ML-based Blood Pressure Estimation using Converted PPG Signal from Video

Conventional cuff-based methods for measuring blood pressure is effective, but its cumbersome and uncomfortable. Earlier Study acquired data via wearable devices or smartphone sensors. This study investigates alternative techniques for measuring blood pressure without the need of cuffs, by developing technologies such as photoplethysmography (PPG) and machine learning. This research endeavors to develop precise and convenient methods for estimating blood pressure without the need for a cuff, by analyzing PPG data acquired from smartphone This study presents a thorough examination of BP through smartphone camera measuring methods, including their accuracy, and machine learning algorithm. The suggested method utilizes a multivariate regression model to estimate SBP and DBP. The model was trained and tested on 56 healthy people. We assessed the suggested technique by comparing the blood pressure (BP) readings obtained from the proposed method with the reference device, using the R-squared (R 2 ) value of the cross-validation as a measure of comparison. The experimental results demonstrate that the approach proposed calculates estimates the R 2 for SBP is 0.6899 and for DBP is 0.6894