CONTACTLESS HEART RATE MONITORING BASED ON SMARTPHONE APPLICATION

• Main Author: Mutmainnah Rahim, Nur
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/86487
• Institution: Institut Teknologi Bandung

Heart disease is the leading cause of death globally, predominantly affecting the elderly, often due to inadequate early monitoring. This study aims to design a heart rate monitoring device targeted at young individuals to prevent future heart conditions. This device is developed as a smartphone application using selfie videos, a method popular among the youth. The application employs an RGB camera and the remote Photoplethysmography (rPPG) method based on the Plane- Orthogonal-to-Skin (POS) technique, implemented using Java and Python programming languages, integrated through the Chaquopy framework. Testing was conducted to evaluate sensitivity to lighting conditions, the impact of variability, and processing duration on 10 subjects, representing the primary challenge of non-contact measurements. Evaluation and analysis through ANOVA statistical tests and p-value indicated that for lighting conditions, the p-value exceeded 0.05, suggesting no significant differences between lighting conditions. However, the application demonstrated significant differences in detecting variability before and after physical activity, with a p-value as low as 0.008, indicating highly significant differences. Additionally, processing duration was measured using two algorithms, ML Kit and GMS Vision, with ML Kit showing significantly enhanced performance, reducing the time from 1.5 hours to just 15 minutes.