Blood pressure monitor measured by light: Blood pressure estimation method using photoplethysmography (PPG).

Estimation of blood pressure with minimal error using photoplethysmography (PPG). It can be applied to blood pressure monitoring with wearable devices!

We introduce a method for estimating blood pressure using photoplethysmographic (PPG) signals and a computer program for blood pressure estimation. In conventional methods for estimating blood pressure using PPG, there have been significant estimation errors due to individual differences and issues related to the black-box nature of nonlinear models, which has prevented widespread adoption. This technology alleviates estimation errors caused by individual differences by introducing feature normalization and bias adjustment for blood pressure, resulting in the generation of a generalized model. To address the black-box issue of nonlinear models, we utilized a multiple regression model to facilitate the interpretation of the blood pressure estimation model. As a result, depending on the training data, it has become possible to construct a generalized blood pressure estimation model with an estimation error of less than 10 mmHg. [Keywords] Health management / Wearable devices / Medical devices / Machine learning / Work style reform *For more details, please refer to the PDF document or feel free to contact us.

others