Date of Award


Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

First Advisor

Kirk Shelley


The pulse oximeter measures the arterial oxygen saturation. It accomplishes this through the use of the photoplethysmograph waveform (PPG) at two or more wavelengths. It has been recognized for some time that the movement of venous blood can be detected using the PPG. We hypothesize that the PPG waveform, obtained non-invasively by modern pulse oximeters, can be analyzed via digital signal processing to infer the venous oxygen saturation. Fundamental to the successful isolation of the venous saturation is the identification of PPG characteristics that are unique to the peripheral venous system. Two such characteristics have been identified. First, the peripheral venous waveform tends to reflect atrial contraction (e.g., a-c-v waveform). Second, ventilation tends to move venous blood preferentially due to the low pressure and high compliance of the venous system. Red (660nm) and IR (940nm) PPG waveforms were collected from 10 cardiac surgery patients using an esophageal PPG probe. These waveforms were analyzed using algorithms written in Mathematica (Wolfram Research). The eight saturation algorithms (ArtSat, VenSat, ArtInstSat, VenInstSat, RespDC, RespAC, Cardiac, and Harmonic) were applied to the data set. Three of the methods (VenSat, VenInstSat, and RespDC) demonstrate significance difference from ArtSat using the Wilcoxon signed-rank test with Bonferroni correction (p<0.0071). This thesis introduces new methods of PPG analysis. Three methods of analysis (VenSat, VenInstSat, and RespDC) succeed in detecting lower saturation blood. The next step is to confirm the accuracy of the measurement by comparing them to a gold standard (i.e., venous blood gas).