Browse
Recent Submissions
Item Machine learning-driven analysis of surface plasmon resonance imaging for automated biomolecular binding detection(2026-03) Tsebesebe, NkgapheT; Mpofu, Kelvin T; Mthunzi-Kufa, PSurface Plasmon Resonance Imaging (SPRI) enables label-free, real-time monitoring of biomolecular interactions across sensor surfaces, making it an essential tool for multiplexed diagnostics. However, high-dimensional SPRI image data introduces challenges for interpreting binding events, particularly in complex sample matrices. This study trains machine learning model based on Multinomial Naïve Bayes (MNB) classifier with SPRI images to automatically classify binding and non-binding events. Statistical feature analysis (mean intensity, standard deviation, entropy, gradient magnitude) revealed significant differences between the SPRI binding and non-binding images, leading to the model (with Laplace smoothing parameter α = 1.0) achieve 97.7% accuracy, 97.8% sensitivity, 97.6% specificity, and an area under the ROC curve (AUC) of 0.98. The approach holds a potential to automated and high-throughput diagnostics, reducing reliance on manual interpretation and potentially extending SPRI platforms diagnostic tools in clinical and research settings.Item Machine learning-enhanced LAMP-fluorescence assay for automated tuberculosis detection(2026-03) Tsebesebe, Nkgaphe T; Mpofu, Kelvin T; Sivarasu, S; Mthunzi-Kufa, PRapid and accurate disease detection is critical for effective clinical management and public health interventions, particularly in low-resource settings. Loop-Mediated Isothermal Amplification (LAMP) is a promising molecular diagnostic technique due to its simplicity, speed, and compatibility with real-time fluorescence detection. However, interpreting fluorescence images, especially in cases with low target concentrations or high background noise remains challenging and error-prone when performed manually or using basic threshold-based methods. This study proposes the integration of a Support Vector Machine (SVM) algorithm with the LAMP-fluorescence assay to automate and enhance the detection of tuberculosis (TB) at low concentrations. Using a dataset of fluorescence images, the model was trained with 5-fold cross-validation across 20 candidate configurations (100 total fits). The optimized model achieved a classification accuracy of 99.76%, effectively distinguishing true positives, false positives, and ambiguous amplification patterns. Diagnostic sensitivity and specificity improved to 95.0% and 97.0%, respectively, supporting earlier and more reliable detection. This enhanced approach demonstrates strong potential for developing robust, scalable, and fielddeployable diagnostic tools for TB and other infectious diseases.Item Proposed amendments to SACAA part 101 regulations progress and future outlook(2026-07) Chirwa, MaryPart 101 of the South African Civil Aviation Regulations governs unmanned aircraft systems (UAS) operations and was introduced early to ensure aviation safety. However, the framework was largely adapted from manned aviation regulations and applies a uniform, certification-driven approach to operations with very different risk profiles. As the UAS sector has evolved, concerns have emerged regarding whether the current regulatory structure continues to support innovation, accessibility, and compliance, particularly for low-risk operations, research activities, and emerging manufacturers. This paper presents a structured review of Part 101 and a proposal for targeted amendments developed through stakeholder consultation, expert engagement, and comparative benchmarking against international regulations and standards. The proposed framework introduces a proportionate, risk-based regulatory architecture, differentiated operational categories, and dedicated provisions for institutional and research operations whilst managing safety and security risks associated with unsecured airspace. The analysis demonstrates that a risk-based approach can reduce regulatory barriers and improve compliance while maintaining core aviation safety objectives.Item Digital governance in the generative AI era: A qualitative analysis of privacy and trust in South African organizations(2026-07) Mlambo, Tlangelani P; Zuva, T; Brown, A; Moletsane, RSouth African organizations are experiencing rapid digital transformation as digital technologies increasingly shape service delivery, data management, and interactions between the state and citizens. In this time where machines such as Generative Artificial Intelligence (GenAI) can generate answers from the user prompt, there’s a need to control their use in order not to harm others. This study examines digital governance in the generative AI era, focusing on how privacy and trust are managed within organizations in South Africa. The study adapts a qualitative research approach utilizing the traditional literature review and secondary data analysis. There’s a disconnect between the initiated policies and the implementation or regulation of these policies, therefore, there is a need to