Abstract
Non-contact neck imaging using a standard smartphone camera and ambient lighting with accurate estimates of heart rate (HR) and respiratory rate (RR) provides superior access to care in resource constrained health care systems. While most existing techniques for non-contact imaging focus on detecting HR from facial regions, limited work has explored HR and RR detection from the neck. Prior studies involve healthy volunteers while relying on specialized cameras or controlled lighting conditions. Data acquired for the study were collected in a cardiology clinic under regular fluorescent lighting, with the participants being assessed for cardiac pathology. The proposed approach uses 15 to 30-second videos to analyze the skin displacement motion of neck pixels. The results demonstrate superior performance with R-values of 0.98 and 0.85 for HR and RR respectively compared to existing imaging photoplethysmography (iPPG) algorithms, highlighting the robustness of this method. The outcomes of this work could facilitate the detection of carotid artery and jugular venous pulsation, providing a more comprehensive assessment of cardiovascular and respiratory health, particularly for patient monitoring.