Simulation-based optimization of a near-infrared spectroscopic subcutaneous fat thickness measuring device

Using Monte Carlo simulations we optimized the wavelength and source-detector distance (SDD) of a reflectance-based spectroscopic device used for measuring subcutaneous fat thickness. As the optical properties of muscle, fat and dermis are wavelength dependent, it is necessary to choose a wavelength that is highly sensitive to fat but insensitive to water and melanin. The SDD is important since it determines average photon penetration depth. With a tissue optics plug-in for the GEANT4/GAMOS system and published ex vivo tissue optical properties we were able to predict the behavior of different device configurations when used with varying thicknesses of fat, melanin concentrations or hydration levels. Our results indicate that the ideal wavelengths for fat measurement are 630-650 nm with an SDD of 2.6-29 cm. We also examined the potential of using near infrared (NIR) spectroscopy to determine tissue hydration levels, but concluded that this wavelength range was not ideal.