develop a comprehensive digital governance framework for South African organizations to help minimize any harm that can be caused by use of Generative Artificial Intelligence. The research shows that in the generative AI era, is necessary for digital governance to move from only technologycentric strategies to a comprehensive framework that will combine institutional, ethical and operational aspects. The study contributes theoretically and practically to a deeper understanding of governance challenges and opportunities in South Africa. The study is limited to South African organizations and on secondary data from existing literature. Future work can extend the scope of the study beyond South Africa and conduct empirical and longitudinal studies to gain more insights on digital governance in the generative AI era.Item Bridging the gap from development to implementation of tools for the South African Transport Department(2026-07) Chirwa, Mary; Mukwena, Shothodzo T; Roux, Michael P; Muronga, KhangweloThe South African government has invested significantly in developing digital decisionsupport systems to improve road infrastructure planning and maintenance. Despite their technical robustness, the impact of these tools, such as the Abnormal Loads System, Road Asset Management Systems (RAMS), the PotholeFixGP application, and the STRUMAN Bridge Management System, has been limited by institutional, governance, and funding challenges. This study examines whether these systems are being used as intended, and identifies the barriers between development, implementation, and sustained network-level use. Drawing on interviews with system developers and implementation stakeholders, the paper presents four case studies highlighting common implementation bottlenecks. Findings show that while technical design was sound across all systems, uptake was constrained by weak institutional ownership, inconsistent operational funding, and misalignment with budgeting and decision-making processes. The study concludes with recommendations for strengthening the long-term integration of digital tools into infrastructure governance frameworks.Item Data governance frameworks for enabling responsible AI in small, medium, and micro enterprises: A systematic literature review(2026-06) Siphambili, Nokuthaba; Nelufule, Nthatheni NDisruptive technologies such as Artificial Intelligence (AI) have brought about changes in how organisations function. The adoption of AI has been applied in various industries, ranging from smart energy, smart transportation, smart health such as cancer treatment to managing automated cybersecurity threats and responding to sophisticated cyber threats. This presents opportunities and challenges as small, medium and micro enterprises (SMMEs) are also adopting AI to drive innovation, efficiency, and competitiveness. When compared to large organisations, SMMEs often lack the resources, expertise, and infrastructure necessary to implement comprehensive data governance frameworks, which are essential for responsible AI deployment. This study aims to investigate how data governance frameworks can enable responsible AI practices, specifically within the context of SMMEs. This study adopts a systematic literature review where the PRISMA framework is used to extract information on the data governance principles, challenges and opportunities that SMMEs can use for responsible AI. This study's findings reveal that effective governance plays a critical role in the adoption of AI within SMMEs. Existing data governance frameworks provide guidance, even though they are complex, which is a limitation for SMMEs. This study highlights opportunities for SMMEs in the data governance frameworks in enabling responsible AI. This study also reveals the need for a universal data governance framework.Item Integrating NIST SP 800-101 and ISO/IEC 27000 for mobile device forensics: A conceptual framework(2026-07) Agenbag, S; Henney, A; Meyer, HeloiseMobile device forensics has become central to digital investigations, yet existing process models for smartphones and other portable devices remain fragmented, platform-specific, and inconsistently aligned with international standards. This paper proposes an integrated conceptual framework for mobile digital forensics that combines the operational phases of NIST SP 800-101 Rev. 1 (Preservation, Acquisition, Examination/Analysis, Reporting) with the governance and validation principles of the ISO/IEC 27000-family standards, including ISO/IEC 27037, 27040, 27041, 27042, 27043, and 27050. The conceptual framework explicitly incorporates contemporary mobile challenges— such as platform diversity, file-system variation, strong encryption, applicationcentric data storage, high data volatility, and cloud dependencies—and maps them to concrete activities and controls at each phase. The paper illustrates how this integrated model can guide evidence handling, tool validation, and artefact analysis in mobile investigations, addressing long-standing inconsistencies in mobile forensic practice. The proposed conceptual framework