Gauging Ambient Environmental Carbon Dioxide Concentration Solely Using Biometric Observations: A Machine Learning Approach

Respiration is vital for human function. Inhaling specific gases can have specific physiological and cognitive impacts. Using a suite of sensors, we can collect detailed information on a range of both physiological and environmental factors. This study builds on previous research exploring how particulate matter affects physiological and cognitive responses, now expanded to include CO₂. We tracked the biometric variables of a cyclist, analyzing 329 specific variables. Simultaneously, an electric vehicle following the cyclist measured CO₂ and other environmental factors. After data collection, we used machine learning models to decipher the interactions between the human body and its surroundings. We found that biometric data alone could be used to accurately estimate the amount of CO₂ inhaled, achieving a good level of precision (r²=0.98) when comparing the estimated CO₂ based on biometrics and the actual observed CO₂ levels. In addition, we developed a ranking system to identify the biometric variables that most significantly predict environmental CO₂ inhalation.