Short Time Blood Pressure Estimation Using RGB-NIR Camera

This study proposes a non-contact blood pressure estimation method using facial video images captured by an RGB-NIR camera. Conventional blood pressure measurement methods often face challenges such as invasive techniques, discomfort caused by cuff inflation, and decreased performance in low-light environments. This study aims to address these limitations. By analyzing pulse wave signals extracted from facial videos, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were estimated using a Random Forest algorithm. To ensure generalizability, the experiment included subjects from various age groups. Additionally, short-time estimation accuracy was evaluated to enhance usability in daily environments. As a result, this method demonstrated the potential for continuous and non-invasive blood pressure monitoring in daily life.Clinical Relevance-The non-contact blood pressure estimation method using RGB-NIR cameras is a technology that reduces patient burden while enabling continuous blood pressure monitoring in low light conditions and daily life settings. This method is useful for regular blood pressure measurement and remote medical care.