Claim Missing Document
Check
Articles

Found 19 Documents
Search

Early Detection of Diabetic Retinopathy Through Explainable AI Models: A Systematic Review Ngwazi, Tinashe; Ndlovu, Belinda; Maguraushe, Kudakwashe
IJID (International Journal on Informatics for Development) Vol. 14 No. 2 (2025): IJID December
Publisher : Faculty of Science and Technology, Universitas Islam Negeri (UIN) Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/ijid.2025.5200

Abstract

Diabetes, if not detected early, can lead to serious complications such as vision loss, known as diabetic retinopathy. Explainable Artificial Intelligence (XAI) can enhance traditional Machine Learning methods, which are not understandable and transparent in diagnostic tasks. This Systematic Literature Review explores data inputs that influence the performance of XAI models in detecting diabetic retinopathy, how XAI techniques can enhance early detection outcomes in diabetic retinopathy, the challenges in implementing these techniques and the ethical implications of using these models in clinical practice. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses approach guided the search in 4 databases, Springer, Science Direct, PubMed and IEEE Xplore. The findings reveal that XAI techniques like Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP) and Gradient-weighted Class Activation Mapping (GRAD-CAM) offer opportunities like early detection outcomes, integration with existing clinical processes, enhancing trust in AI systems, improving accuracy and personalised treatment. XAI can also facilitate collaboration among clinicians, maintaining fairness in AI systems and supporting adherence to ethical standards. However, research on clinical validation of these models, as well as standardised performance evaluation metrics, is lacking.
Quantum Computing Cryptography: A Systematic Review of Innovations, Applications, Challenges, and Algorithms Maitireni, Peter; Ncube, Vusumuzi; Ndlovu, Belinda; Sibanda, Thando
Journal of Information System and Informatics Vol 7 No 4 (2025): December
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v7i4.1331

Abstract

This study explores how to build quantum-resistant systems to safeguard digital infrastructure in the post-quantum era by uncovering the innovations, applications, algorithms, and challenges of Quantum Computing cryptography. Utilizing the Preferred Reporting Items for Systematic Reviews and Meta-Analyses approach a search was conducted across the following databases for the years 2021–2025: PubMed, IEEE Xplore, ScienceDirect, SpringerLink, and Google Scholar. We shortlisted 15 studies from 519 screened articles for a comprehensive evaluation based on their relevance. Findings show strong adoption in finance, healthcare, IoT, cybersecurity, and e-government, with lattice-based PQC emerging as the most dominant cryptographic family, followed by QKD and hybrid PQC–QKD models. The review highlights key obstacles, including transition complexity, lack of global standards, high implementation costs, and integration difficulty. The study contributes by providing the first sector-aligned synthesis of innovations, identifying algorithmic trends, and mapping global research disparities through a conceptual model. It also presents a structured set of future research directions to guide policymakers, cryptographers, and practitioners preparing for quantum-enabled threats.
A Systematic Review of Post-Quantum Cryptography for Healthcare Data Protection: Performance, Readiness, and Deployment Challenges Ngwenya, Taboka; Ndlovu, Belinda
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.11836

Abstract

The traditional cryptographic methods used to protect healthcare data, especially for the long-term storage of medical imaging records, are becoming increasingly threatened by the quick development of quantum computing. The purpose of this study is to assess the challenges, efficacy, and preparedness of integrating Post-Quantum Cryptography (PQC) into healthcare information systems. Twenty peer-reviewed studies published between 2020 and 2025 were analysed following the Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) protocol. The review was conducted using a systematic research design that included qualitative thematic synthesis, predetermined eligibility criteria, and database searching. According to the results, lattice-based PQC schemes, specifically, CRYSTALS-Kyber for encryption and CRYSTALS-Dilithium for authentication, show great promise because of their effectiveness, resilience, and suitability for decentralized architectures like blockchain and Internet-of-Medical-Things environments. Nonetheless, the review points out a notable deficiency of empirical assessment in actual healthcare settings, particularly with regard to cloud-based platforms and Picture Archiving and Communication Systems utilized in medical imaging processes. Scalability limitations, intricate key-management specifications, system interoperability restrictions, and the requirement for conformity with regulatory and compliance frameworks are some of the major issues noted. The results indicate that lattice-based PQC schemes have great promise, deployment readiness remains largely at the conceptual and experimental stage, particularly for cloud-based PACS environments. Real-world implementation validation in a healthcare setting has not been achieved.
Explainable Transformer and Machine Learning Models in Predicting Tuberculosis Treatment Outcomes. A Systematic Review Sibanda, Shumirai; Ndlovu, Belinda
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.11846

