Characterization of spatial and temporal variations in the optical properties of tissuelike media with diffuse reflectance imaging

Quality Assessment of Fruits and Vegetables Based on Spatially Resolved Spectroscopy: A Review

Damage occurs easily and is difficult to find inside fruits and vegetables during transportation or storage, which not only brings losses to fruit and vegetable distributors, but also reduces the satisfaction of consumers. Spatially resolved spectroscopy (SRS) is able to detect the quality attributes of fruits and vegetables at different depths, which is of great significance to the quality classification and defect detection of horticultural products. This paper is aimed at reviewing the applications of spatially resolved spectroscopy for measuring the quality attributes of fruits and vegetables in detail. The principle of light transfer in biological tissues, diffusion approximation theory and methodologies are introduced, and different configuration designs for spatially resolved spectroscopy are compared and analyzed. Besides, spatially resolved spectroscopy applications based on two aspects for assessing the quality of fruits and vegetables are summarized. Finally, the problems encountered in previous studies are discussed, and future development trends are presented. It can be concluded that spatially resolved spectroscopy demonstrates great application potential in the field of fruit and vegetable quality attribute evaluation. However, due to the limitation of equipment configurations and data processing speed, the application of spatially resolved spectroscopy in real-time online detection is still a challenge.

Steady-state total diffuse reflectance with an exponential decaying source

The increasing preclinical and clinical utilization of digital cameras for photographic measurements of tissue conditions motivates the study of reflectance measurements obtained with planar illumination. We examine herein a formula that models the total diffuse reflectance measured from a semi-infinite medium using an exponentially decaying source, assuming continuous plane wave epi-illumination. The model is validated with experimental reflectance measurements from tissue mimicking phantoms. The need for adjusting the blood absorption spectrum due to pigment packaging is discussed along with the potential applications of the proposed formulation.

The utility of direct-current as compared to frequency domain measurements in spectrally-constrained diffuse optical tomography toward cancer imaging

This work investigates, by means of analytical and simulation studies, the performance of spectrally-constrained image reconstruction in Continuous-Wave or Direct-Current (DC) and Frequency Domain (FD) near-infrared optical tomography. A recent analytic approach for estimating the accuracy of target recovery and the level of background artifact for optical tomography at single wavelength, based on the analysis of parametric reconstruction uncertainty level (PRUL), is extended to spectrally-constrained optical tomography. The analytical model is implemented to rank three sets of wavelengths that had been used as spectral prior in an independent experimental study. Subsequent simulation appraises the recovery of oxygenated hemoglobin (HbO), deoxygenated hemoglobin (Hb), water (H2O), scattering amplitude (A), and scattering power (b) using DC-only, DC-excluded FD, and DC-included FD, based on the three sets of wavelengths as the spectral prior. The simulation results support the analytic ranking of the performance of the three sets of spectral priors, and generally agree with the performance outcome of DC-only versus that of DC-excluded FD and DC-included FD. Specifically, this study indicate that: 1) the rank of overall quality of chromophore recovery is Hb, H2O, and HbO from the highest to lowest; and in the scattering part the A is always better recovered than b. This outcome does suggest that the DC-only information gives rise to unique solution to the image reconstruction routine under the given spectral prior. 2) DC-information is not-redundant in FD-reconstruction, as the artifact levels of DC-included FD reconstruction are always lower than those of DC-excluded FD. 3) The artifact level as represented by the noise-to-contrast-ratio is almost always the lowest in DC-only, leading to generally better resolution of multiple targets of identical contrasts over the background than in FD. However, the FD could outperform DC in the recovery of scattering properties including both A and b when the spectral prior is less optimal, implying the benefit of phase-information in scattering recovery in the context of spectrally-constrained optical tomography.

Fluorescence imaging to quantify the fluorescent microspheres in cardiac tissue

To quantify the fluorescent microsphere (FM) content in cardiac tissue, which is an indicative of blood flow, fluorescence imaging of both sides of the pig heart slice was employed. Despite the light scattering inside the tissue and contributions from multiple tissue layers to the total emission, it is shown that the fluorescence intensity at any pixel is proportional to the FM content and the fluorescence image may be transformed to the image of the FM concentration. A convenient standard for the emission-FM concentration transformation is proposed. The approach has several advantages in comparison with the traditional "digestion & extraction" method such as: non-destructiveness, high spatial resolution, high throughput, repeatability and simplicity of operation.

