Predicting Heartbeats in Real-Time: A Continuous Monitoring Approach with Face Recognition Algorithm

Healthcare practitioners heavily depend on Electrocardiogram (ECG) machines to acquire critical insights into heart function through waveform analysis. The information can be used for monitoring and understanding overall heart health. However, the onset of the Covid-19 pandemic prompted global governments to implement stringent guidelines, emphasizing touchless healthcare methods. To overcome this problem, an innovative and reliable method for continuous heart rate estimation without physical contact has been introduced. The approach leverages face recognition algorithms to predict real-time heart pulse rates based on color intensity measurements. To accommodate inputs from videos or images, a segmentation algorithm is integrated. Rigorous experiments involving volunteers validate the results, revealing relative errors below 6%, a margin widely accepted in the field. The work concludes with a comparative analysis, assessing the proposed approach against existing literature. The system excels in providing a touchless and efficient approach to the crucial task.