Hybrid Optical Unobtrusive Blood Pressure Measurements

Figures

• 

  Figure 1. Use case of the ICPPG system (combination of contact photoplethysmography (cPPG) and imaging photoplethysmography (iPPG)). The ICPPG system can be implemented on a wristwatch with a contact photoplethysmography (PPG) sensor at the back side and a camera on the top. The pulse transit time (PTT) can be calculated by the temporal difference between the morphologic valleys of PPG signals at the wrist and the iPPG signal at the face.

• 

  Figure 2. IPPG signal extraction and region of interest (ROI) selection. (a) An example of fixed ROI (blue block) and the tracking ROI (red blocks) with the increasing number of frames. It was found that the tracking ROI is able to adapt to the movement of the participant while fixed ROI cannot; (b) An example of the 2-D correlation between the skin within ROI and the original template of the skin that was selected in the first video frame vs. number of frames. It shows that the 2-D correlation of the skin within tracking ROI is higher and more stable than fixed ROI.

• 

  Figure 3. PTT calculation. The IC-PTT (combination of imagery and contact PTT) (and the Ref-PTT, a conventional physiological measurement system) calculated by the temporal difference between the onset points for each beat in the pair of IC-PPG signals (and the Ref-PPG signal).

• 

  Figure 4. Data collection. (a) Setup for data collection; (b) Schematic overview of the data collection on the subject; (c) Sessions with and without exercise. Note that IC-PTT is the PTT produced by the ICPPG system and Ref-PTT is the PTT produced by the reference system.

• 

  Figure 5. Morphologic differences between PPG signals. An example of the morphologic difference of PPG signals at different sites with different sensing modalities.

• 

  Figure 6. The correlation between IC-PTT and Ref-PTT for relaxation and exercise scenarios. (a) The correlation between IC-PTT and Ref-PTT of relaxation sessions; (b) The correlation between IC-PTT and Ref-PTT of exercise sessions.

• 

  Figure 7. The variability of IC-PTT and Ref-PTT. (a) The mean value and standard deviation of IC-PTT and Ref-PTT for each subject during the experiment; (b) Standard deviations of IC-PTT and Ref-PTT. There is a significant difference between the variability of IC-PTT and Ref-PTT.

• 

  Figure 8. Correlation between IC-PTT (and Ref-PTT) and BP. (a) PTT (both IC-PTT and Ref-PTT) varied with BP (systolic BP (SBP), diastolic BP (DBP), and mean BP (mBP)) over time during the experiment; (b) Correlation coefficient between IC-PTT, SBP, DBP, and mBP per subject; (c) The absolute correlations coefficient between IC-PTT and BP (SBP, DBP, and mBP) and that between Ref-PTT and BP (SBP, DBP, and mBP).