Rapid and accurate determination of tissue optical properties using least-squares support vector machines

Diffuse reflectance spectroscopy (DRS) has been extensively applied for the characterization of biological tissue, especially for dysplasia and cancer detection, by determination of the tissue optical properties. A major challenge in performing routine clinical diagnosis lies in the extraction of the relevant parameters, especially at high absorption levels typically observed in cancerous tissue. Here, we present a new least-squares support vector machine (LS-SVM) based regression algorithm for rapid and accurate determination of the absorption and scattering properties. Using physical tissue models, we demonstrate that the proposed method can be implemented more than two orders of magnitude faster than the state-of-the-art approaches while providing better prediction accuracy. Our results show that the proposed regression method has great potential for clinical applications including in tissue scanners for cancer margin assessment, where rapid quantification of optical properties is critical to the performance.

Angular domain transillumination imaging optimization with an ultrafast gated camera

By employing high-aspect-ratio parallel microchannels as an angular filter, quasiballistic photons sensitive to internal structures in a turbid medium can be captured. Scattered photons exiting the turbid medium typically exhibit trajectories with random angles compared to the initial trajectory and are mostly rejected by the filter. However, angular filter arrays cannot differentiate between quasiballistic photons (early arriving) and photons that happen to attain a scattered trajectory that is within the acceptance angle (late arriving). Therefore, we have two objectives: (1) to experimentally characterize the angular distribution and proportion of minimally deviated quasiballistic photons and multiply scattered photons in a turbid medium and (2) to combine time and angular gating principles so that early and late arriving photons can be distinguished. From the angular distribution data, the angular filter with angular acceptance about 0.4 deg yields the highest image contrast for transillumination images. The use of angular domain imaging(ADI) with time-gating enables visualization of submillimeter absorbing objects with approximately seven times higher image contrast compared to ADI in a turbid medium with a scattering level of six times the reduced mean free path.

Direct-current-based image reconstruction versus direct-current included or excluded frequency-domain reconstruction in diffuse optical tomography

We study the level of image artifacts in optical tomography associated with measurement uncertainty under three reconstruction configurations, namely, by using only direct-current (DC), DC-excluded frequency-domain, and DC-included frequency-domain data. Analytic and synthetic studies demonstrate that, at the same level of measurement uncertainty typical to optical tomography, the ratio of the standard deviation of mu(a) over mu(a) reconstructed by DC only is at least 1.4 times lower than that by frequency-domain methods. The ratio of standard deviations of D (or mu(s)') over D (or mu(s)') reconstructed by DC only are slightly lower than those by frequency-domain methods. Frequency-domain reconstruction including DC generally outperforms that excluding DC, but as the amount of measurements increases, the difference between the two diminishes. Under the condition of a priori structural information, the performances of three reconstruction configurations are seemingly equivalent.

Solution of transport equations in layered media with refractive index mismatch using the PN-method

The PN-method is a spectral discretization technique used to obtain numerical solutions to the radiative transport equation. To the best of our knowledge, the PN-method has yet to be generalized to the case of refractive index mismatch in layered slabs used to numerically simulate skin. Our main contribution is the application of a collocation method that takes into account refractive index mismatch at layer interfaces. The stability, convergence, and accuracy of the method are established. Example calculations demonstrating the flexibility of the method are performed.

In vivo fluorescence imaging: a personal perspective

In vivo fluorescence imaging with near-infrared (NIR) light holds enormous potential for a wide variety of molecular diagnostic and therapeutic applications. Because of its quantitative sensitivity, inherent biological safety, and relative ease of use (i.e., with respect to cost, time, mobility, and its familiarity to a diverse population of investigators), fluorescence-based imaging techniques are being increasingly utilized in small-animal research. Moreover, there is substantial interest in the translation of novel optical techniques into the clinic, where they will prospectively aid in noninvasive and quantitative screening, disease diagnosis, and post-treatment monitoring of patients. Effective deep-tissue fluorescence imaging requires the application of exogenous NIR-emissive contrast agents. Currently, available probes fall into two major categories: organic and inorganic NIR fluorophores (NIRFs). In the studies reviewed herein, we utilized polymersomes (50 nm to 50 microm diameter polymer vesicles) for the incorporation and delivery of large numbers of highly emissive oligo (porphyrin)-based, organic NIRFs.

