Machine Learning Applied to Edge Computing and Wearable Devices for Healthcare: Systematic Mapping of the Literature.

Area of Science:

• Computer Science
• Biomedical Engineering
• Artificial Intelligence

Background:

• The convergence of machine learning (ML), edge computing, and wearable devices is a rapidly evolving area in healthcare.
• This field leverages real-time data processing at the network edge to enhance medical applications.

Purpose of the Study:

• To systematically map and analyze the literature on ML, edge computing, and wearable devices in healthcare.
• To identify key concepts, techniques, architectures, and emerging trends in this interdisciplinary field.

Main Methods:

• Systematic literature review of 171 studies.
• Rigorous selection process focusing on 28 key articles for in-depth analysis.
• Analysis of research trends, common ML models, edge platforms, and healthcare applications.

Main Results:

• Significant research increase in the last six years, particularly the last three.
• Dominant applications include fall detection, cardiovascular monitoring, and disease prediction.
• Prevalence of neural network models like Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs).
• Diverse edge computing platforms utilized, including Raspberry Pi and smartphones.

Conclusions:

• The field is nascent with substantial opportunities for future research and development.
• Need for standardized architectures and further exploration of hardware/software for ML-driven healthcare.
• Identified research directions to foster continued innovation in healthcare technologies.