Several machine learning (ML) techniques have demonstrated efficacy in precisely forecasting HIV risk and identifying the most eligible individuals for HIV testing in various countries. Nevertheless, there is a data gap on the utility of ML algorithms in strengthening HIV testing worldwide. This systematic review aimed to evaluate how effectively ML algorithms can enhance the efficiency and accuracy of HIV testing interventions and to identify key outcomes, successes, gaps, opportunities, and limitations in their implementation. This review was guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. A comprehensive literature search was conducted via PubMed, Google Scholar, Web of Science, Science Direct, Scopus, and Gale OneFile databases. Out of the 845 identified articles, 51 studies were eligible. More than 75% of the articles included in this review were conducted in the Americas and various parts of Sub-Saharan Africa, and a few were from Europe, Asia, and Australia. The most common algorithms applied were logistic regression, deep learning, support vector machine, random forest, extreme gradient booster, decision tree, and the least absolute shrinkage selection operator model. The findings demonstrate that ML techniques exhibit higher accuracy in predicting HIV risk/testing compared to traditional approaches. Machine learning models enhance early prediction of HIV transmission, facilitate viable testing strategies to improve the efficiency of testing services, and optimize resource allocation, ultimately leading to improved HIV testing. This review points to the positive impact of ML in enhancing early prediction of HIV spread, optimizing HIV testing approaches, improving efficiency, and eventually enhancing the accuracy of HIV diagnosis. We strongly recommend the integration of ML into HIV testing programs for efficient and accurate HIV testing.

Background: HIV testing is the cornerstone of HIV prevention and a pivotal step in realizing the Joint United Nations Program on HIV/AIDS (UNAIDS) goal of ending AIDS by 2030. Despite the availability of relevant survey data, there exists a research gap in using machine learning (ML) to analyze and predict HIV testing among adults in South Africa. Further investigation is needed to bridge this knowledge gap and inform evidence-based interventions to improve HIV testing. Objective: This study aims to determine consistent predictors of HIV testing by applying supervised ML algorithms in repeated adult population-based surveys in South Africa. Methods: A retrospective analysis of multiwave cross-sectional survey data will be conducted to determine the predictors of HIV testing among South African adults aged 18 years and older. A supervised ML technique will be applied across the five cycles of the South African National HIV Prevalence, Incidence, Behavior, and Communication Survey (SABSSM) surveys. The Human Science Research Council (HSRC) conducted the SABSSM surveys in 2002, 2005, 2008, 2012, and 2017. The available SABSSM datasets will be imported to RStudio (version 4.3.2; Posit Software, PBC) to clean and remove outliers. A chi-square test will be conducted to select important predictors of HIV testing. Each dataset will be split into 80% training and 20% test samples. Logistic regression, support vector machines, random forests, and decision trees will be used. A cross-validation technique will be used to divide the training sample into k-folds, including a validation set, and models will be trained on each fold. The models’ performance will be evaluated on the validation set using evaluation metrics such as accuracy, precision, recall, F1-score, area under curve-receiver operating characteristics, and confusion matrix. Results: The SABSSM datasets are open access datasets available on the HSRC database. Ethics approval for this study was obtained from the University of Johannesburg Research and Ethics Committee on April 23, 2024 (REC-2725-2024). The authors were given access to all five SABSSM datasets by the HSRC on August 20, 2024. The datasets were explored to identify the independent variables likely influencing HIV testing uptake. The findings of this study will determine consistent variables predicting HIV testing uptake among the South African adult population over the course of 20 years. Furthermore, this study will evaluate and compare the performance metrics of the 4 different ML algorithms, and the best model will be used to develop an HIV testing predictive model. Conclusions: This study will contribute to existing knowledge and deepen understanding of factors linked to HIV testing beyond traditional methods. Consequently, the findings would inform evidence-based policy recommendations that can guide policy makers to formulate more effective and targeted public health approaches toward strengthening HIV testing. International Registered Report Identifier (IRRID): DERR1-10.2196/59916