Intrinsic Raman spectroscopy for quantitative biological spectroscopy part I: theory and simulations

We present a novel technique, intrinsic Raman spectroscopy (IRS), to correct turbidity-induced Raman spectral distortions, resulting in the intrinsic Raman spectrum that would be observed in the absence of scattering and absorption. We develop an expression relating the observed and intrinsic Raman spectra through diffuse reflectance using the photon migration depiction of light transport. Numerical simulations are employed to validate the theoretical results and study the dependence of this expression on sample size and elastic scattering anisotropy.

Assessment of near-infrared path length in fibrous phantom and muscle tissue

The first derivative of the pseudo-absorption spectrum of a water-loaded cotton wool (water-CW) phantom, which mimics muscle tissues, was used to determine the light path length in the near-infrared (NIR) region. The light path length increased as the density of the turbid medium decreased. It is independent of both water content in the range of 75-85% (by weight) and the diffuse reflecting reference used to determine the pseudo-absorbance. The path length determination procedure was verified by measurements of diffuse reflectance in chicken breast tissue for which the path length of 1.8 mm (differential path length factor, DPF = 2.1) was found to be similar to the path length of NIR light of 1.5-2.2 mm (DPF = 1.8-2.6) in a water-CW phantom of density similar to chicken breast. We conclude that the NIR light path length can serve as a characteristic of muscle tissue density.

Sub-epidermal imaging using polarized light spectroscopy for assessment of skin microcirculation

BACKGROUND/AIMS

Many clinical conditions that affect the microcirculation of the skin are still diagnosed and followed up by observational methods alone in spite of the fact that non-invasive, more user-independent and objective methods are available today. Limited portability, high cost, lack of robustness and non-specificity of findings are among the factors that have hampered the implementation of these methods in a clinical setting. The aim of this study is to present and evaluate a new, portable and easy-to-use imaging technology for investigation of the red blood cell (RBC) concentration in the skin microvasculature based on the method of polarization light spectroscopy using modified standard digital camera technology.

METHODS

The use of orthogonal linear polarization filters over both the flash source and the detector array removes the polarization-retaining light reflected from the epidermal layer. Only the depolarized light backscattered from the papillary dermal matrix reaches the detector array. By separating the RGB color planes of an image acquired in this manner and applying a dedicated image processing algorithm, spectroscopic information about the chromophores in the dermal tissue can be attained. If the algorithm is based on a differential principle in which the normalized differences between the individual values of the red and green color plane are calculated, tissue components with similar spectral signature in both planes are suppressed, while components with different spectral signatures such as RBCs are enhanced.

RESULTS

In vitro fluid models compare well with theory and computer simulations in describing a linear relationship between the imager output signal termed the tissue viability index (TiVi(index)) and RBC concentration in the physiological range of 0-4% RBC fraction of tissue volume (cc=0.997, n=20). The influence of oxygen saturation on the calculated RBC concentration is limited to within -3.9% for values within the physiological range (70-100% oxygen saturation). Monte Carlo simulations provide information about the sampling depth (about 0.5 mm on the average) of the imaging system. In vivo system evaluation based on iontophoresis of acetylcholine displays a heterogeneous pattern of vasodilatation appearing inside the electrode area after about 10 min. Topical application of methyl nicotinate and clobetasol propionate further demonstrates the capacity to document the extent and intensity of both an increase (erythema) and a decrease (blanching) in the skin RBC concentration without movement artifact and with compensation for irregularity in pigmentation.

