VPE-Net: Simultaneous Measurement of Heart Rate, Respiration Rate, and Blood Pressure from PPG

Cardiopulmonary diseases are the primary reason behind worldwide mortality.Sudden and abnormal changes in heart rate (HR), respiration rate (RR), and blood pressure (BP) are some of the primary indicators of these diseases.Hence, unobtrusively monitoring these parameters is crucial for their use in the home environment.However, no prior study has been found that extracted all these parameters simultaneously from PPG.In this paper, we have proposed a first-of-its-kind deep learning framework, 'VPE-Net,' for estimating RR, HR, and BP (both systolic and diastolic) simultaneously from PPG without extracting any feature manually.The proposed model is designed by incorporating the framework of both gated and long-shortterm recurrent convolutional networks and yields normalized mean absolute error of 0.0714 breath rate per minute, 0.0619 beats per minute, 0.162 mmHg, and 0.0321 mmHg for RR, HR, SBP, and DBP respectively.