A System for Objectively Measuring Behavior and the Environment to Support Large-Scale Studies on Childhood Obesity.

Area of Science:

• Digital Health
• Public Health Technology
• Behavioral Science

Background:

• Internet of Things (IoT) advancements enable detailed human behavior quantification.
• High-rate, multi-source data streams offer insights into behavior-environment interactions.
• Tackling obesity requires innovative approaches to understand behavioral and environmental factors.

Purpose of the Study:

• To present an integrated system for collecting and extracting behavioral and environmental indicators.
• To improve public health policies for obesity prevention and management.
• To detail the design, implementation, and evaluation of algorithms for behavioral data analysis.

Main Methods:

• Passive data collection via smartphone and smartwatch applications.
• Integration of multiple technologies, algorithms, and components into a unified system.
• Development and evaluation of algorithms for processing large-scale (Big Data) behavioral and environmental data.

Main Results:

• Accurate step counting with an absolute error of 8-9 steps.
• Effective detection of visited locations with an F1-score of 0.86.
• Precise estimation of gross sleep time with an error of less than 12 minutes.

Conclusions:

• The developed system demonstrates significant potential for public health authorities and researchers.
• The system provides a valuable tool for understanding human behavior and its environmental influences.
• The findings support the use of integrated IoT systems for data-driven public health initiatives.