-image translation for biomedical use
How fish climb trees : illuminating the experiences of academic mothers in South Africa
Full text linkJournal article published in African Perspectives of Research in Teaching and Learning Journal Issue 3, Volume 9, 2025 Special IssueThe title of this contribution emerged from the truism that academic mothers seem to do the impossible. Mothers spend a significant amount of time and energy on parenting, often exceeding the hours of a normal job. When adding the responsibilities of an academic portfolio, it is not surprising that the wellbeing and mental health of academic mothers are constantly hanging in the balance. The COVID-19 pandemic redirected our attention to the importance of general wellbeing, but specifically for women who need to manage work, household, and childcare responsibilities. Using semi-structured interviews, this interpretive empirical exposition shines the spotlight on the experiences of four women in higher education in South Africa who have had to manage childcare in addition to their academic work. Using Crenshaw’s notion of intersectionality, the paper highlights how women are “taxed” for choosing to have children and pursue academia. Through centering their voices, this contribution advocates for taking sensitive discussions out of the bathroom and into the boardroom. It discusses the importance of creating enabling environments for these academic mothers through education, sensitive policy-making, realistic goal-setting, empathetic HR practices, and changing how we think about the value of what mothers do
Challenges faced by secondary school teachers in implementing inclusive education in Rakgwadi Circuit, Sekhukhune South
Full text linkJournal article published in African Perspectives of Research in Teaching and Learning Journal Issue 2, Volume 9, 2025Implementing inclusive education practices and principles remains a significant challenge in many South African secondary schools, particularly within the mainstream system. These challenges must be identified and systematically addressed through a comprehensive, all-encompassing, inclusive education approach. A qualitative case study was employed to explore the challenges faced by teachers in implementing inclusive education in the Rakgwadi Circuit, Sekhukhune South, located in the Limpopo Province. The study was framed within the constructivism paradigm. Purposive sampling was utilised to select participants, comprising four principals and twenty teachers teaching mathematics, mathematical literacy, and languages. Data was collected through individual interviews, focus group discussions, and open-ended questionnaires. Thematic analysis was conducted to identify recurring themes and patterns related to implementing inclusive education. The study revealed several significant barriers to inclusive education, including inadequate teacher training, overcrowded classrooms, lack of support from the Department of Basic Education, insufficient school infrastructure, limited parental involvement, and language barriers. These factors collectively hindered the effective teaching and learning process. The findings highlight the critical need for a comprehensive approach to inclusive education that systematically addresses the identified challenges. The study suggests enhancing advocacy and awareness while improving continuous professional development for all teachers to effectively implement inclusive practices
The role of e-learning in open distance learning : a case study of a selected institution
Full text linkJournal article published in African Perspectives of Research in Teaching and Learning Journal Issue 1, Volume 9, 2025A qualitative approach was used to study the role of eLearning in Open Distance Learning at Unisa. This is an exploratory study. The importance of Open Distance and eLearning has motivated the researchers to embark on the study. For this limited scope dissertation, the researcher physically interviewed ten students selected using purposive sampling. The participants were limited to Unisa students enrolled in the College of Education. Participants were selected based on their experience in ODeL mode of learning. The data
was analysed, presented, coded, organised, categorised and themes were identified. The findings suggest that participants understood student support in the context of ODeL. The responses revealed that ODeL reduces costs, is a flexible, enriched learning experience, increases success rate and encourages life-long learning for all age groups and cultures. The ODeL module should continuously be improved to remain credible. It is important to note that this paper could not address all issues of student support in ODeL. Although the participants were a small sample, it is believed that the findings of this study make a valuable contribution to eLearning in Open Distance Learning. Student monitoring, and effective mechanisms to curb internet access challenges for online students can ensure success. Furthermore, the findings revealed that students need to be taught computer literacy before registering for this specific online modul
Undergraduate students’ experiences of online learning in accounting education during COVID-19: a South African case study
Full text linkJournal article published in African Perspectives of Research in Teaching and Learning Journal Issue 4, Volume 9, 2025This article reports on students’ experiences of learning in an undergraduate Accounting Education module at one university in KwaZulu-Natal, South Africa. Qualitative research was applied as a research methodology to explore students’ experiences. Ten students were purposively selected from an Accounting Education class that had participated in online learning during the COVID-19 era. The choice of a phenomenological case study as a design type enabled the researcher to focus on a few participants who had experiences of learning Accounting Education online, and who could relate these experiences. Face-to-face semi-structured interviews and focus-group discussion were used to generate data, and thematic analysis was used to analyse this data. The Digital Equity Theory has been used to conceptualise the study through its dimensions of digital equity. Findings show that undergraduate Accounting Education students experienced unreliable internet and network connections, and a flexible learning environment and enhanced opportunities for learning. Understanding students’ experiences of learning online in an undergraduate Accounting Education module has implications for online learning as it may provide valuable lessons from a shift to online learning which Accounting Education teachers in emerging economies can learn from, and advocate for institutional reforms needed to ensure resilience in future disruptions