There is a significant portion of the South African population with unknown HIV status, which slows down epidemic control despite the progress made in HIV testing. Machine learning (ML) has been effective in identifying individuals at higher risk of HIV infection, for whom testing is strongly recommended. However, there are insufficient predictive models to inform targeted HIV testing interventions in South Africa. By harnessing the power of supervised ML (SML) algorithms, this study aimed to identify the most consistent predictors of HIV testing in repeated adult population-based surveys in South Africa. The study employed four SML algorithms, namely, decision trees, random forest, support vector machines (SVM), and logistic regression, across the five cross-sectional cycles of the South African National HIV Prevalence, Incidence, and Behavior and Communication Survey (SABSSM) datasets. The Human Science Research Council (HSRC) conducted the SABSSM surveys and made the datasets available for this study. Each dataset was split into 80% training and 20% testing sets with a 5-fold cross-validation technique. The random forest outperformed the other models across all five datasets with the highest accuracy (80.98%), precision (81.51%), F₁-score (80.30%), area under the curve (AUC) (88.31%), and cross-validation average (79.10%) in the 2002 data. Random forest achieved the highest classification performance across all the dates, especially in the 2017 survey. SVM had a high recall (89.12% in 2005, 86.28% in 2008) but lower precision, leading to a suboptimal F₁-score in the initial analysis. We applied a soft margin to the SVM to improve its classification robustness and generalization, but the accuracy and precision were still low in most surveys, increasing the chances of misclassifying individuals who tested for HIV. Logistic regression performed well in terms of accuracy = 72.75, precision = 73.64, and AUC = 81.41 in 2002, and the F₁-score = 73.83 in 2017, but its performance was somewhat lower than that of the random forest. Decision trees demonstrated moderate accuracy (73.80% in 2002) but were prone to overfitting. The topmost consistent predictors of HIV testing are knowledge of HIV testing sites, being a female, being a younger adult, having high socioeconomic status, and being well-informed about HIV through digital platforms. Random forest’s ability to analyze complex datasets makes it a valuable tool for informing data-driven policy initiatives, such as raising awareness, engaging the media, improving employment outcomes, enhancing accessibility, and targeting high-risk individuals. By addressing the identified gaps in the existing healthcare framework, South Africa can enhance the efficacy of HIV testing and progress towards achieving the UNAIDS 2030 goal of eradicating AIDS.

Background: The global targets for HIV testing for achieving the Joint United Nations Programme on HIV/AIDS (UNAIDS) 95-95-95 targets are still short. Identifying gaps and opportunities for HIV testing uptake is crucial in fast-tracking the second (initiate people living with HIV on antiretroviral therapy) and third (viral suppression) UNAIDS goals. Machine learning and health technologies can precisely predict high-risk individuals and facilitate more effective and efficient HIV testing methods. Despite this advancement, there exists a research gap regarding the extent to which such technologies are integrated into HIV testing strategies worldwide. Objective: The study aimed to examine the characteristics, citation patterns, and contents of published studies applying machine learning and emerging health technologies in HIV testing from 2000 to 2024. Methods: This bibliometric analysis identified relevant studies using machine learning and emerging health technologies in HIV testing from the Web of Science database using synonymous keywords. The Bibliometrix R package was used to analyze the characteristics, citation patterns, and contents of 266 articles. The VOSviewer software was used to conduct network visualization. The analysis focused on the yearly growth rate, citation analysis, keywords, institutions, countries, authorship, and collaboration patterns. Key themes and topics were driven by the authors’ most frequent keywords, which aided the content analysis. Results: The analysis revealed a scientific annual growth rate of 15.68%, with an international coauthorship of 8.22% and an average citation count of 17.47 per document. The most relevant sources were from high-impact journals such as the Journal of Internet Medicine Research, JMIR mHealth and uHealth, JMIR Research Protocols, mHealth, AIDS Care-Psychological and Socio-Medical Aspects of AI, and BMC Public Health, and PLOS One. The United States of America, China, South Africa, the United Kingdom, and Australia produced the highest number of contributions. Collaboration analysis showed significant networks among universities in high-income countries, including the University of North Carolina, Emory University, the University of Michigan, San Diego State University, the University of Pennsylvania, and the London School of Hygiene and Tropical Medicine. The discrepancy highlights missed opportunities in strategic partnerships between high-income and low-income countries. The results further demonstrate that machine learning and health technologies enhance the effective and efficient implementation of innovative HIV testing methods, including HIV self-testing among priority populations. Conclusions: This study identifies trends and hotspots of machine learning and health technology research in relation to HIV testing across various countries, institutions, journals, and authors. The trends are higher in high-income countries with a greater focus on technology applications for HIV self-testing among young people and priority populations. These insights will inform future researchers about the dynamics of research outputs and help them make scholarly decisions to address research gaps in this field.

