JOURNAL ARTICLE
Privacy-Preserving Synthetic Data for HIV Risk Prediction in People Who Inject Drugs in Kenya: Benchmarking Differentially Private Generative Models
Fedha Daniel Murunga1, Betty Mayeku1, Juma Kilwake1, Mercy Nyakowa2
1 Department of Computer Science, Kibabii University, Bungoma, Kenya2 Kenya Ministry of Health, National AIDS & STI Control Program, Nairobi, Kenya
International Journal of Engineering and Computer Science, Vol. 15(07), pp. 28694-28698 · 2026-07-10 · DOI: 10.18535/ijecs.v15i07.5572
0
Reads
0
Downloads
0
Citations
Abstract
People who inject drugs (PWIDs) in Kenya face HIV prevalence three to four times the general population rate, yet criminalization, stigma, and hard-to-reach sampling produce small, class-imbalanced datasets that are ethically difficult to share across institutional boundaries. This tension whereby the epidemiological need to predict risk meets the moral and legal need to protect subjects from re-identification, is a long-standing AI ethics problem that synthetic data is increasingly proposed to solve. Using a de-identified dataset of 6,142 PWID records from Nairobi and Coastal Kenya (17.8% HIV-positive), the study benchmarked eight augmentation strategies, comprising four deep generative methods (VAE, Tabular GAN, CTGAN, and PATE-GAN at ε = 1.0) and four classical baselines, across four classifiers, yielding 32 configurations. PATE-GAN paired with Random Forest achieved the highest precision-recall AUC (0.9165), matching or exceeding every non-private method, while providing a formal (ε = 1.0, δ = 10⁻⁵)-DP guarantee. Every method cleared a distance-to-closest-real privacy threshold. Formal DP carries near-zero utility cost in low-dimensional socio-behavioral tabular data. Remaining risks (group-level harms, surveillance creep, and consent boundaries for synthetic data derived from vulnerable populations) are discussed, and five governance-layer deployment commitments are proposed
Full text
Reader not loading? Open the PDF directly.
Cite this publication
Fedha Daniel Murunga, Betty Mayeku, Juma Kilwake, Mercy Nyakowa (2026). Privacy-Preserving Synthetic Data for HIV Risk Prediction in People Who Inject Drugs in Kenya: Benchmarking Differentially Private Generative Models. International Journal of Engineering and Computer Science, 15(07), 28694-28698. https://doi.org/10.18535/ijecs.v15i07.5572Download .ris (Zotero / Mendeley / EndNote)
Related publications
- From Numbers to Narratives: A Generative Artificial Intelligence Framework for Automated Risk Report Production in IT Program Portfolio Governance
International Journal of Engineering and Computer Science · 2026
- “Death may suck, but it’s better than the emergency room”, Lived experiences of healthcare avoidance and systemic harm among people experiencing homelessness: A meta-synthesis
Journal of Community Nursing and Primary Care · 2026
- Cognitive Supplier Risk Management Systems for Medical Device: AI Framework for Predictive Quality, Compliance, and Supply Chain Resilience
International Journal of Engineering and Computer Science · 2026
- LLM Safety as a Quality Gate: Integrating Bias, Fairness, and Robustness Evaluation into CI/CD via GitHub Apps
International Journal of Engineering and Computer Science · 2025
- AI-Assisted Tasawuf Education: Developing the AI-Murshid Framework
International Journal of Scientific Research and Management (IJSRM) · 2026