Contactless Vital Signs Monitoring Analysis

Problem

Conventional vital signs measurement often relies on wristbands, patches, or fingertip sensors, which add wear burden, maintenance, and infection risk, especially in home care and elderly monitoring. A camera-based approach is harder because vital signs are subtle physiological signals: skin color changes, micro-movement, and respiration-related cues are easily disturbed by lighting, pose, and video quality.

How It Works

The skill invokes a cloud API through scripts/contactless_vital_signs_monitoring.py using a local video path or public --url. It supports heart rate, respiration rate, SpO2 estimation, and HRV analysis, extracting hemodynamic cues from subtle facial skin color changes using photoplethysmography-style processing. A typical workflow is: capture a 10-30 second video at 30-50 cm with even lighting; obtain and pass a valid --open-id; run the analysis; and receive a JSON/basic report or a cloud-hosted report list. Historical report lookup is expected to query the cloud interface and render a Markdown table with report name, duration, analysis time, heart rate, and a link to the full report.

Boundaries

Accuracy depends on face orientation, adequate uniform lighting, no mask over the mouth/nose, and acceptable video length and quality. Supported formats include mp4/avi/mov, with a recommended maximum size of 10MB. The output is for health reference only and cannot replace clinical measurement or medical diagnosis; seek timely care if abnormal symptoms occur.