Fair non-contact blood pressure estimation using imaging photoplethysmography

Application of Real-Time Palm Imaging with Nelder-Mead Particle Swarm Optimization/Regression Algorithms for Non-Contact Blood Pressure Detection

In response to the rising prevalence of hypertension due to lifestyle changes, this study introduces a novel approach for non-contact blood pressure (BP) monitoring. Recognizing the "silent killer" nature of hypertension, this research focuses on developing accessible, non-invasive BP measurement methods. This study compares two distinct non-contact BP measurement approaches: one combining the Nelder-Mead simplex method with particle swarm optimization (NM-PSO) and the other using machine learning regression analysis. In the NM-PSO method, a standard webcam captures continuous images of the palm, extracting physiological data through light wave reflection and employing independent component analysis (ICA) to remove noise artifacts. The NM-PSO achieves a verified root mean square error (RMSE) of 2.71 mmHg for systolic blood pressure (SBP) and 3.42 mmHg for diastolic blood pressure (DBP). Alternatively, the regression method derives BP values through machine learning-based regression formulas, resulting in an RMSE of 2.88 mmHg for SBP and 2.60 mmHg for DBP. Both methods enable fast, accurate, and convenient BP measurement within 10 s, suitable for home use. This study demonstrates a cost-effective solution for non-contact BP monitoring and highlights each method's advantages. The NM-PSO approach emphasizes optimization in noise handling, while the regression method leverages formulaic efficiency in BP estimation. These results offer a biomimetic approach that could replace traditional contact-based BP measurement devices, contributing to enhanced accessibility in hypertension management.

Fair non-contact blood pressure estimation using imaging photoplethysmography

Hypertension is typically manifested as a latent symptom that requires detection through specialized equipment. This poses an inconvenience for individuals who need to undergo long-term blood pressure monitoring in their daily lives. Therefore, there is a need for a portable, non-contact method for estimating blood pressure. However, current non-contact blood pressure estimation methods often rely on relatively narrow datasets, lacking a broad range of blood pressure distributions. Additionally, their applicability is confined to controlled experimental environments. This study proposes a non-contact blood pressure estimation method suitable for various life scenarios, encompassing multiple age groups, diverse ethnicities, and individuals with different skin tones. The aim is to enhance the practicality and accuracy of existing non-contact blood pressure estimation methods. The research extracts the imaging photoplethysmogram (IPPG) signal from facial videos and processes the signal through four layers of filtering operations to obtain an IPPG signal reflecting pulse wave variations. A CNN+BiLSTM+GRU network structure is constructed to improve the accuracy of current non-contact blood pressure estimation methods. In comparison to existing approaches, the mean absolute error (MAE) for systolic blood pressure (SBP) and diastolic blood pressure (DBP) is reduced by 13.6% and 16.4%, respectively.