Optimization of fingernail sensing technique based on optical experimentation and modeling

The purpose of this paper is to experimentally determine the optimal optical parameters for transmitting light through the human fingernail bed in order to measure the change in blood volume that occurs when force is exerted on the fingertip. This “fingernail sensing” technique, which is a form of photoplethysmography, involves the placement of LEDs and photodetectors on the fingernail surface. As forces are applied to the fingertip, the blood perfusion in the fingernail bed is affected, resulting in various red and white regions of coloration beneath the fingernail. Thus, the transmittance of light across the fingernail bed changes as a function of applied force, resulting in a change in voltage from the photodetector circuit. Much research has previously been done to build and calibrate black-box models that estimate fingertip force based on the photodetector outputs. However, the effect of varying the wavelength and optical path length has never been thoroughly investigated. Due to recent advances in the manufacturing of fingernail sensors and the availability of surface mount LEDs of certain wavelengths, we now perform a thorough experimental characterization of the sensitivity of the transmittance to wavelength and optical path length. Results show the sensitivity is maximized when using green light (525 nm) and when the surface mount LED and photodiode are placed as close together as possible. Using data from the experiments, we also calibrate a physically-based model of the optical transmittance, which can be used to optimize fingernail sensor design.