A Portable Multi-Modal Cushion for Continuous Monitoring of a Driver’s Vital Signs

Figures

• 

  Figure 1. Model of active electrode (altered from [17]).

• 

  Figure 2. Principle of reflective PPG (inspired by [37]).

• 

  Figure 3. Principle of MIM (altered from [19]).

• 

  Figure 4. System overview. On the left side, a block diagram of the system is depicted. The orange highlighted block depicts the top side of the PCB (shown on the top right). The blue highlighted block depicts the bottom side of the PCB. On the right side, the cushion with the controller box is depicted.

• 

  Figure 5. Simulator and protocol.

• 

  Figure 6. Workflow for processing of signals. Please note that the peak detection for the cECG signals is performed with the Pan–Tompkins algorithm and the algorithm of Brüser et al. is used to extract the HR of the SCG.

• 

  Figure 7. Signals with good quality. The reference peaks of the conductive ECG and impedance pneumography are shown as dashed red or blue lines, respectively.

• 

  Figure 8. Signals with bad quality. The reference peaks of the conductive ECG and impedance pneumography are shown as dashed red or blue lines, respectively.

• 

  Figure 9. Example of unreliable respiratory reference. The respiratory signals extracted from rPPG1 and rPPG3 are shown for comparison.

• 

  Figure 10. AUC of HR for different stages for cECG and reflective PPG. The bar shows the median across all participants and the lines show the standard deviation.

• 

  Figure 11. AUC of RR for different stages for cECG, MIM and reflective PPG. The bar shows the median across all participants and the lines show the standard deviation.

• 

  Figure 12. AUC of RR for different stages for SCG. The bar shows the median across all participants and the lines show the standard deviation.