What does this research mean for the field?

While AI-driven IoT systems show promise in predictive accuracy and lower response latency for disease detection and monitoring, significant heterogeneity in methodologies, reliance on small datasets, and implementation challenges such as data privacy and interoperability currently restrict their mass clinical implementation. Novelty: ClaimNovelty.SYNTHESIS. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This review aims to analyze AI-enhanced IoT systems for real-time disease prediction and patient monitoring.

June 27, 2026Open Access

Artificial Intelligence assisted Internet of Things systems for real time health condition prediction a systematic review

Key Points

This review aims to analyze AI-enhanced IoT systems for real-time disease prediction and patient monitoring.
Conducted a systematic review of recent literature across major scientific databases.
Selection based on clearly defined inclusion and exclusion criteria for relevance and transparency.
Evaluation of machine learning and deep learning models in IoT systems for predictive accuracy.
Identified promising predictive accuracy and low response latency in selected AI-IoT studies.
Noted significant heterogeneity in datasets and methodologies limiting direct comparisons.
Highlighted ongoing issues with data privacy, cybersecurity, and regulatory compliance affecting clinical implementation.

Abstract

The combination of Artificial Intelligence (AI) and Internet of Things (IoT) has contributed to the development of real-time health monitoring and predictive healthcare systems to a significant level. This literature review critically evaluates the recent AI-aided IoT platforms that have been developed in the detection of diseases, early diagnosis, and ongoing remote monitoring of patients. The search process was organized in a systematic search of major scientific databases with pre-defined keywords and the selection of studies was conducted according to clear inclusion and exclusion criteria to provide transparency and relevance of the methods. As shown by the reviewed literature, machine learning and deep learning models used in this context of IoT-enabled sensor architecture are promising in terms of predictive accuracy and lower response latency. Nevertheless, there is a significant level of heterogeneity in datasets, preprocessing algorithms, validation methods, and metrics, which restricts the direct comparison of the reported results of accuracy. Most research works use small or field-specific data sets, and this could limit the ability to generalize to wider and more diverse clinical groups. Along with technical variability, a number of implementation issues have become permanent, such as issues with data privacy, cybersecurity risks, algorithmic bias, interoperability constraints, and regulatory compliance needs. These aspects are serious obstacles to mass clinical implementation. Despite the above frameworks indicating the prospect of AI-driven IoT systems to improve the process of early detection, enhance the ability to manage chronic diseases, and facilitate remote healthcare delivery, additional real-life validation is necessary. In the future, standardized benchmarking protocols, large-scale multicenter clinical studies, explainable AI integration, secure data governance frameworks, and pathways aligned with regulatory pathways should be emphasized to ensure safe, ethical, and scalable use of AI in healthcare.

Mark Helpful

Bookmark

Relay

View Full Paper