aims to support more defensible, repeatable mobile device forensics and to provide a basis for future empirical evaluation and tool-agnostic checklists.Item 5G-enabled automated plant watering system for small farmers using srsRAN and Open5GS(2026-07) Gumbi, Zekhethelo S; Mtsotso, Steve; Mamushiane, Lusani; Kobo, Hlabishi ISmart irrigation systems rely on sensing, automation, and communication to improve water-use efficiency and reduce manual intervention. In rural South Africa, however, small-scale farmers continue to face persistent constraints including water scarcity, prolonged dry seasons, limited digital connectivity, and increasing irrigation energy costs, which obstruct the adoption of precision irrigation technologies. This paper presents an automated plant watering system implemented over an opensource private 5G standalone (SA) network using srsRAN and Open5GS. The proposed system integrates soil moisture and environmental sensing, threshold-based irrigation control, and real-time communication between a low-cost sensing node and a Python-based edge application server. In the deployed architecture, the sensing and actuation node connects via Wi-Fi to a 5G user equipment (UE), which provides broadband access to the private 5G network. The system is experimentally evaluated using both network-level and application-level performance measurements. The private 5G testbed achieved round-trip time (RTT) between 19.557 ms and 159.896 ms, uplink throughput of 16.4 Mbps, downlink throughput of 50.9 Mbps, and zero packet loss during testbed evaluation. During application operation, end-to-end RTT ranged from approximately 60 ms to 170 ms for local operation and increased beyond 300 ms when a cloud-based monitoring platform was introduced. The results show that open-source private 5G SA infrastructure can support responsive smart irrigation services while providing a practical and affordable platform for experimentation and pilot-oriented deployment. This study contributes experimental evidence for broadband-based smart irrigation over private 5G SA, an area that remains relatively underexplored compared with low power wide area network (LPWAN) agricultural Internet of Things (IoT) approaches.Item A review of injury risk curve development to support human factors and ergonomic design under extreme loading conditions(2026-02) Mokgotho, Lebone Y; Sekhuthe, Karabelo L; Modungwa, Dithoto MConducting risk assessments in underbody blast (UBB) scenarios on landmine-protected vehicles is important for evaluating vehicle safety, crew injury risk, and the effectiveness of blast protection systems. Due to the hazardous, costly, and resource-constrained nature of UBB testing, injury risk assessment commonly relies on survival analysis methods to estimate injury probabilities from limited and censored experimental data. This paper presents a structured review of survival analysis methods used to develop lower-limb injury risk criteria under UBB loading conditions. Parametric, non-parametric, and semi-parametric survival models are reviewed and compared with respect to their assumptions, data requirements, and suitability for constrained experimental environments. Emphasis is placed on their application to injury risk curve development for lower extremity injuries. The review highlights methodological strengths, limitations, and practical implications of model selection for engineering decision-making, including risk threshold definition, test evaluation, and validation of protective systems. A limited illustrative dataset is used to demonstrate the application of selected methods. The findings support more consistent and transparent use of survival analysis techniques in injury risk assessment for extreme loading scenarios.Item A hybrid, transparent trust and risk assessment framework for cryptocurrency exchanges(2026-06) Mawhayi, B; Botha, Johannes G; Leenen, LCryptocurrency exchanges act as critical intermediaries within the digital asset ecosystem, yet users currently rely on largely opaque, platform-defined trust scores to assess their reliability and risk. Existing industry frameworks, notably those produced by CoinGecko and CoinMarketCap, provide useful signals related to liquidity and volume integrity but suffer from limited transparency, fixed weighting schemes, and the absence of sentiment-based assessment. This paper presents HTREx (Hybrid Trust and Risk Evaluation framework for exchanges), a semi-automated, modular trust and risk assessment framework for cryptocurrency exchanges that addresses these limitations. The framework integrates five dimensions of exchange integrity: user sentiment, regulatory compliance, technical security, transparency, and incident history. Sentiment is quantified using transformer-based natural language processing applied to user-generated content from mobile application reviews and online forums. Compliance is assessed through structured extraction of regulatory and operational disclosures from Terms of Service documents using large language models. Security, transparency, and incident history are