Fig. 1. Mechanical analysis of HB structure. (a) Deformation and von Mises stress distribution of a double-layer PDMS hemispherical structure under an average stress of 4 kPa (radius of primary and secondary structures is 50 μm and 9 μm respectively); (b) deformation process and electrical response of pressure sensors with HB structure (depth of color indicates degree of deformation, and the darker the color, the greater the deformation)

Fig. 2. HB sensor and pulse wave measurement system. (a) Structure schematic diagram of HB sensor; (b) structure schematic diagram of pulse wave measurement system

Fig. 3. Schematic diagram of the calculation of I_ar

Fig. 4. Schematic diagram of the calculation of HR

Fig. 5. Structural characterization of HB sensor. (a) Cross-sectional SEM image of primary and secondary layer; (b) tilted SEM image of primary layer; (c) plane SEM image of secondary layer uncoated with rGO

Fig. 6. Test results of response curves of sensors with different structures. (a) Relationship between relative current variation and pressure applied on the sensor; (b) elastic modulus of sensors (S₁ and S₂ are sensitivity of the HB sensor within two linear ranges respectively, E₁, E₂ and E_HB are elastic modulus of single layer 1 sensor, single layer 2 sensor and HB sensor respectively)

Fig. 7. Application of HB sensor for pulse monitoring. (a) Recorded pulse waveforms of the volunteer at rest (in a seated position) (upper right illustration shows single-cycle pulse signals, R₁ and R₂ are intensity of P₁ and P₂ peaks, top left illustration shows HB sensor attached to radial artery of twrist); (b) recorded pulse waveforms of volunteer after exercise (inset graph shows detailed pulse signals); FFT results of recorded pulse waveforms at rest (c) and after exercise (d); comparison of I_ar (p<0.0001) (e) and RWTT (p<0.0001) (f) between rest and after exercise

Fig. 8. Application of HB sensor in quantitative pulse diagnosis of traditional Chinese medicine. (a) Dynamic pulse pressure and corresponding pulse wave; (b) baseline pressure extracted from original signal; (c) pulse wave signal extracted from original signal; (d)‒(f) original value, average value and spectrum of pulse wave under different pulse pressures

Fig. 9. Pulse waveform and amplitude quantification at different finger pressures. (a) Average pulse waveform under different finger pressures; (b) average pulse amplitude (R₁) under different finger pressures; (c) sectional view of anatomical structure at radial artery

Fig. 10. Framework diagram of CAMKformer model

Fig. 11. Bland-Altman diagram. (a) SBP; (b) DBP

Fig. 12. Visualization analysis of the importance of pulse wave features in different types of volunteers (color represents degree of impact of data on model classification results). (a) Healthy volunteers under normal blood pressure; (b) high blood pressure volunteers under high blood pressure; (c) hypertensive volunteers under high blood pressure