Analysis and Synthesis of Respiratory Rate for Female Patients

In this paper, a new method for estimating the respiratory rate (RR) signal using pulse oximeter is proposed. The PPG signal acquired from female patients is composed by three main elements which are the PPG signal itself, the motion artifacts and the respiratory rate. This paper analyzes eight (8) female patients from the Beth Israel Deaconess Medical Centre in Boston. The information (data) is collected from `physionet.org'. The PPG normally has a significant amount of power ranging from 0 to 10 Hz and the motion artifacts are around 0 to 0.5 Hz. Therefore, the first step is to remove the motion artifacts and then extract the RR from rest of the PPG+RR signal. A record of 480 seconds (8 minutes) is analyzed from each female patient. All operations are performed in time domain digital signal processing. The algorithm performs several operations using median and moving average filters to estimate the RR component. The results give an efficiency of 96.5% considering an average error of +/- 0.62 breathes per minute.