a, The sensor system consists of a peripherally mounted sensor capable of MWPPG measurements, and a chest-mounted sensor capable of ECG and triaxial accelerometry. Both sensors are coordinated wirelessly via a Bluetooth-enabled control module. b, Time-synchronized measurement of ECG and PPG enables estimation of systemic blood pressure via arterial pulse wave velocity, while MWPPG enables quantification of local/peripheral arteriolar states. c, Combining these hemodynamic sensors enables differentiation of hemodynamic states in dimensions of clinically pertinent hemodynamic parameters of blood pressure and vascular resistance. d, Exploded rendering of the multispectral device with silicone encapsulation layers, fPCB consisting of four islands separated via flexible serpentine interconnects and support circuitry. e, Left: bottom view of the multiwavelength sensor with multiple pairs of individually controllable broadband and NIR LEDs. Source-detector distances are 3 mm, 6.5 mm, 10 mm and 13.5 mm. Right: time-sequential LED sequencing allows for spatially resolved spectroscopy useful in tissue oximetry. f, Side-by-side views of two different form factors achievable by folding the fPCB. g, Device placement as a finger wrap. h, Mechanical twisting of the peripheral multiwavelength device in its short form. i, Diagram indicating various possible placements of the devices and associated normalized waveforms, offset for clarity, from central and peripheral locations, and with both arterial-dominated or capillary-dominated vascular beds (more detail in Supplementary Fig. 2). Obtained during deep breathing exercises, the signal contains both low-frequency oscillations due to respiration and high-frequency oscillations due to the cardiac cycle. j, Monochromator scan of an encapsulated device showing relative responsivity (left axis) at 9 different wavelengths throughout the visible and NIR ranges as well as LED pair emittance spectrum (black, right axis). k, Diagram of the measurement principles showing simultaneous wavelength- and distance-based light paths through a photoacoustic tomography rendering of real human skin. Portions of this figure were created with BioRender.com.