PPG Signal Detection Before and After Human Movement Based on ResNet

As a physiological signal of the human body, the pulse wave contains a large amount of cardiovascular information of the human body. Each inflection point of its waveform has a clear physiological meaning and represents the transformation of each mechanical process in the cardiac cycle. When the human body performs load exercise, due to the increase in cellular oxygen consumption, the heart beats faster, the blood flow speed increases, and the physiological parameters of the human body change. These changes will be reflected in the pulse waveform. This article studies the identification and classification of fingertip pulse data before and after human movement based on resnet. In this study, we used 100 fingertip pulse data as a data set for training and validation. The results show that the resnet34 network has the best prediction effect, and the prediction accuracy reaches 83.33%.