evaluated through a combination of publicly verifiable indicators, third-party assessments, and a recency-weighted incident scoring model. All components are normalised and aggregated into a composite score using user-adjustable weights, enabling personalised risk prioritisation while retaining a defensible default configuration for comparative analysis. The framework is demonstrated using four prominent exchanges—Kraken, Coinbase, Binance, and Uniswap—highlighting clear differences between centralised and decentralised platforms and illustrating how sentiment, compliance, and historical incidents materially influence overall trust assessments. The results suggest that transparent, extensible, and user-configurable scoring models can provide a more interpretable and context-sensitive evaluation of exchange risk than existing monolithic trust scores, with direct relevance for both retail and institutional participants.Item Evaluating an investigative process for cryptocurrency-related crimes(2026-03) Botha, Johannes G; Singh, Kreaan D; Leenen, LThis paper evaluates a previously proposed investigative process for cryptocurrency-related crimes, originally introduced by the authors (Botha, Singh, & Leenen, 2025a), through the application of a real-world case study. The process covers crime reporting and case registration, on-chain analysis, off-chain analysis, and the transformation of investigative intelligence into court-admissible evidence. The current study focuses on a new, active case involving an elderly South African (SA) woman who was defrauded of a substantial portion of her pension through a fraudulent investment scheme known as ###-Platform (redacted). The case is presently under investigation by the Directorate for Priority Crime Investigation (DPCI), a specialised unit of the South African Police Services (SAPS) tasked with addressing serious economic crimes and commonly referred to as the Hawks. By systematically applying the proposed investigative process to this case, the study assesses the framework's practical utility, adaptability, and effectiveness in real-world conditions. The analysis further reflects on legal, technical, and procedural challenges encountered during the investigation, offering critical insights for law enforcement, regulators, and cybersecurity professionals. It also highlights broader systemic vulnerabilities that facilitate such scams, particularly among elderly and non-technical populations. The findings underscore the need for enhanced public education, improved regulatory oversight, and international cooperation in combating cryptocurrency fraud. Ultimately, the paper contributes to the evolving discourse on financial crime in the digital age and aims to support the development of more secure and accountable crypto-investment environments.Item 5G core network load testing traffic generator: Emulating real-world traffic scenarios(2026-12) Mukute, T; Lysko, Albert A; Mwangama, JThis paper introduces the Core Network Traffic Generator, a comprehensive performance evaluation and compliance testing tool for 5G core networks that addresses significant limitations in existing solutions. Our tool uniquely combines control plane and data plane traffic generation with high-precision performance metrics collection. Built on well-established projects, pycrate for protocol message handling and eBPF/BCC for kernel-level monitoring, it collects SCTP transport metrics with sub-microsecond precision—a capability absent in current tools. Key features include simultaneous emulation of gNodeB and multiple UEs, granular 5G procedure timing analysis, detailed SCTP performance visualisation, and automated compliance validation against 3GPP standards. Validated with major open-source implementations (Open5GS, free5GC, OAI) and proven in research publications, the system operates on commodity hardware and is available under the MIT license, offering an accessible yet powerful solution for both academic research and operational network testing.Item Reflections and inferences on processes and challenges in procuring new subsidised service contracts in South Africa(2026-07) Mabece, RL; Chauke, Risenga S; Mokonyama, Mathetha TSubsidised public transport contracts for road-based public transport services have been operating in the form of ever-green contracts for many years. Majority of the contracts stem as far back as 1997 and keep being extended without going out to a new procurement process as envisaged in the Constitution. Some provinces have started to try and change the trajectory through tendered and negotiated contracts approaches as provided for in terms of sections 41 and 42 of the National Land Transport Act No. 5 of 2009 (NLTA), and others have tried to issue new tenders without following the provisions of the NLTA. Through a techno-legal review, interpretation and analysis of primary and secondary sources the paper provides a critical exposition of the regulatory framework and practices regarding contracting for subsidised public transport service contracts. The paper is necessary to address some of the