Emerging sensing and modeling technologies for wearable and cuffless blood pressure monitoring

Figures

• 

  Fig. 1. Flow diagram of wearable and cuffless BP monitoring system. Sensors read specific signals from the body and send them to the model for processing and analysis to obtain the BP estimate.

• 

  Fig. 2. Flexible TAG sensors for epidermal pulse monitoring. a Pulse sensing mechanism. Reprinted with permission from ref. 21. Copyright 2020 Wiley. b A piezoresistive sensor adhering to the human wrist seamlessly. Reprinted with permission from ref. 22. Copyright 2017 Wiley. c A typical pulse waveform consisting of three characteristic peaks. Reprinted with permission from ref. 21. Copyright 2020 Wiley. d Piezocapacitive sensing mechanism. e A piezocapacitive sensor made of a micropillar dielectric layer, and its pressure transfer curve. Reprinted with permission from ref. 24. Copyright 2019 American Chemical Society. f Piezoresistive sensing mechanism. g A piezoresistive sensor made of a carbon-decorated fabric and its pressure transfer curve. Reprinted with permission from ref. 25. Copyright 2015 Wiley. h Piezoelectric sensing mechanism. i A piezoelectric sensor made of a PZT film and its transfer curve. Reprinted with permission from ref. 26. Copyright 2017 Wiley. j Triboelectric sensing mechanism. k A triboelectric sensor made of woven PTFE stripes and a PET film, together with its sensing mechanism. Reprinted with permission from ref. 27. Copyright 2018 Wiley.

• 

  Fig. 3. ECG electrodes. a Schematic diagram of in-pillow capacitive cloth electrodes40. b Schematic diagrams of the standard Ag/AgCl electrode and prepared electrode78.

• 

  Fig. 4. Working principle of PPG195. a Reflective mode. b transmitting mode. c example of PPG signal. d Reflective mode epidermal and transmissive mode flexible hybrid organic/inorganic NIR PPG for BP monitoring. Reprinted with permission from ref. 59. Copyright 2017 Wiley.

• 

  Fig. 5. Schematic diagram of the stretchable ultrasound device. Reprinted with permission from ref. 70. Copyright 2018 Nature.

• 

  Fig. 6. Schematic diagram of BP information in different temporal resolutions. From fine to coarse-grained: continuous BP waveform, beat-to-beat BP, and snapshot BP.

• 

  Fig. 7. Illustration of two mainstream definitions of PTT and PAT. Mukkamala et al.84: ①—PAT, ②③—PTT; Ding et al.85: ①—PTTEP, ②—PTTIP, ③—PTTPP, or PAT.

• 

  Fig. 8. General pipeline of machine learning algorithms for BP monitoring. Traditional ML contains a separate feature extraction step, while DL can perform end-to-end inference.