This PDF file contains the front matter associated with SPIE Proceedings Volume 9325, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and Conference Committee listing.

Solid phantoms that serve as a proxy for human tissue provide a convenient test subject for optical medical imaging devices. In order to determine the quantitative performance of a given system, the absolute optical properties of the test subject must be known. Currently there is no national scale applicable to the scattering and absorption properties of solid diffuse tissue phantoms that would provide traceability and estimated measurement uncertainties for optical imaging applications. This paper describes progress in the development of a facility dedicated to the determination of the optical properties of solid biomedical phantoms. A brief description of the system, data analysis, Graphical User Interface (GUI), and measurement uncertainties is presented. The design is based on a double-integrating sphere, steady-state domain approach. The initial evaluation of the system includes the measurement of solid phantoms and a comparison to the manufacturer’s values that were determined by a time resolved approach. The initial results indicate that measurement agreement is within the estimated uncertainties with the coverage factor k=2.

We propose a simple and reliable solid phantom for mimicking localized absorption changes within a diffusive medium. The phantom is based on the Equivalence Relation stating that any realistic absorption inhomogeneity can be mimicked by a totally absorbing sphere of adequate volume. Applying this concept, we constructed a solid phantom holding a movable black inclusion to be positioned beneath the source-detector pair (perturbed case) or far from it (unperturbed case). Different absorption perturbations can be mimicked by changing the volume and the position of the black object both in transmittance and reflectance configuration. Time-resolved measurements of transmittance images and a lateral reflectance scan are presented.

Optical Coherence Tomography (OCT) has become a standard tool for diagnosing retinal disease in many ophthalmology clinics. Nonetheless, the technical and clinical research communities still lack a standardized phantom that could aid in evaluating and normalizing the various scan protocols and OCT machines employed at different institutions. Existing retinal phantoms designed for OCT imaging mimic some important features of the retina, such as the thickness and scattering properties of its many layers. However, the morphology of the foveal pit and the visible tapering of the retinal layers underlying the surface surrounding the pit remains a challenge to replicate in current phantoms. Recent attempts at creating a realistic foveal pit include molding, ablation and laser etching but have not proved sufficient to replicate this particular anatomical feature. In this work, we demonstrate a new fabrication procedure that is capable of replicating the tapered appearance of the retinal layers near the foveal pit using a combination of spin-coating and replica molding. The ability to create an anatomically correct foveal pit will allow for a new phantom better suited for intra- and inter-system evaluation and for improved testing of retinal segmentation algorithms.

This paper introduces novel methods to fabricate optical phantoms that simulate the morphologic, optical, and microvascular characteristics of skin tissue. The multi-layer skin-simulating phantom was fabricated by a light-cured 3D printer that mixed and printed the colorless light-curable ink with the absorption and the scattering ingredients for the designated optical properties. The simulated microvascular network was fabricated by a soft lithography process to embed microchannels in polydimethylsiloxane (PDMS) phantoms. The phantoms also simulated vascular anomalies and hypoxia commonly observed in cancer. A dual-modal multispectral and laser speckle imaging system was used for oxygen and perfusion imaging of the tissue-simulating phantoms. The light-cured 3D printing technique and the soft lithography process may enable freeform fabrication of skin-simulating phantoms that embed microvessels for image and drug delivery applications.

The emerging technique of three-dimensional (3D) printing provides a revolutionary way to fabricate objects with biologically realistic geometries. Previously we have performed optical and morphological characterization of basic 3D printed tissue-simulating phantoms and found them suitable for use in evaluating biophotonic imaging systems. In this study we assess the potential for printing phantoms with irregular, image-defined vascular networks that can be used to provide clinically-relevant insights into device performance. A previously acquired fundus camera image of the human retina was segmented, embedded into a 3D matrix, edited to incorporate the tubular shape of vessels and converted into a digital format suitable for printing. A polymer with biologically realistic optical properties was identified by spectrophotometer measurements of several commercially available samples. Phantoms were printed with the retinal vascular network reproduced as ~1.0 mm diameter channels at a range of depths up to ~3 mm. The morphology of the printed vessels was verified by volumetric imaging with μ-CT. Channels were filled with hemoglobin solutions at controlled oxygenation levels, and the phantoms were imaged by a near-infrared hyperspectral reflectance imaging system. The effect of vessel depth on hemoglobin saturation estimates was studied. Additionally, a phantom incorporating the vascular network at two depths was printed and filled with hemoglobin solution at two different saturation levels. Overall, results indicated that 3D printed phantoms are useful for assessing biophotonic system performance and have the potential to form the basis of clinically-relevant standardized test methods for assessment of medical imaging modalities.

Fabrication of optical tissue phantom (OTP) simulating whole skin structure has been regarded as laborious and time consuming work. This study fabricated multilayer OTP optically and structurally simulating epidermis-dermis structure including blood vessel. Spin coating method was used to produce thin layer mimicking epidermal layer, then optimized for reference epoxy and silicone matrix. Adequacy of both materials in phantom fabrication was considered by comparison the fabrication results. In addition similarities between OTP and biological tissue in optical property and thickness was measured to evaluate this fabrication process.