in-session questions raised previously by delegates of the Southern African Transport Conference. The paper is also necessary to guide practitioners engaged in the procurement of these services. It also reflects on practical challenges in applying both processes and make recommendations on future public procurement processes considering the current applicable legal framework informed by the NLTA as amended by the National Land Transport Amendment Act No. 23 of 2023 and new contracting regulations. Specific practical recommendations are made pursuant of alignment with the Constitution and improved public transport service delivery.Item Road safety risks among motorcycle food delivery riders in South Africa’s gig economy: Findings from a survey in Pretoria(2026-07) Kwange, Mamokgolwane L; Dube, Mxolisi SSouth Africa’s app-based food and parcel delivery boom has driven a surge in motorcycle use for last mile logistics employing tens of thousands of riders mainly in urban areas. This growth has raised road safety alarms with motorcycle fatalities rising from 1.9% to 3.9% of total road deaths between 2023 and 2024. This quantitative survey of 109 motorcycle delivery riders in Pretoria captured self-reported data on crashes, near misses, protective gear practices and risky behaviours linked to gig economy pressures. Findings reveal a cycle of fatigue, and conspicuity gaps heightening vulnerability. Using the Safe System Approach recommendations include mandatory training, app rewards, speed controls and lanes, road upgrades, vehicle standard enforcement and better post-crash support. This study fills a key data gap on South African delivery riders advocating for systemic reforms to improve safety in the expanding gig-economy urban transport landscape.Item Bridging global metrics and local realities - A modelling framework for sustainability in ports(2026-07) Weerts, Steven P; Taljaard, Susan; Slinger, J; Vreugdenhil, H; Nzuza, CSustainability has become a strategic business imperative for ports, extending beyond environmental stewardship to encompass long-term economic viability, social responsibility, and operational resilience. Increasingly, ports require a societal license to operate, sustained access to funding, and the ability to maintain competitiveness in a rapidly evolving global maritime sector. Demonstrating sustainability performance, however, requires a robust and transparent method for assessing diverse and complex data across environmental, social, and economic dimensions. A sustainability index provides such a mechanism by offering a conceptual and mathematical framework that evaluates, normalizes, and aggregates multiple raw indicators into a single, standardized numerical score. This integrated model enables ports to measure performance consistently, benchmark progress, support informed decision-making, and communicate sustainability outcomes to stakeholders. By transforming complex datasets into actionable insights, a sustainability index enhances resource management, promotes accountability, and supports the transition toward more sustainable and competitive port operationsItem Women in Laboratory Leadership: Journey, Challenges and Opportunities(2026-02) Zwane, G; Kesa, H; Dikgale, Moyahabo S; Naicker, D; Mketo, N; Kumari, S; Campbell, G; Rasephei, K; Vermaak, KThe Circuit Media Pty Ltd is delighted to announce the dates to the 1st Africa Women in Laboratories Conference 2026. This convergence will bring together, under one roof, all laboratory managers and staff across a multiplicity of sectors for the purposes of enhancing attitudes, sharing knowledge, trends, approaches, policies and mechanisms to develop lab skills, training, promoting women and gender mainstreaming. The Women in Laboratories Africa Conference is an opportunity for you to create new professional networks and contacts, develop new laboratory partnerships, meet new people, rub shoulders with women working in laboratories across Africa, share experiences, learn new skills and benefit from idea exchanges and training. Experience powerful presentations, keynote sessions, future insights, skills, and trends from high-level executives in the day-to-day management of laboratories in Africa and the world.Item A unified framework for secure and interoperable digital identity in South Africa(2026-05) Mthethwa, Sthembile N; Ntshangase, Cynthia S; Myaka, Zanele S; Ndhlovu, Nomalisa; Lefophane, SamuelDigital identity represents a technological advancement facilitating the secure access and management of identities for both individuals and organisations within digital environments. Despite considerable progress in digital identity systems over the past decade, their implementation has been hindered by challenges, including fragmentation, lack of interoperability, and inadequate governance framework. This study analyses global digital identity frameworks to identify best practices and opportunities for accelerating adoption, focusing on key principles such as privacy protection, trust, inclusivity, and cross-border interoperability. Additionally, it outlines