Photoplethysmograph quality estimation through multichannel filtering

Information about the quality of a recorded physiological waveform can be valuable for the detection of critical medical conditions. This work presents a new point-by-point signal quality index (SQI) based on adaptive multichannel prediction which does not rely on ad-hoc morphological feature extraction from the target waveform. An application of the SQI to photoplethysmograph waveforms showed that the SQI is monotonically related to SNR (simulated by adding white noise) and subjective human quality assessment of 1,313 waveform epochs. A receiver-operating-characteristic (ROC) curve analysis, with the human “bad” quality label as negative and the human “good” quality label as positive, yielded an area under the ROC curve of 0.863. For photoplethysmograph waveforms, a SQI greater than 0.8 seems in general to be indicative of good signal quality.