Alcohol Impairment Detection Through Heart Rate Variability Analysis Using Gyrocarotidography
Excessive alcohol consumption can impair decision-making and balance, putting workers and those around them at risk, especially in safety-sensitive occupations. Continuous monitoring of employees may help prevent workplace accidents. With advancements in wearable technologies such as smartwatches, headphones, and earbuds, there is a growing trend in real-time monitoring using wearable devices. To this end, this study introduces a carotid-mounted wearable system designed to assess heart rate variability (HRV) using an Inertial Measurement Unit (IMU) sensor for alcohol-induced impairment detection. The proposed system utilizes a novel signal modality called Gyrocarotidography (GCG)−a form of carotid ballistography (CBG) −to collect gyroscope-based signals and extract HRV features to analyze the effects of alcohol consumption while validating the results against a reference electrocardiogram (ECG) device. Statistically significant HRV features are then used to classify participants as impaired and non-impaired. The findings demonstrate a strong correlation between HRV features extracted from the IMU sensor and the ECG reference, confirming the feasibility of IMU-based HRV monitoring. Significant differences in HRV metrics before and after alcohol consumption further support the potential of this method, with the classification model achieving 83% accuracy. The proposed carotid-mounted IMU system offers a non-invasive, wearable solution for real-time HRV monitoring and alcohol impairment detection, and it provides clinicians and occupational health professionals with a tool for assessing alcohol-induced autonomic changes through HRV− offering a practical pathway for impairment detection and workplace safety enhancement.
Authors
- Edward J. Park (ORCID: https://orcid.org/0000-0003-0035-061X)
- Saboora M. Roshan (ORCID: https://orcid.org/0000-0001-7978-1310)
Institutions
- Simon Fraser University (CA)
Publication Details
- Journal
- IEEE Access
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1109/access.2025.3630346
- Primary Topic
- Non-Invasive Vital Sign Monitoring
- Type
- article
- Field-Weighted Citation Impact
- 0.00