the essential requirements for a well-governed digital identity ecosystem and proposes a national digital identity framework consistent with international standards, aimed at enhancing the security, resilience, and global recognition digital identities issued in South African.Item Using various 2 Dimensional (2D) cancer cell lines to screen for anti-cancer drugs for applications in the development of a predictive high throughput (HTS) drug screening platform(2023-10) Nxumalo, Precious ZThe efficacy of widely available cancer therapy options is mainly based on the molecular and biochemical profile from patient cohort from high income countries. Thus, there is a critical need to study on low-middle income countries cancer patient such as African cohort. In addition, gynecological cancer is under-studied and hence is an unmet medical need due to high chemo-resistance and recurrence rates in African patients. The project goals are to develop a high-throughput in-vitro drug screening platform that will use patient-derived samples to predict how adjuvant therapy would affect African ovarian cancer patients. By employing our proposed approach, we hope to find drug or drug combinations that can combat chemotherapy resistance and ultimately offer tailored-specific therapeutic alternatives for African cancer patients within clinically appropriate timeframes to the prescribing oncologist. We have demonstrated the cytotoxicity of clinically approved drugs on various 2D cancer cell lines. We believe drug sensitivity screening would be a gold standard to evaluate the drug dosage response. Thus, we aim to implement this method with automated drug screening pipeline in cancer cell lines and patient derived samples due to the controlled conditions and repeatable procedures. Based on our findings we are certain that by choosing the correct tumor model at the stage of ex-vivo testing we will establish an automated technology platform for drug sensitivity screening on patient samples, thereby allowing immediate benefit to the patient who donates the sample for analysis and thus contributing towards efforts for individualized treatment as a part of personalized medicine.Item Machine learning-based prediction of remaining useful life of mobile devices for circular economy e-waste management(2026-05) Thovhale, Mulisa; Rananga, S; Ebrahim, Rozeena; Masonta, Moshe TThis paper presents a Machine Learning (ML) approach for predicting the Remaining Useful Life (RUL) of mobile phone devices to support circular economy strategies aimed at reducing e-waste. The study addresses the challenge of premature device disposal by developing a prediction model that estimates how long a device can continue functioning before reaching end of life. A synthetic dataset representing 1,000 devices operating over 1,460 days was generated using mathematical degradation equations, and four ML models were evaluated: Random Forest, Gradient Boosting, Support Vector Regression and a Long Short-Term Memory (LSTM) network. The LSTM achieved the best performance, with a Mean Absolute Error (MAE) of 153 days and an R² (coefficient of determination) of 0.47. The results show that RUL prediction can support repair, refurbishment and recycling decisions, enabling more sustainable device lifecycle management.Item PrivSev: Privacy-Preserving artificial intelligence in 6G Open Radio Access Networks: A survey(2026-05) Nelufule, Nthatheni; Siphambili, Nokuthaba; Shadung, Lesiba DThe disaggregated, multi-vendor architecture of Open Radio Access Networks (O-RAN) in 6G, promises an unprecedented flexibility through the AI-native intelligence, and cost efficiency. However, these benefits also introduce challenges such as severe privacy and security risks which include the model inversion, data poisoning, and unauthorized access across distributed edge nodes; mainly Open Radio Unit (O-RU), Open Distributed Unit (O-DU), Open Centralized Unit (O-CU), and RAN Intelligent Controllers (RICs). In this paper, a systematic review based on the PRISMA framework was used to synthesize 42 peer-reviewed articles published between 2020 and 2026, particularly on the privacy-preserving AI techniques, Federated Learning (FL), Differential Privacy (DP), Secure Multi-Party Computation (SMPC), and emerging hybrid technologies applied to 6G O-RAN environments. The key research findings revealed that the combination of the Zero trust Architecture (ZTA) and FL can achieve up to 32% energy savings and Near-RT compliance, while the combination of DP and FL helps to secure the RIC and FBMP, and the Intrusion Detection System (IDS) helps to enable lightweight Multi-Party Computation (MPC). The notion of introducing a three-way FL, DP and SMPC integration for O-RAN remains unexplored, and this work bridges this gap, by introducing Privacy-Preserving (PrivSev), which is a novel layered hybrid framework that applies lightweight DP at the edge clients and threshold SMPC at the Non-RT RIC. The projected performance of the proposed framework tested against the reviewed benchmarks promises a 92% baseline accuracy retention.