The growing availability of electronic health records (EHRs) has accelerated the use of artificial intelligence (AI) and machine learning (ML) in public health. Yet, how well these methods work in low- and middle-income countries (LMICs), remains poorly understood. This review synthesises studies on ML-based prediction using EHR or routinely collected electronic clinical data, with direct LMIC evidence analysed alongside high-income country (HIC) studies, included as methodological benchmarks. Following PRISMA guidelines, searches across five major databases identified 64 eligible studies published between Jan 2018–Mar 2025. Of these, 12 (18.8%) were conducted exclusively in LMIC settings, 44 (68.8%) in HICs, and 8 (12.5%) drew on mixed or multi-setting data. Retrospective designs predominated (81.3%). Disease progression (40.6%), mortality (34.4%), and treatment response (25.0%) were the most common prediction targets. Deep learning architecture was the most frequently applied category overall (39.1%, n = 25), driven by HIC studies with access to large curated datasets; among LMIC-focused studies, traditional ML and ensemble methods were each applied in 33.3% of studies. Evaluation practices were dominated by discrimination metrics, particularly AUROC; external validation was reported in only 5 studies (7.8%) and calibration in only 4 (6.2%). Explainability assessment was reported in 1 of 12 LMIC studies (8.3%) compared with 16 of 44 HIC studies (36.4%), with governance and ethical considerations inconsistently documented in LMIC settings. This review highlights key methodological and contextual gaps and offers guidance for developing interpretable, reliable, and context-appropriate AI tools for public health decision-making in LMIC settings.
Joe Phiri, Aaron Zimba, C. Njovu et al.· Discover Artificial Intellig...· 0 citations
Urban traffic congestion is a major challenge affecting sustainability, energy efficiency, and mobility in smart cities. The emergence of 5G/6G and beyond networks, combined with intelligent wireless communication and sensing, offers unprecedented opportunities for real-time traffic monitoring, AI-driven congestion modeling, and dynamic routing optimization. Despite these advancements, existing 5G/6G-enabled intelligent transportation systems studies remain fragmented with limited integration between wireless communication, sensing, vehicular traffic modeling, which reduces their effectiveness in addressing real-world urban traffic congestion challenges. A systematic literature review was conducted using the PRISMA framework, synthesizing 49 peer-reviewed studies published between 2019 and 2026 to identify key trends, methodologies and applications that leverage 5G/6G-enabled sensing and communication for vehicular congestion reduction. However, current review studies primarily focus on isolated aspects such as communication architectures, AI techniques, or vehicular networking frameworks, and lack a comprehensive and systematic synthesis that examines intelligent wireless communication and sensing for sustainable traffic modeling. Findings indicate that existing studies widely explore the integration of 5G/6G-enabled vehicular communication, IoT sensing, AI-driven traffic prediction, and edge intelligence for real-time congestion management. Nevertheless, full integration across communication, sensing, and traffic modeling remains limited, with privacy, scalability, cost, and interoperability continuing to pose major challenges. Integration of 5G/6G, IoT sensing, and AI traffic modeling enables real-time congestion management and efficient routing, directly supporting sustainability goals such as reduced energy use, lower emissions, and improved urban mobility. To the best of our knowledge, this is the first PRISMA-guided systematic review that provides a unified cross-layer synthesis of intelligent wireless communication, sensing technologies, and AI-driven traffic modeling specifically for sustainable vehicular congestion reduction in 5G/6G-enabled ITS. By examining 5G/6G-enabled ITS, this study complements existing surveys through a unified cross-layer synthesis of communication, sensing, traffic modeling, and sustainability aspects, while providing practical insights for developing efficient and AI-driven urban traffic systems. Furthermore, the findings support policy decisions that prioritize 5G/6G infrastructure investment, edge computing deployment and interoperable ITS standards to enable scalable real-world implementation.
Benjamin Simwinga, Aaron Zimba, Mayumbo Nyirenda· Discover Networks· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
