Windows Into Human Health Through Wearables Data Analytics.

Area of Science:

• Biomedical Engineering
• Digital Health
• Algorithm Development

Background:

• Wearable sensors are increasingly integrated into commercial products, collecting raw physiological data.
• This data is being investigated across diverse clinical domains and patient populations.
• Research focuses on algorithm development to translate sensor data into actionable health information.

Purpose of the Study:

• To review key fitness and health indicators derived from wearable sensor data.
• To describe various algorithmic approaches used to generate these indicators.
• To highlight advances in wearable technology for health monitoring.

Main Methods:

• Systematic search of the PubMed database using terms for popular wearable devices and sensor types (e.g., Fitbit, Apple Watch, accelerometry, ECG, PPG, continuous glucose monitoring).
• Focus on algorithms from 2014-2017 dominating the commercial market and highly represented in biomedical literature.
• Summarized studies on commercially available device algorithms and pivotal sensor algorithm development.

Main Results:

• Algorithms are crucial for processing raw sensor data into meaningful health indicators.
• Key areas include fitness metrics (activity, energy expenditure) and health conditions (arrhythmia, diabetes management).
• Advances in algorithms enhance the analytical capabilities of wearable devices.

Conclusions:

• Wearable health technologies convert physiological data into digital health information.
• The utility of wearables heavily relies on signal processing and algorithms.
• Ongoing algorithmic and machine learning advancements will expand the capabilities of wearable health indicators.