SignificanceOver the past decade, machine learning (ML) algorithms have rapidly become much more widespread for numerous biomedical applications, including the diagnosis and categorization of disease and injury.AimHere, we seek to characterize the recent growth of ML techniques that use imaging data to classify burn wound severity and report on the accuracies of different approaches.ApproachTo this end, we present a comprehensive literature review of preclinical and clinical studies using ML techniques to classify the severity of burn wounds.ResultsThe majority of these reports used digital color photographs as input data to the classification algorithms, but recently there has been an increasing prevalence of the use of ML approaches using input data from more advanced optical imaging modalities (e.g., multispectral and hyperspectral imaging, optical coherence tomography), in addition to multimodal techniques. The classification accuracy of the different methods is reported; it typically ranges from ∼70% to 90% relative to the current gold standard of clinical judgment.ConclusionsThe field would benefit from systematic analysis of the effects of different input data modalities, training/testing sets, and ML classifiers on the reported accuracy. Despite this current limitation, ML-based algorithms show significant promise for assisting in objectively classifying burn wound severity.
The complex cerebrovascular network is critical to controlling local cerebral blood flow (CBF) and maintaining brain homeostasis. Alzheimer’s disease (AD) and neurological injury can result in impaired CBF regulation, blood–brain barrier breakdown, neurovascular dysregulation, and ultimately impaired brain homeostasis. Measuring cortical hemodynamic changes in rodents can help elucidate the complex physiological dynamics that occur in AD and neurological injury. Widefield optical imaging approaches can measure hemodynamic information, such as CBF and oxygenation. These measurements can be performed over fields of view that range from millimeters to centimeters and probe up to the first few millimeters of rodent brain tissue. We discuss the principles and applications of three widefield optical imaging approaches that can measure cerebral hemodynamics: (1) optical intrinsic signal imaging, (2) laser speckle imaging, and (3) spatial frequency domain imaging. Future work in advancing widefield optical imaging approaches and employing multimodal instrumentation can enrich hemodynamic information content and help elucidate cerebrovascular mechanisms that lead to the development of therapeutic agents for AD and neurological injury.
KEYWORDS: Machine learning, Color imaging, Hyperspectral imaging, Data modeling, Multispectral imaging, Digital color imaging, Diffuse optical imaging, Biopsy
Accurately classifying burn severity is crucial to inform proper treatment. Here, we quantitatively compare the efficacy of machine learning (ML) burn classification algorithms using multispectral imaging versus conventional digital color imaging data as inputs. We imaged 80 porcine burns that underwent biopsy and histology for ground truth categorization into “skin graft needed” versus “no graft needed” groups. The accuracy of our ML algorithm with a transfer learning architecture was 97.5% for the multispectral model, 57.5% for the digital-color model, and 57.5% for the multispectral+digital-color model. This result strongly supports the use of multispectral imaging over digital-color imaging for burn classification.
Absorption of light by epidermal melanin can confound light-based measurements of tissue oxygenation, reducing the ability of these technologies to reliably identify hypoxia. Spatial frequency domain imaging (SFDI) may help address this critical concern, because the different spatial frequencies of light penetrate different depths into the tissue, facilitating separation between contributions of different tissue layers to the detected signal. Here, we rigorously investigate the relationship between the skin tone of healthy subjects and the tissue oxygen saturation (StO2) measured with multispectral SFDI. This study helps to quantify the degree to which skin tone influences SFDI measurements of tissue oxygenation.
The conservation orbital angular momentum and polarization for beams propagating through scattering bio-soft matter enables multiplexed signaling. By utilizing nonlinear optical effects in the scattering bio-soft-matter, we investigate the conservation of polarization and OAM through self-trapping and pump/probe coupled waveguides of light in sheep red blood cell suspensions at 532 nm and 780nm wavelengths. This study provides a basis for further exploration into optical signaling in soft matter systems.
Significance: Spatial frequency domain imaging (SFDI) is a wide-field diffuse optical imaging technique for separately quantifying tissue reduced scattering (μs ′ ) and absorption (μa) coefficients at multiple wavelengths, providing wide potential utility for clinical applications such as burn wound characterization and cancer detection. However, measured μs ′ and μa can be confounded by absorption from melanin in patients with highly pigmented skin. This issue arises because epidermal melanin is highly absorbing for visible wavelengths and standard homogeneous light–tissue interaction models do not properly account for this complexity. Tristimulus colorimetry (which quantifies pigmentation using the L * “lightness” parameter) can provide a point of comparison between μa, μs ′ , and skin pigmentation.
