Efficient construction of robust artificial neural networks for accurate determination of superficial sample optical properties

Abstract

Abstract:In general, diffuse reflectance spectroscopy (DRS) systems work with photon diffusion models to determine the absorption coefficient μ a and reduced scattering coefficient μ s ' of turbid samples. However, in some DRS measurement scenarios, such as using short source-detector separations to investigate superficial tissues with comparable μ a and μ s ', photon diffusion models might be invalid or might not have analytical solutions. In this study, a systematic workflow of constructing a rapid, accurate photon transport model that is valid at short source-detector separations (SDSs) and at a wide range of sample albedo is revealed. To create such a model, we first employed a GPU (Graphic Processing Unit) based Monte Carlo model to calculate the reflectance at various sample optical property combinations and established a database at high speed. The database was then utilized to train an artificial neural network (ANN) for determining the sample absorption and reduced scattering coefficients from the reflectance measured at several SDSs without applying spectral constraints. The robustness of the produced ANN model was rigorously validated. We evaluated the performance of a successfully trained ANN using tissue simulating phantoms. We also determined the 500-1000 nm absorption and reduced scattering spectra of in-vivo skin using our ANN model and found that the values agree well with those reported in several independent studies.

Toward reliable retrieval of functional information of papillary dermis using spatially resolved diffuse reflectance spectroscopy

“…Our DRS system, which has been described in detail previously, consists of a supercontinuum source (NKT photonics, Denmark), an 1x4 optical switch (Piezosystem Jena, Germany) and a spectrometer equipped with a back-thinned CCD (QE65000, Ocean Optics, FL) capable of collecting light in the wavelength range from 500 to 1000 nm [13]. Two custom made optical fiber probes were employed in this study and the side view of the probes are shown in Fig.…”

“…Cappon et al utilized an optical fiber probe to collect reflectance at SDSs of 0.23, 0.59, and 1.67 mm for estimating the superficial samples having optical properties comparable to those of brain tissues; they calculated the average probing depth of the probe was roughly less than 0.5 mm [12]. More recently, we employed a Monte Carlo simulation trained Artificial Neural Network (ANN) to recover skin optical properties from diffuse reflectance measured at SDSs of 1 and 2 mm [13].…”

Light distribution modulated diffuse reflectance spectroscopy

“…To achieve real time sample optical property recovery, we constructed an artificial neural network (ANN), which is trained by a database composed of numerous Monte Carlo simulation results, to efficiently determine the sample optical properties from the measured reflectance. The essential steps for constructing ANN models for this purpose were elaborated in our previous paper [13]. We employed the Matlab (Mathworks, MA) training function "trainbr" based on Bayesian regularization to update the weight and bias values according to LevenbergMarquardt optimization.…”

“…We constructed ANNs for this purpose. The essential steps of the construction and validation of these ANNs are very similar to those listed in our previous study [13] and are described in detail as follows.…”

“…for μ s ' to account for the human skin optical properties variation range from 600 to 1000 nm [13,15,16]. We divided μ a and μ s ' ranges both into 60 equal steps to construct a training database containing the diffuse reflectances of 3600 optical property combinations.…”