Abstract

Tuberculosis (TB) remains a major health challenge, and predicting treatment outcomes continues to be difficult in real-world settings. Recent advances in Artificial Intelligence (AI), particularly transformer-based models, have shown promise in modelling longitudinal, multimodal, and heterogeneous TB data. However, their clinical adoption is constrained by limited interpretability, fairness concerns, and deployment challenges. This study presents a systematic literature review of explainable transformer and machine learning models used for TB prognosis. Following PRISMA guidelines, searches across ACM, IEEE Xplore, PubMed, and ScienceDirect identified 17 peer-reviewed studies published between 2020 and 2025 that met the inclusion criteria. The review synthesises evidence on predictive performance, explainability techniques, and deployment considerations. Findings indicate that transformer-based and deep learning models generally outperform conventional machine learning approaches on longitudinal and multimodal data. In contrast, traditional models remain competitive for tabular clinical datasets. Explainability approaches are dominated by feature importance methods and SHAP, with limited use of intrinsic transformer interpretability mechanisms. Persistent challenges include data scarcity, limited generalisability, computational overhead, insufficient evaluation of fairness, and weak alignment with real-world TB care workflows. Building on these findings, the study proposes the Explainable Transformer Adoption Model for TB Prognosis (ETAMTB) as a conceptual clinical adoption framework integrating multimodal transformers, explainability layers, clinician-facing interfaces, and deployment enablers. Overall, the review concludes that effective AI adoption in TB care requires balancing predictive performance, interpretability, and equity, and that explainable transformers should currently be viewed as promising but largely experimental tools rather than deployment-ready solutions.
Transformer-Based Models for Electronic Health Records and Omics in Healthcare: A Systematic Literature Review Machemedze, Joshua; Ndlovu, Belinda
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.11893

Abstract

Electronic Health Records (EHRs) have become central to modern healthcare. The emergence of transformer-based models has profoundly influenced how EHRs are used for modelling complex, longitudinal data. Integration with omics technologies improves the precision of disease identification and risk assessment during modelling. While several reviews have examined transformers in healthcare broadly, a systematic synthesis focused on their architectural design, empirical performance and integration of EHRs with omics data remains limited. This study presents a systematic literature review of transformer-based models applied to electronic health records (EHRs) and omics data, and of their integration into healthcare. Following PRISMA guidelines, peer-reviewed studies were retrieved from IEEE Xplore, ACM Digital Library, PubMed, and ScienceDirect, resulting in 14 eligible empirical studies published between 2020 and 2025. The review analyses transformer architectures, submodules, application domains, comparative performance, interpretability mechanisms, and limitations. Findings indicate that architectural design drives task-specific advantages in disease prediction, phenotyping, medication recommendation, and omics analysis. The integration of self-attention with deep learning, temporal modelling, and a pre-trained biomedical transformer improves performance. However, most studies remain centred on EHR, with limited empirical integration of omics data. Persistent challenges include limited generalisability, high computational cost, data quality issues, and insufficient interpretability for clinical deployment. The primary contribution of this review lies in synthesising architectural trends and methodological gaps. By consolidating current evidence, the study provides clear directions for the development of explainable, generalisable, and multimodal transformer-based systems in precision healthcare.
Transformer-based Models for Cardiovascular Disease Predictions from Electronic Health Records: A Systematic Review Chikumo, Onayi Theresa; Ndlovu, Belinda
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.11899

Abstract

This systematic literature review (SLR) analyses 16 studies published between 2020 and 2025 that applied transformer-based or other machine learning models to predict cardiovascular disease (CVD) using electronic health records (EHRs). Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, the review ensures transparency in the identification, screening, and quality appraisal of eligible studies. The key findings reveal a rapid shift from traditional machine learning models, such as Random Forest, toward transformer architectures like the Bidirectional Encoder Representation from Transformers for Electronic Health Record (BEHRT) and its variants. These models demonstrate a superior discrimination (Area Under Curve:0.84 to 0.93) due to their capacity to model long-term temporal dependencies. Explainable AI (XAI) tools, such as attention visualisation, were frequently employed, yet clinical interpretability and integration into decision support remain underexplored. The review also highlights opportunities in federated and privacy-preserving learning, multimodal data fusion, and hybrid architectures that integrate transformers with traditional machine learning methods. This review addresses a gap in the past literature by being the first SLR to compare transformer variants for the prediction of CVDs. Other SLRs examined general CVD risk models, but the present SLR analyses interpretability, external validation and methodological limitations to transformer models. The findings of the recent SLR reported challenges that include data-shift limitations, model-poor population generalisation and their limitations to clinical adoption, which highlights the need for more evaluation protocols and clinicians’ interpretability frameworks.
Multi-Agent Retrieval Augmented Generation for Clinical Decision Support: A Systematic Review and Integrative Conceptual Framework Mugambiwa , Tarisai; Ndlovu, Belinda
Journal of Applied Informatics and Computing Vol. 10 No. 1 (2026): February 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i1.11900

