A Sensor Platform for Athletes' Training Supervision: A Proof of Concept Study.

Area of Science:

• Sports Science
• Biomedical Engineering
• Physiological Monitoring

Background:

• Professional athletes require real-time, non-invasive monitoring for tailored training strategies.
• Current methods may be invasive or lack real-time data crucial for performance enhancement.
• Identifying the anaerobic threshold is key for optimizing training intensity.

Purpose of the Study:

• To present a novel sensor system for real-time, non-invasive monitoring of athletes' physiological parameters and performance during training.
• To identify the anaerobic threshold by measuring lactate concentration and respiratory parameters.
• To develop a predictive model for key physiological metrics.

Main Methods:

• A sensor system was developed to measure lactate concentration from saliva samples.
• Respiratory parameters (maximal oxygen consumption, maximal carbon dioxide production, respiratory rate) were estimated from exhaled breath.
• A supervised learning model, Partial Least Squares (PLS), was used to predict physiological parameters.

Main Results:

• The sensor non-invasively measures lactate levels and estimates key respiratory metrics.
• A PLS model was successfully built to predict physiological parameters of interest.
• The system demonstrated low power consumption and wireless communication capabilities.

Conclusions:

• The developed sensor offers a non-invasive, real-time solution for monitoring athletes' physiological status and performance.
• This technology facilitates personalized training by providing actionable data on lactate and respiratory function.
• The system's advantages, including non-invasive sampling and wireless communication, support its application in actual sports settings.