CONCLUSIONS

Polarization light spectroscopy imaging for assessment of RBC concentration in the skin microvasculature is a robust and accessible technique for the clinical setting. Additionally, the technique has pre-clinical research applications for investigation of the spatial and temporal aspects of skin erythema and blanching as well as a potential role in drug development, skin care product development and skin toxicological assessment.

Effects of turbid media optical properties on object visibility in subsurface polarization imaging

We studied the effectiveness of using polarized illumination and detection to enhance the visibility of targets buried in highly scattering media. The effects of background optical properties including scattering coefficient, absorption coefficient, and anisotropy on image visibility were examined. Both linearly and circularly polarized light were used in the imaging. Three different types of target were investigated: scattering, absorption, and reflection. The experimental results indicate that target visibility improvement achieved by a specific polarization method depends on both the background optical properties and the target type. By analyzing all polarization images, it is possible to reveal certain information about target or the scattering background.

Determination of optical properties of normal and adenomatous human colon tissues in vitro using integrating sphere techniques

AIM: The purpose of the present study is to compare the optical properties of normal human colon mucosa/submucosa and muscle layer/chorion, and adenomatous human colon mucosa/submucosa and muscle layer/chorion in vitro at 476.5, 488, 496.5, 514.5 and 532 nm. We believe these differences in optical properties should help differential diagnosis of human colon tissues by using optical methods.

METHODS: In vitro optical properties were investigated for four kinds of tissues: normal human colon mucosa/submucosa and muscle layer/chorion, and adenomatous human colon mucosa/submucosa and muscle layer/chorion. Tissue samples were taken from 13 human colons (13 adenomatous, 13 normal). From the normal human colons a total of 26 tissue samples, with a mean thickness of 0.40 mm, were used (13 from mucosa/submucosa and 13 from muscle layer/chorion), and from the adenomatous human bladders a total of 26 tissue samples, with a mean thickness of 0.40 mm, were used (13 from mucosa/submucosa and 13 from muscle layer/chorion). The measurements were performed using a double-integrating-sphere setup and the optical properties were assessed from these measurements using the adding-doubling method that was considered reliable.

RESULTS: The results of measurement showed that there were significant differences in the absorption coefficients and scattering coefficients between normal and adenomatous human colon mucosa/submucosa at the same wavelength, and there were also significant differences in the two optical parameters between both colon muscle layer/chorion at the same wavelength. And there were large differences in the anisotropy factors between both colon mucosa/submucosa at the same wavelength, there were also large differences in the anisotropy factors between both colon muscle layer/chorion at the same wavelength. There were large differences in the value ranges of the absorption coefficients, scattering coefficients and anisotropy factors between both colon mucosa/submucosa, and there were also large differences in these value ranges between both colon muscle layer/chorion. There are the same orders of magnitude in the absorption coefficients for four kinds of colon tissues. The scattering coefficients of these tissues exceed the absorption coefficients by at least two orders of magnitude.

CONCLUSION: There were large differences in the three optical parameters between normal and adenomatous human colon mucosa/submucosa at the same laser wavelength, and there were also large differences in these parameters between both colon muscle layer/chorion at the same laser wavelength. Large differences in optical parameters indicate that there were large differences in compositions and structures between both colon mucosa/submucosa, and between both colon muscle layer/chorion. Optical parameters for four kinds of colon tissues are wavelength dependent, and these differences would be useful and helpful in clinical applications of laser and tumors photodynamic therapy (PDT).

Comparative analysis of selected radiative transfer approaches for aquatic environments

A comparative analysis is presented of simple approaches to radiative transfer in plane-parallel layers, such as the self-consistent Haltrin approach, the Chandrasekhar-Klier exact solution for isotropic scatters, an extended version of two-flux radiative Kubelka-Munk theory, the neutron-diffuse Gate-Brinkworth theory, and different versions of the delta-Eddington theory. It is shown that the Haltrin approach is preferable to others and can be used for the solution of an inverse optical problem of the estimation of absorption and backscattering coefficients of aquatic environments from measured apparent optical properties. Two different methods of transformation from measured irradiance reflectance at combined illumination to irradiance reflectance induced by diffuse illumination only are developed. An analysis of the use of the different models for estimation of the effect of the bottom albedo is also presented.