Measurement of arterial oxygen saturation (SpO2) using PPG optical sensor

The pulse oximeters are used to measure the heart rate and arterial oxygen saturation (SpO
2
). The photoplethysmographic (PPG) signals from pulse oximeters are highly influenced by the noise generated due to movement. The signal is also affected by the ambience light signal. It is essential to obtain a clean signal for the measurement of arterial oxygen saturation (SpO
2
). In this paper, an Adaptive Noise Cancelling (ANC) technique is being used to remove noise from the signal. A tri-axial based accelerometer data is also taken during the data acquisition time. Normalized Least Mean Square (NLMS) based adaptive filter algorithm is adopted. Tri-axis accelerometer data gives the directional data along the entire three axes. They are summed up and they are fed into the adaptive filter as reference noise signal. In the filter, the coefficients are updated based on the error signal and they are updated till the error becomes minimum. From the obtained clean signal oxygen saturation is calculated by finding the normalized ratio of ratios. Accelerometer based adaptive noise cancellation is found to reduce the errors in arterial oxygen saturation. The mean result for the mentioned algorithm is around 0.2409. The result is also found to be in correlation with the reference pulse oximeter.