Background: Testing for HIV using rapid test devices assists in determining HIV status and ascertains if treatment is required. Rapid HIV quality assurance (QA) training is conducted to ensure accuracy of test results in non-laboratory settings. Objectives: This study aimed to determine the factors associated with the accuracy of rapid HIV testing among primary healthcare (PHC) workers who received HIV testing QA training in the Eastern Cape, South Africa. Method: A cross-sectional analytical study design was used to purposively recruit 167 PHC workers with prior training on HIV testing QA. Data were collected in 2022 using a self-administered questionnaire. STATA version 17.0 was used for data analysis. Results: Tester accuracy measured by a proficiency testing (PT) score of greater than 80% was achieved among 64.7% of the testers. Comprehensive HIV QA training was significantly associated with a PT score of greater than 80% (P = 0.001). On multivariable analyses, a PT score of greater than 80% was less likely with rural facilities (adjusted odd ratios [aOR] = 0.56, 95% confidence interval [CI]: 0.36–0.92, P = 0.020), Grade 12 education (aOR = 0.40, 95% CI: 0.19–0.85, P = 0.017), mentorship (aOR = 0.55, 95% CI: 0.35–0.85, P = 0.008), and attitudes towards inaccurate HIV results (aOR = 0.13, 95% CI: 0.02–0.82, P = 0.03); while implementation of QA processes (aOR = 3.94, 95% CI: 1.22–12.74, P = 0.022) and elements of QA in the HIV Testing Services register (aOR = 4.93, 95% CI: 1.45–16.74, P = 0.011) were associated with a PT score of greater than 80%. Conclusion: Exposure to comprehensive rapid HIV QA training were associated with higher tester accuracy. A framework for QA training is required for standardisation of training in the country.

Background: In South Africa, there is no centralized HIV surveillance system where key populations (KPs) data, including gay men and other men who have sex with men, female sex workers, transgender persons, people who use drugs, and incarcerated persons, are stored in South Africa despite being on higher risk of HIV acquisition and transmission than the general population. Data on KPs are being collected on a smaller scale by numerous stakeholders and managed in silos. There exists an opportunity to harness a variety of data, such as empirical, contextual, observational, and programmatic data, for evaluating the potential impact of HIV responses among KPs in South Africa. Objective: This study aimed to leverage and harness big heterogeneous data on HIV among KPs and harmonize and analyze it to inform a targeted HIV response for greater impact in Sub-Saharan Africa. Methods: The Boloka data repository initiative has 5 stages. There will be engagement of a wide range of stakeholders to facilitate the acquisition of data (stage 1). Through these engagements, different data types will be collated (stage 2). The data will be filtered and screened to enable high-quality analyses (stage 3). The collated data will be stored in the Boloka data repository (stage 4). The Boloka data repository will be made accessible to stakeholders and authorized users (stage 5). Results: The protocol was funded by the South African Medical Research Council following external peer reviews (December 2022). The study received initial ethics approval (May 2022), renewal (June 2023), and amendment (July 2024) from the University of Johannesburg (UJ) Research Ethics Committee. The research team has been recruited, onboarded, and received non–web-based internet ethics training (January 2023). A list of current and potential data partners has been compiled (January 2023 to date). Data sharing or user agreements have been signed with several data partners (August 2023 to date). Survey and routine data have been and are being secured (January 5, 2023). In (September 2024) we received Ghana Men Study data. The data transfer agreement between the Pan African Centre for Epidemics Research and the Perinatal HIV Research Unit was finalized (October 2024), and we are anticipating receiving data by (December 2024). In total, 7 abstracts are underway, with 1 abstract completed the analysis and expected to submit the full article to the peer-reviewed journal in early January 2024. As of March 2025, we expect to submit the remaining 6 full articles. Conclusions: A truly “complete” data infrastructure that systematically and rigorously integrates diverse data for KPs will not only improve our understanding of local epidemics but will also improve HIV interventions and policies. Furthermore, it will inform future research directions and become an incredible institutional mechanism for epidemiological and public health training in South Africa and Sub-Saharan Africa. International Registered Report Identifier (IRRID): DERR1-10.2196/63583