Abstract

Multi agent retrieval augmented generation (RAG) systems are increasingly explored as advanced architectures for clinical decision support combining information retrieval, reasoning and verification through coordinated agent interactions. This study systematically reviews applications of agentic and multi agent RAG in clinical decision support systems (CDSS) and synthesizes an integrative conceptual framework linking technical design to technology adoption considerations. Following PRISMA guidelines, searches were conducted from PubMed, IEEE Xplore and ScienceDirect using structured Boolean strings combining terms for multi agent architectures, RAG and CDSS.The search yielded 12 studies published between 2020 and 2025 that met the inclusion criteria. The review synthesises evidence on multi agent role configurations retrieval and reasoning strategies, verification mechanisms and reported clinical contexts. Across studies, dominant challenges include data and corpus limitations retrieval quality dependency, limited clinical validation and computational overhead, alongside governance concerns such as privacy, bias and accountability. Building on the synthesis, we propose a four-agent CDSS framework retriever, reasoner, verifier, safety and map its deployment determinants to Technology Acceptance Model constructs perceived usefulness, perceived ease of use, trust and diffusion of Innovations attributes. The review concludes with design-oriented recommendations for safer, explainable, and adoption-ready multi-agent RAG CDSS, particularly for low-resource contexts.
A Personalized Generative AI Model for Diabetes Drug Discovery: Integrating Molecular and Clinical Data Using Variational Autoencoders (VAE) Ndlovu, Belinda
The Indonesian Journal of Computer Science Vol. 15 No. 1 (2026): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v15i1.5084

Abstract

Diabetes drug discovery remains slow, costly, and insufficiently personalised, particularly in resource-constrained healthcare settings. This study proposes and empirically evaluates a personalised, generative artificial intelligence framework that integrates molecular and clinical data to generate diabetes drug candidates. Guided by the CRISP-DM framework, a hybrid Clinical–Molecular Variational Autoencoder (VAE) architecture was developed, combining molecular representations with anonymised patient metabolic profiles, including HbA1C, fasting glucose, BMI, cholesterol, and age. Molecular data were sourced from PubChem and ChEMBL, and generated compounds were evaluated using drug-likeness metrics, molecular validity checks, and downstream effectiveness classification. The model successfully generated chemically valid, drug-like molecules with average Quantitative Estimate of Drug-likeness (QED) scores above 0.5. At a fixed decision threshold, effectiveness classification achieved an accuracy of 0.80; however, probability calibration analysis revealed limited discriminative reliability across thresholds (AUC = 0.49), highlighting the impact of class imbalance. Unlike prior molecule-centric generative drug discovery approaches, this study presents one of the first empirically evaluated Clinical–Molecular dual-VAE frameworks for personalised diabetes drug discovery, explicitly integrating patient metabolic profiles while revealing calibration limitations in generative pharmaceutical pipelines.
Developing a graph-based machine learning model for identifying money laundering networks associated with sanctioned entities in a bank in Zimbabwe Ndlovu, Belinda; Kiwa, Fungai Jacqueline; Muduva, Martin; Chipfumbu, Colletor T.; Marambi, Sheltar; Maphosa, Amazing
Indonesian Journal of Innovation and Applied Sciences (IJIAS) Vol. 6 No. 1 (2026): February-May
Publisher : CV. Literasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47540/ijias.v6i1.2306

Abstract

Money laundering networks associated with sanctioned entities pose a significant risk to financial systems, often operating through complex relational transaction structures that evade traditional rule-based monitoring. While graph neural networks have demonstrated promise in financial crime detection, limited work has formally modelled sanction-linked transaction networks within highly imbalanced banking datasets under consistent comparative evaluation. This study proposes a directed weighted graph-based learning framework for identifying sanction-associated money laundering networks using real-world banking transaction data. Transactions were modelled as relational graphs, with accounts as nodes and transfers as weighted edges, and evaluated using a Graph Convolutional Network (GCN) against classical and ensemble classifiers. The proposed model achieved an accuracy of 88.18%, F1-score of 0.7345, ROC-AUC of 0.8968, and a superior Matthews Correlation Coefficient compared to baseline methods. Results demonstrate that relational graph modelling improves the detection of structurally coordinated laundering behaviours that are not captured by independent transaction classifiers. These findings support the integration of graph neural network architectures into anti-money laundering systems to enhance sanction-linked detection capabilities in complex financial networks.