Experimental and theoretical evaluation of a fiber-optic approach for optical property measurement in layered epithelial tissue

Multispectral snapshot imaging of skin microcirculatory hemoglobin oxygen saturation using artificial neural networks trained on in vivo data

SIGNIFICANCE

Developing algorithms for estimating blood oxygenation from snapshot multispectral imaging (MSI) data is challenging due to the complexity of sensor characteristics and photon transport modeling in tissue. We circumvent this using a method where artificial neural networks (ANNs) are trained on in vivo MSI data with target values from a point-measuring reference method.

AIM

To develop and evaluate a methodology where a snapshot filter mosaic camera is utilized for imaging skin hemoglobin oxygen saturation (SO2), using ANNs.

APPROACH

MSI data were acquired during occlusion provocations. ANNs were trained to estimate SO2 with MSI data as input, targeting data from a validated probe-based reference system. Performance of ANNs with different properties and training data sets was compared.

RESULTS

The method enables spatially resolved estimation of skin tissue SO2. Results are comparable to those acquired using a Monte-Carlo-based approach when relevant training data are used.

CONCLUSIONS

Training an ANN on in vivo MSI data covering a wide range of target values acquired during an occlusion protocol enable real-time estimation of SO2 maps. Data from the probe-based reference system can be used as target despite differences in sampling depth and measurement position.

Beyond diffuse correlations: deciphering random flow in time-of-flight resolved light dynamics

Diffusing wave spectroscopy (DWS) and diffuse correlation spectroscopy (DCS) can assess blood flow index (BFI) of biological tissue with multiply scattered light. Though the main biological function of red blood cells (RBCs) is advection, in DWS/DCS, RBCs are assumed to undergo Brownian motion. To explain this discrepancy, we critically examine the cumulant approximation, a major assumption in DWS/DCS. We present a precise criterion for validity of the cumulant approximation, and in realistic tissue models, identify conditions that invalidate it. We show that, in physiologically relevant scenarios, the first cumulant term for random flow and second cumulant term for Brownian motion alone can cancel each other. In such circumstances, assuming pure Brownian motion of RBCs and the first cumulant approximation, a routine practice in DWS/DCS of BFI, can yield good agreement with data, but only because errors due to two incorrect assumptions cancel out. We conclude that correctly assessing random flow from scattered light dynamics requires going beyond the cumulant approximation and propose a more accurate model to do so.

In vivo characterization of light scattering properties of human skin in the 475- to 850-nm wavelength range in a Swedish cohort

We have determined in vivo optical scattering properties of normal human skin in 1734 subjects, mostly with fair skin type, within the Swedish CArdioPulmonary bioImage Study. The measurements were performed with a noninvasive system, integrating spatially resolved diffuse reflectance spectroscopy and laser Doppler flowmetry. Data were analyzed with an inverse Monte Carlo algorithm, accounting for both scattering, geometrical, and absorbing properties of the tissue. The reduced scattering coefficient was found to decrease from 3.16 ± 0.72 to 1.13 ± 0.27 mm-1 (mean ± SD) in the 475- to 850-nm wavelength range. There was a negative correlation between the reduced scattering coefficient and age, and a significant difference between men and women in the reduced scattering coefficient as well as in the fraction of small scattering particles. This large study on tissue scattering with mean values and normal variation can serve as a reference when designing diagnostic techniques or when evaluating the effect of therapeutic optical systems.

Efficient estimation of subdiffusive optical parameters in real time from spatially resolved reflectance by artificial neural networks

Subdiffusive reflectance captured at short source-detector separations provides increased sensitivity to the scattering phase function and hence allows superficial probing of the tissue ultrastructure. Consequently, estimation of subdiffusive optical parameters has been the subject of many recent studies focusing on lookup-table-based (LUT) inverse models. Since an adequate description of the subdiffusive reflectance requires additional scattering phase function related optical parameters, the LUT inverse models, which grow exponentially with the number of estimated parameters, become excessively large and computationally inefficient. Herein, we propose, to the best of our knowledge, the first artificial-neural-network-based inverse Monte Carlo model that overcomes the limitations of the LUT inverse models and thus allows efficient real-time estimation of optical parameters from subdiffusive spatially resolved reflectance. The proposed inverse model retains the accuracy, is about four orders of magnitude faster than the LUT inverse models, grows only linearly with the number of estimated optical parameters, and can be easily extended to estimate additional optical parameters.

Modelling spatially-resolved diffuse reflectance spectra of a multi-layered skin model by artificial neural networks trained with Monte Carlo simulations

A robust modelling method was proposed to extract chromophore information in multi-layered skin tissue with spatially-resolved diffuse reflectance spectroscopy. Artificial neural network models trained with a pre-simulated database were first built to map geometric and optical parameters into diffuse reflectance spectra. Nine fitting parameters including chromophore concentrations and oxygen saturation were then determined by solving the inverse problem of fitting spectral measurements from three different parts of the skin. Compared to the Monte Carlo simulation accelerated by a graphics processing unit, the proposed modelling method not only reduced the computation time, but also achieved a better fitting performance.