Light/tissue interactions, like diffuse reflectance, endogenous fluorescence and Raman scattering, are a powerful means for providing skin diagnosis. Instrument calibration is an important step. We thus developed multilayered phantoms for calibration of optical systems. These phantoms mimic the optical properties of biological tissues such as skin. Our final objective is to better understand light/tissue interactions especially in the case of confocal Raman spectroscopy.

The phantom preparation procedure is described, including the employed method to obtain a stratified object. PDMS was chosen as the bulk material. TiO2 was used as light scattering agent. Dye and ink were adopted to mimic, respectively, oxy-hemoglobin and melanin absorption spectra. By varying the amount of the incorporated components, we created a material with tunable optical properties.

Monolayer and multilayered phantoms were designed to allow several characterization methods. Among them, we can name: X-ray tomography for structural information; Diffuse Reflectance Spectroscopy (DRS) with a homemade fibered bundle system for optical characterization; and Raman depth profiling with a commercial confocal Raman microscope for structural information and for our final objective.

For each technique, the obtained results are presented and correlated when possible.

A few words are said on our final objective. Raman depth profiles of the multilayered phantoms are distorted by elastic scattering. The signal attenuation through each single layer is directly dependent on its own scattering property. Therefore, determining the optical properties, obtained here with DRS, is crucial to properly correct Raman depth profiles. Thus, it would be permitted to consider quantitative studies on skin for drug permeation follow-up or hydration assessment, for instance.

We present a procedure to generate digital phantoms with a hyperspectral image projector (HIP) consisting of two liquid crystal on silicon (LCoS) spatial light modulators (SLMs). The digital phantoms are 3D image data cubes of the spatial distribution of spectrally resolved abundances of intracellular light-absorbing oxy-hemoglobin molecules in single erythrocytes. Spectrally and spatially resolved image data indistinguishable from the real scene may be used as standards to calibrate image sensors and validate image analysis algorithms for their measurement quality, performance consistency, and inter-laboratory comparisons for quantitative biomedical imaging applications.

Scattering property of Intralipid is widely used for calibration and simulation of turbid media, especially biological tissues, in optical spectroscopic studies. The desired phantom turbidity level matching that of target tissue scattering properties is vital in the right preparation of phantoms mimicking the tissue. A simplified two fiber oblique illumination-collection geometry setup is used along with iterative inverse Monte Carlo simulations on the diffuse reflectance obtained experimentally for estimating the reduced scattering coefficient (μś) of Intralipid-20% for wavelengths ranging from 500 nm to 880 nm. Basic Monte Carlo for Multi Layered media (MCML) code is modified to incorporate the two fiber inverse model of diffuse reflectance with oblique broadband illumination and perpendicular collection of diffusively reflected light from the sample. Wavelength dependent true phase function of Intralipid is incorporated in the model and a semi-empirical concentration scaling methodology is used to obtain volume concentration dependence on the μś. In the inverse modelling, the modified Twersky equation for correlated scattering has been used to obtain the μś profile of Intralipid-20% for its volume concentration ranging from 16% to 100%. The results are shown to be in good agreement with the optical characterization studies of Intralipid-20% involving bulkier instrumentation for the wavelength range under consideration. The study presented in this paper gives an insight for an in vivo fiber based methodology for quantifying the variation of optical scattering during tissue malignancy.

The Diffuse Photon Density Wave (DPDW) methodology is widely used in a number of biomedical applications. Here we present results of Monte Carlo simulations that employ an effective numerical procedure, based upon a description of radiative transfer in terms of the Bethe-Salpeter equation, and compare them with measurements from Intralipid aqueous solutions. In our scheme every act of scattering contributes to the signal. We find the Monte Carlo simulations and measurements to be in a very good agreement for a wide range of source –detector separations.

Auto-fluorescence spectroscopy based on spectral line shape and intensity has been in use as a promising technique for detecting varying degrees of tissue malignancy. Tissue is a turbid medium with multi-layered structure constituting of different fluorophores, absorbers and scattering molecules. Tumor progression in tissues is ac- companied by varying degrees of biochemical and morphological changes. These include changes in nuclear size and density, epithelial thickness and increase in the hemoglobin (Hb) concentration associated with changes in metabolic activity. These variations in overall tissue scattering and absorption properties in turn modulate the fluorescence spectrum emitted and derived from tissues. Estimation of fluorescence conversion efficiency in the turbid tissue needs to take into account these effects of absorption and scattering in order to be evolved as a parameter for tissue discrimination. In this study, we set to investigate the factors affecting tissue fluorescence conversion efficiency by making use of physical models of the tissue. Liquid tissue models were prepared with different concentrations of absorbing and scattering media to simulate biological tissues of various degrees of malignancy. The results indicate that emitted fluorescence from the tissue model is subjected to variations by multiple scattering events and absorption. The fluorescence conversion efficiency of the models were derived and correlated to the experimental results with possible diagnostic significance.