Analysis of BRICS' response to the socio-economic impact of the GOVID-19 pandemic in South Africa
Full text linkThesis (M. A. (International Politics)) -- University of Limpopo, 2025.This study analyses the response of BRICS (Brazil, Russia, India, China, and South
Africa) to the socio-economic impact of COVID-19 in South Africa. Using a qualitative
approach, this study evaluates the effectiveness of BRICS’ cooperation in addressing
the economic, social and health consequences of pandemic. The findings revealed a
mixed bag of solidarity, limited tangible support, and constrained cooperation due to
geopolitical, economic, and institutional factors. Despite some remarkable
initiatives, BRICS’ response has been hindered by a lack of coordination and limited
engagement with local stakeholders. The study highlights opportunities for enhanced
cooperation in health diplomacy, trade, and economic development, emphasising the
need for strengthened coordination and prioritisation of local needs. This study
contributes to understanding the role of emerging economies in global health
governance and crisis respons
Statistics of extremes with application to extreme floods in Kwazulu Natal Province, South Africa
Full text linkThesis (M. Sc. (Statistics)) -- University of Limpopo, 2025Extreme rainfall has become a prevailing natural disaster in the region of
Southern Africa. Flooding is one of the natural disasters that pose damage
to property, infrastructure, and human lives. This study conducted a comprehensive
extreme value analysis of monthly maximum rainfall recorded at
five selected meteorological stations in KwaZulu-Natal province, South Africa;
namely Mandini, Mount Edgecombe, Richards Bay Airport, Port Edward, and
Virginia, using data spanning from 1952 to 2022 as provided by the South
African Weather Service (SAWS). The aimed to compare the performance of
advanced extreme value theory (EVT) models, specifically the generalised extreme
value distribution (GEVD), generalised extreme value distribution for
r-largest order statistics (GEVDr) and the blended generalised extreme value
distribution (bGEVD), in modelling extreme rainfall events. Stationarity assessments
using the Augmented Dickey-Fuller (ADF), Kwiatkowski-Phillips-
Schmidt-Shin (KPSS), and Phillips-Perron (PP) tests produced mixed results,
while the Mann-Kendall (M-K) trend test indicated a monotonic decreasing
trend in rainfall. Parameter estimation for the GEVD was performed using
maximum likelihood estimation (MLE) and Bayesian Markov Chain Monte
Carlo (MCMC) methods, both yielding positive shape parameters consistent
with the Fr´echet class of distributions. Goodness-of-fit evaluations through
Anderson-Darling (A-D) and Kolmogorov-Smirnov (K-S) tests, alongside diagnostic
plots, confirmed the suitability of the GEVD model for the data. Additionally,
the Shapiro-Wilk test demonstrated the non-normality of the rainfall
datasets. Optimal block sizes for the r-largest order statistics model varied
across stations, with r-values ranging from 2 to 4. Both the standard GEVD
and r-largest GEVD models provided consistent return level estimates, suggesting
strong model performance. The bGEVD model further revealed a negative
time trend in rainfall maxima, resulting in lower return level estimates
compared to the other models. Return levels were calculated for return periods
ranging from 5 to 250 years, highlighting that extreme rainfall events
become increasingly likely with longer return periods. Overall, the findings
of the study offer valuable insights into the behaviour of extreme rainfall in
KwaZulu-Natal province, with significant implications for risk management,
infrastructure planning, and disaster preparedness
Determination of the diagnostic accuracy of rapid molecular assays for diagnosis of bloodstream infections in Pietersburg Hospital, Limpopo Province in South Africa
Full text linkThesis (M. Sc. (Medical Sciences)) -- University of Limpopo, 2025Background Bloodstream infections (BSIs) are infections caused by the presence of viable microorganisms in the bloodstream and are an essential cause of morbidity and mortality, especially among immunocompromised individuals. Blood culture is still considered the gold standard for diagnosing BSI; however, prolonged turnaround times limit its use for rapid clinical decision-making. Rapid molecular techniques such as BioFire FilmArray Blood Culture Identification 2 (BCID2) and metagenomics sequencing offer faster turnaround time for pathogen detection than blood cultures, in which bacteria, fungi, and their resistance profiles are detected.
Aim
This study aimed to evaluate the diagnostic accuracy of rapid molecular assays for the identification of pathogens that causes bloodstream infections at Pietersburg Hospital in Limpopo Province, South Africa.
Objectives
Compare the effectiveness of VITEK2 and BCID2 in identifying pathogens present in positive blood culture samples. Metagenomics was used to resolve the discrepant results between the VITEK2 and BCID2.
To assess the sensitivity, specificity, and overall diagnostic accuracy of the BCID2 and metagenomics in relation to the VITEK2.
Methods
This prospective diagnostic cross-sectional study was conducted at Pietersburg Hospital, Polokwane, Limpopo Province, South Africa. Participants included patients suspected of BSI from various regional hospitals and clinics of Limpopo Province.