Measurements of extrinsic fluorescence in Intralipid and polystyrene microspheres

The fluorescence of Intralipid and polystyrene microspheres with sphere diameter of 1 µm at a representative lipid and microsphere concentration for simulation of mucosal tissue scattering has not been a subject of extensive experimental study. In order to elucidate the quantitative relationship between lipid and microsphere concentration and the respective fluorescent intensity, the extrinsic fluorescence spectra between 360 nm and 650 nm (step size of 5 nm) were measured at different lipid concentrations (from 0.25% to 5%) and different microsphere concentrations (0.00364, 0.0073, 0.0131 spheres per cubic micrometer) using laser excitation at 355 nm with pulse energy of 2.8 µJ. Current findings indicated that Intralipid has a broadband emission between 360 and 650 nm with a primary peak at 500 nm and a secondary peak at 450 nm while polystyrene microspheres have a single peak at 500 nm. In addition, for similar scattering properties the fluorescence of Intralipid solutions is approximately three-fold stronger than that of the microsphere solutions. Furthermore, Intralipid phantoms with lipid concentrations ~2% (simulating the bottom layer of mucosa) produce up to seven times stronger fluorescent emission than phantoms with lipid concentration ~0.25% (simulating the top layer of mucosa). The fluoresence decays of Intralipid and microsphere solutions were also recorded for estimation of fluorescence lifetime.

Microcirculation assessment using an individualized model for diffuse reflectance spectroscopy and conventional laser Doppler flowmetry

Microvascular assessment would benefit from co-registration of blood flow and hemoglobin oxygenation dynamics during stimulus response tests. We used a fiber-optic probe for simultaneous recording of white light diffuse reflectance (DRS; 475-850 nm) and laser Doppler flowmetry (LDF; 780 nm) spectra at two source-detector distances (0.4 and 1.2 mm). An inverse Monte Carlo algorithm, based on a multiparameter three-layer adaptive skin model, was used for analyzing DRS data. LDF spectra were conventionally processed for perfusion. The system was evaluated on volar forearm recordings of 33 healthy subjects during a 5-min systolic occlusion protocol. The calibration scheme and the optimal adaptive skin model fitted DRS spectra at both distances within 10%. During occlusion, perfusion decreased within 5 s while oxygenation decreased slowly (mean time constant 61 s; dissociation of oxygen from hemoglobin). After occlusion release, perfusion and oxygenation increased within 3 s (inflow of oxygenized blood). The increased perfusion was due to increased blood tissue fraction and speed. The supranormal hemoglobin oxygenation indicates a blood flow in excess of metabolic demands. In conclusion, by integrating DRS and LDF in a fiber-optic probe, a powerful tool for assessment of blood flow and oxygenation in the same microvascular bed has been presented.

Accurate extraction of optical properties and top layer thickness of two-layered mucosal tissue phantoms from spatially resolved reflectance spectra

We are reporting on an experimental investigation of a movable diffuse reflectance spectroscopy system to extract diagnostically relevant optical properties of two-layered tissue phantoms simulating mucosae that are covered with stratified squamous epithelium. The reflectance spectra were measured at multiple sourcedetector separations using two imaging fiber bundles in contact with the phantoms, one with its optical axis perpendicular to the sample surface (perpendicular probe) and the other with its distal end beveled and optical axis tilted at 45 deg (oblique probe). Polystyrene microspheres and purified human hemoglobin were used to make tissue phantoms whose scattering and absorption properties could be well controlled and theoretically predicted. Monte Carlo simulations were used to predict the reflectance spectra for system calibration and an iterative curve fitting that simultaneously extracted the top layer reduced scattering coefficient, thickness, bottom layer reduced scattering coefficient, and hemoglobin concentration of the phantoms. The errors of the recovered parameters ranged from 7% to 20%. The oblique probe showed higher accuracy in the extracted top layer reduced scattering coefficient and thickness than the perpendicular probe. The developed system and data analysis methods provide a feasible tool to quantify the optical properties in vivo.

In vivo spatial frequency domain spectroscopy of two layer media

Monitoring of tissue blood volume and local oxygen saturation can inform the assessment of tissue health, healing, and dysfunction. These quantities can be estimated from the contribution of oxyhemoglobin and deoxyhemoglobin to the absorption spectrum of the dermis. However, estimation of blood related absorption in skin can be confounded by the strong absorption of melanin in the epidermis and epidermal thickness and pigmentation varies with anatomic location, race, gender, and degree of disease progression. Therefore, a method is desired that decouples the effect of melanin absorption in the epidermis from blood absorption in the dermis for a large range of skin types and thicknesses. A previously developed inverse method based on a neural network forward model was applied to simulated spatial frequency domain reflectance of skin for multiple wavelengths in the near infrared. It is demonstrated that the optical thickness of the epidermis and absorption and reduced scattering coefficients of the dermis can be determined independently and with minimal coupling. Then, the same inverse method was applied to reflectance measurements from a tissue simulating phantom and in vivo human skin. Oxygen saturation and total hemoglobin concentrations were estimated from the volar forearms of weakly and strongly pigmented subjects using a standard homogeneous model and the present two layer model.

