Analysis and Synthesis of Respiratory Rate for Male Patients

In this paper an algorithm to estimate the respiratory rate (RR) using the oxygen saturation signal is presented. This is achieved by using the median and moving average filters with an adaptive window sizes technique according to the characteristics of the input data. This proposed algorithm is tested for eight (8) male patients, and the information is provided by "Physionet.org". This algorithm receives an oxygen saturation signal or pulse signal (PPG signal) as input. Therefore, this algorithm first removes motion artifacts (MA) from the PPG signal. In the first step, the MA frequencies from 0 to 0.5 Hz are vanished. Then, the second step extracts the RR part from the PPG signal. This is achieved by decreasing its frequency power above 0.8 Hz. The simulation results show an efficiency of 95% having a maximum error of 3 breaths on each calculation and an average error of 0.57 breaths per minute. However, the algorithm shows robustness by obtaining accurate results in different PPG signals.