Blood samples sent to the National Health Laboratory Service (NHLS) from May to December 2023 were included.
Blood culture bottles were incubated in a BacT/ALERT 3D system. Positive cultures were processed using VITEK 2 and BCID2, with discrepancies resolved by 16s metagenomics sequencing. Data were analyzed using SPSS (v29.0.0.0) and MedCalc (v22.026).
Results
Of the 247 samples, 78% (193/247) were Gram-negative bacteria, 17% (43/247) Gram-positive bacteria, and 4% (11/247) yeast. The ICU and paediatric units contributed 34% (83/247) and 29% (72/247) of samples, respectively. A total of 72% (179/247) were monomicrobial, while 28% (68/247) were polymicrobial. A total of nine off-panel organisms were detected in monomicrobial while seven were detected in polymicrobial samples. BCID2 achieved sensitivity, specificity, and accuracy rates of 95.7%, 95.5%, and 96%, respectively. CTX-M was the most common resistance gene (40%), while van A/B and VIM were rare (0.5% each). Metagenomics confirmed BCID2 in 72% of cases and VITEK 2 in 48%, detecting additional off-panel pathogens.
Conclusions The BioFire FilmArray Blood Culture Identification 2 test gave a very good diagnostic performance with 96% and a faster detection of pathogens in 60 minutes, making it appropriate in emergency cases. VITEK 2 had a better diagnostic performance but relatively slower and surpassed BCID2 in some pathogens. Metagenomics sequencing, being the most comprehensive technique, confirmed 72% of results from BCID2 and identified novel ones. These findings highlight the complementary roles of BCID2, VITEK 2 and metagenomics in optimizing bloodstream infection diagnosis in poor resource settingsNational Research Foundation (NRF) and BioMérieu
Psychological contract, employee vitality and organasational citizentionship behaviour among employees at a selected local municipality in the Sekhukhune District Municipality, South Africa
Full text linkThesis (M. Com. (Human Resource Management)) -- University of Limpopo, 2025The research adopts a quantitative approach to investigate the relationship between psychological contract, employee vitality, and organisational citizenship behaviour within the selected local municipality in the Sekhukhune district. The study encompasses 377 employees as its target population, all employed by the municipality. Sample size determination utilised an online Rao soft sample calculator. The exact sample size to be targeted after using an online Rao soft sample calculator was 95, attempts to get more participants were made. At the end, a sample size of 97 participated in this study. Data collection relied on self-administered questionnaires (Cross-sectional survey) known for their validity and reliability. Both convenience and quota sampling methods were employed to generate data for hypothesis testing. Analysis involved IBM - SPSS version 29.0 for statistical analysis. The relationship between the variables were tested using Pearson correlations (both primary and secondary hypotheses). Regression analysis was also carried out to test and measure the relationship between variables (primary hypotheses) and independent t-test was conducted to compare the gender differences. Findings suggest positive correlations between psychological contract and organisational citizenship behaviour, negative associations between employee vitality and psychological contract, and between employee vitality and organisational citizenship behaviour. Gender exhibited no significant differences in perceptions of psychological contract, employee vitality, and organisational citizenship behaviour. Additionally, positive relationships were found between psychological contract and various organisational citizenship behaviour dimensions, as well as between employee vitality and sportsmanship. Conversely, a negative relationship was found between employee vitality and several organisational citizenship behaviour dimensions. The study concludes by recommending that public sector entities should ensure that they understand and meet their employees’ expectations in order to positively influence their behaviour and enhance their organisational citizenship behaviour, particularly in uncertain contexts resembling the COVID-19 pandemic or technological changesNational Research Funding (NRF
The experiences of fathers of children with autism spectrum disorders at the selected special school in the Limpopo Province
Full text linkThesis (M. (Nursing)) -- University of Limpopo, 2025Background Autism spectrum disorder has a negative impact on the fathers of children with ASD as they come across several challenges while raising their children with ASD, which include stress, financial problems, and stigma. Fathers of children with ASD need to cope with having a child with ASD. Purpose The purpose of the study was to determine the experiences of fathers of children with ASD at the selected special school in the Limpopo province. Study methods This was a qualitative, explorative, and descriptive research design. Non-probability purposive sampling was used to select 15 fathers of children with ASD at the selected special school in the Limpopo Province. Data was collected using semi-structured interviews and analysed using Tesch's steps for data analysis. Results Four themes and 17 sub-themes were formulated from the data gathered by using Tesch's 8 steps of data analysis. Fathers of children with ASD were found to have various experiences of diagnosing and caring for a child with ASD. The fathers were found to be playing different roles in caring for the ASD child such as the caregiving role, family provider role, and protective and supportive role. However, fathers of children with ASD experienced challenges related to ASD, service provision, socially and psychologically. Several coping strategies were used to overcome the challenges. Conclusion The experiences and challenges faced by fathers of children with ASD highlighted in this study can be integrated to develop an intervention programme that will serve as a guide to support fathers of children with ASD