Although contact-based vital sign detection is pervasive in healthcare settings, it can cause discomfort and lacks long-term monitoring capabilities, which is imperative to detect disease. Non-invasive methods of radio-emitting vital sign detection have emerged to enhance patient health insight; however, due to their long wavelengths, radio waves are vulnerable to reflections from surroundings, leading to inaccurate vital sign readings. To address this issue, I propose a contactless, high-resolution vital sign detection system based on light detection and ranging (LiDAR). LiDAR emits light with shorter wavelengths and superior resolution than radio, resulting in more accurate sensing. My proposed system combines spatial cleaning algorithms and temporal signal processing to analyze continuous 3D LiDAR point clouds and isolate minute chest displacements to extract accurate respiratory rates. I evaluate the system's predicted signal against an ground-truth force respiration belt using both time and frequency domain analysis, and provide a publicly available LiDAR-synchronized vital sign dataset for future work.
