Emerging AIoT-Based Approaches for Early Detection of Cardiac Diseases: A Review

Authors

• Balqees Talal Hasan, Department of Artificial Intelligence, College of Information Technology, Ninevah University, Mosul, IraqFollow
• Zaid J. Al-Araji, Department of Computer Networks and Internet, College of Information Technology, Ninevah University, Mosul, IraqFollow
• Zainab AL-Badrany, Department of Computer Networks and Internet, College of Information Technology, Ninevah University, Mosul, IraqFollow
• Sarah Kanaan Hamzah, Department of Unmanned Aircraft Systems Engineering, College of Engineering, Alnoor University, Nineveh, IraqFollow
• Abdulmajeed Sulaiman, Department of Computer Networks and Internet, College of Information Technology, Ninevah University, Mosul, IraqFollow

Abstract

Cardiovascular disease continues to pose a serious global health burden, emphasizing the need for smart, ongoing, and clinically trustworthy systems for early detection of this condition. The convergence of IoT-enabled sensing technologies with AI has led to the rise of an innovative paradigm known as AIoT, integrating technologies such as machine learning, deep learning, edge and fog computing, and cloud-based analytics with the aim of real-time cardiac monitoring and diagnostic decision support. Despite rapid expansion, current literature appears disparate in its coverage of sensing modalities, learning algorithms, system architectures, data sets, evaluation criteria, and clinical applications. To remedy this, a systematic review of recent literature covering AIoT-based cardiac disease detection was performed, primarily focused on literature from 2020–2025 and also including pertinent works from 2026 as it was being finalized. All reviewed studies were classified based on the AI approach utilized, the IoT sensing method used, the kind of cardiovascular disease targeted, the platform it was deployed on, the data set, the performance measurement, the overall system structure, the advantages of the system, and the disadvantages of the system. Results reveal that ECG-based wearable devices are the prevalent choice of hardware in most existing AIoT-based cardiac monitoring systems, and that classification algorithms using machine learning classifiers and both CNN and recurrent networks, hybrid models as well as edge/fog assisted AIoT architectures are dominant choices of the algorithm and system architecture, respectively. AIoT systems also facilitate the detection of various conditions including arrhythmias, myocardial infarction, monitoring of heart failure and the realization of personalized risk prediction and remote patient monitoring services. Nonetheless, data heterogeneity, difficulties with noise and artifact handling, inadequate clinical validation of methods, under-performance analysis in real-time applications, interoperability limitations, privacy and security concerns, and interpretability limitations are significant shortcomings of present works. This research offers a comprehensive taxonomy and comparative summary of current research trends in the domain of AIoT-based cardiac disease detection systems by highlighting the methodological gaps and providing recommendations for designing reliable, interpretable, and clinically relevant AIoT-based cardiac monitoring and healthcare solutions.

Recommended Citation

Hasan, Balqees Talal; Al-Araji, Zaid J.; AL-Badrany, Zainab; Hamzah, Sarah Kanaan; and Sulaiman, Abdulmajeed (2026) "Emerging AIoT-Based Approaches for Early Detection of Cardiac Diseases: A Review," Mesopotamian Journal of Artificial Intelligence in Healthcare: Vol. 4: Iss. 1, Article 3.