Broadband ultraviolet-visible optical property measurement in layered turbid media

The ability to accurately measure layered biological tissue optical properties (OPs) may improve understanding of spectroscopic device performance and facilitate early cancer detection. Towards these goals, we have performed theoretical and experimental evaluations of an approach for broadband measurement of absorption and reduced scattering coefficients at ultraviolet-visible wavelengths. Our technique is based on neural network (NN) inverse models trained with diffuse reflectance data from condensed Monte Carlo simulations. Experimental measurements were performed from 350 to 600 nm with a fiber-optic-based reflectance spectroscopy system. Two-layer phantoms incorporating OPs relevant to normal and dysplastic mucosal tissue and superficial layer thicknesses of 0.22 and 0.44 mm were used to assess prediction accuracy. Results showed mean OP estimation errors of 19% from the theoretical analysis and 27% from experiments. Two-step NN modeling and nonlinear spectral fitting approaches helped improve prediction accuracy. While limitations and challenges remain, the results of this study indicate that our technique can provide moderately accurate estimates of OPs in layered turbid media.

Inverse Monte Carlo method in a multilayered tissue model for diffuse reflectance spectroscopy

Model based data analysis of diffuse reflectance spectroscopy data enables the estimation of optical and structural tissue parameters. The aim of this study was to present an inverse Monte Carlo method based on spectra from two source-detector distances (0.4 and 1.2 mm), using a multilayered tissue model. The tissue model variables include geometrical properties, light scattering properties, tissue chromophores such as melanin and hemoglobin, oxygen saturation and average vessel diameter. The method utilizes a small set of presimulated Monte Carlo data for combinations of different levels of epidermal thickness and tissue scattering. The path length distributions in the different layers are stored and the effect of the other parameters is added in the post-processing. The accuracy of the method was evaluated using Monte Carlo simulations of tissue-like models containing discrete blood vessels, evaluating blood tissue fraction and oxygenation. It was also compared to a homogeneous model. The multilayer model performed better than the homogeneous model and all tissue parameters significantly improved spectral fitting. Recorded in vivo spectra were fitted well at both distances, which we previously found was not possible with a homogeneous model. No absolute intensity calibration is needed and the algorithm is fast enough for real-time processing.

Hybrid diffusion-P3 equation in N-layered turbid media: steady-state domain

This paper discusses light propagation in N-layered turbid media. The hybrid diffusion-P3 equation is solved for an N-layered finite or infinite turbid medium in the steady-state domain for one point source using the extrapolated boundary condition. The Fourier transform formalism is applied to derive the analytical solutions of the fluence rate in Fourier space. Two inverse Fourier transform methods are developed to calculate the fluence rate in real space. In addition, the solutions of the hybrid diffusion-P3 equation are compared to the solutions of the diffusion equation and the Monte Carlo simulation. For the case of small absorption coefficients, the solutions of the N-layered diffusion equation and hybrid diffusion-P3 equation are almost equivalent and are in agreement with the Monte Carlo simulation. For the case of large absorption coefficients, the model of the hybrid diffusion-P3 equation is more precise than that of the diffusion equation. In conclusion, the model of the hybrid diffusion-P3 equation can replace the diffusion equation for modeling light propagation in the N-layered turbid media for a wide range of absorption coefficients.

Spatial frequency domain spectroscopy of two layer media

Monitoring of tissue blood volume and oxygen saturation using biomedical optics techniques has the potential to inform the assessment of tissue health, healing, and dysfunction. These quantities are typically estimated from the contribution of oxyhemoglobin and deoxyhemoglobin to the absorption spectrum of the dermis. However, estimation of blood related absorption in superficial tissue such as the skin can be confounded by the strong absorption of melanin in the epidermis. Furthermore, epidermal thickness and pigmentation varies with anatomic location, race, gender, and degree of disease progression. This study describes a technique for decoupling the effect of melanin absorption in the epidermis from blood absorption in the dermis for a large range of skin types and thicknesses. An artificial neural network was used to map input optical properties to spatial frequency domain diffuse reflectance of two layer media. Then, iterative fitting was used to determine the optical properties from simulated spatial frequency domain diffuse reflectance. Additionally, an artificial neural network was trained to directly map spatial frequency domain reflectance to sets of optical properties of a two layer medium, thus bypassing the need for iteration. In both cases, the optical thickness of the epidermis and absorption and reduced scattering coefficients of the dermis were determined independently. The accuracy and efficiency of the iterative fitting approach was compared with the direct neural network inversion.