A Systematic Review of Machine Learning and IoT Applied to the Prediction and Monitoring of Cardiovascular Diseases

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Biomedical Engineering

Background:

• Cardiovascular disease (CVD) is a leading global cause of mortality.
• Early detection and continuous monitoring are crucial for managing CVD.
• Emerging technologies like IoT, IoMT, and ML offer potential solutions.

Purpose of the Study:

• To systematically review the application of IoT, IoMT, and ML in CVD detection, prediction, and monitoring.
• To identify prevalent technologies, machine learning algorithms, and datasets used in CVD research.
• To highlight the most frequently studied cardiovascular diseases and current research challenges.

Main Methods:

• Systematic review of 164 high-impact journal papers.
• Categorization of studies into IoT/IoMT for CVD detection (82 papers) and ML for CVD prediction (85 papers).
• Analysis of identified technologies, algorithms, datasets, and disease focus.

Main Results:

• Neural networks demonstrated high accuracy (>90%), followed by Random Forest, XGBoost, k-NN, and SVM.
• IoT/IoMT technologies enable real-time CVD prediction.
• Ensemble techniques showed excellent performance in accuracy metrics.
• Hypertension and arrhythmia were the most frequently studied CVDs.

Conclusions:

• IoT/IoMT technologies are effective for real-time cardiovascular disease prediction.
• Machine learning, particularly ensemble methods, achieves high accuracy in CVD analysis.
• Lack of publicly available datasets is a significant barrier for advancing ML in CVD prediction.