Analysis of human physiology through wearable devices for the identification of alcohol intoxication

The increasing adoption of wearable devices in everyday life offers new possibilities for health monitoring, including the detection of alcohol intoxication. This technology holds significant potential in preventing alcohol-related incidents, particularly in reducing the risk of drunk driving and enhancing workplace safety. Photoplethysmography (PPG), commonly used in wearable devices, presents a promising approach for detecting alcohol-induces physiological changes in a practical, non-intrusive and continuous manner. The goal of this thesis is to introduce a novel method for monitoring alcohol intoxication thorough PPG, with a particular focus on frequency domain analysis by monitoring changes in the signal, recognizing that these changes can vary across different individuals. The study involved over 15 subjects, all wearing a Polar OH1+ sensor on the upper arm to record PPG. By monitoring the different changes throughout the experiments, a tailored algorithm was developed to detect alcohol intoxication. The performance of this approach was evaluated though controlled experiments, with results demonstrating the algorithm’s ability to reliably detect alcohol-induced changes across different subjects. The findings contribute to a broader goal of integrating alcohol monitoring into wearable devices, potentially enabling real-time detection for professionals in safety-sensitive roles, or as a preventive tool in everyday civilian use. This research lays the groundwork for future advancements in alcohol detection technologies within consumer decives like smartwatches, offering a practical and non-invasibe solution for promoting safety in both personal and professional contexts.