Aim: We systematically compare SFDI and colorimetry parameters to quantify confounding effects of pigmentation on measured skin μs ′ and μa. We assess the correlation between SFDI and colorimetry parameters as a function of wavelength.
Approach: μs ′ and μa from the palm and ventral forearm were measured for 15 healthy subjects with a wide range of skin pigmentation levels (Fitzpatrick types I to VI) using a Reflect RS® (Modulim, Inc., Irvine, California) SFDI instrument (eight wavelengths, 471 to 851 nm). L * was measured using a Chroma Meter CR-400 (Konica Minolta Sensing, Inc., Tokyo). Linear correlation coefficients were calculated between L * and μs ′ and between L * and μa at all wavelengths.
Results: For the ventral forearm, strong linear correlations between measured L * and μs ′ values were observed at shorter wavelengths (R > 0.92 at ≤659 nm), where absorption from melanin confounded the measured μs ′ . These correlations were weaker for the palm (R < 0.59 at ≤659 nm), which has less melanin than the forearm. Similar relationships were observed between L * and μa.
Conclusions: We quantified the effects of epidermal melanin on skin μs ′ and μa measured with SFDI. This information may help characterize and correct pigmentation-related inaccuracies in SFDI skin measurements.
Significance: Quantitative measures of blood flow and metabolism are essential for improved assessment of brain health and response to ischemic injury.
Aim: We demonstrate a multimodal technique for measuring the cerebral metabolic rate of oxygen (CMRO2) in the rodent brain on an absolute scale (μM O2 / min).
Approach: We use laser speckle imaging at 809 nm and spatial frequency domain imaging at 655, 730, and 850 nm to obtain spatiotemporal maps of cerebral blood flow, tissue absorption (μa), and tissue scattering (μs ′ ). Knowledge of these three values enables calculation of a characteristic blood flow speed, which in turn is input to a mathematical model with a “zero-flow” boundary condition to calculate absolute CMRO2. We apply this method to a rat model of cardiac arrest (CA) and cardiopulmonary resuscitation. With this model, the zero-flow condition occurs during entry into CA.
Results: The CMRO2 values calculated with our method are in good agreement with those measured with magnetic resonance and positron emission tomography by other groups.
Conclusions: Our technique provides a quantitative metric of absolute cerebral metabolism that can potentially be used for comparison between animals and longitudinal monitoring of a single animal over multiple days. Though this report focuses on metabolism in a model of ischemia and reperfusion, this technique can potentially be applied to far broader types of acute brain injury and whole-body pathological occurrences.
KEYWORDS: Skin, Optical properties, In vivo imaging, Tissue optics, Scattering, Remote sensing, Range imaging, Monte Carlo methods, Mie scattering, Light wave propagation
Spatial frequency domain imaging (SFDI) is a wide-field spectral imaging technique that can be used to characterize optical properties of in-vivo tissue. Typically, SFDI uses light transport modeling based on Monte Carlo simulations to analyze the detected diffuse reflectance. Here, we examined the effect of using a semi-infinite homogeneous tissue model to determine optical properties of in-vivo human skin across a full range of pigmentation levels. We analyzed µs’ curves and performed correlation analysis between µs’ and degree of pigmentation determined using a tristimulus colorimeter. Our results suggested that pigmentation’s effect on µs’ is minimal at near-infrared wavelengths.
Spatial Frequency Domain Imaging (SFDI) is a non-invasive diffuse optical imaging technology that quantifies tissue optical properties by projecting spatially modulated light onto a region of interest. By detecting and fitting the diffuse reflectance to a light transport model, tissue absorption (μa) and scattering (μs') coefficients are calculated. However, when measuring skin, bulk μa and μs' from a homogeneous light transport model may not correspond with the actual properties of individual skin layers, especially for highly pigmented skin. To obtain physiologically accurate skin optical property values, we propose an iterative method based on a two- layered Monte Carlo model. We initially assume that μs' in darker skin is the same as that in lighter skin, and set the epidermal a as the free parameter when fitting the diffuse reflectance. To test this algorithm, we analyzed data from the forearms of 6 subjects having various levels of pigmentation, at 8 visible-to-near-infrared wavelengths. We compared the measured reflectance to both the homogeneous model and our layered model to quantify fit accuracy. At 471 nm and 526 nm for patients of Fitzpatrick skin types IV-VI (relatively dark skin), the two-layered model provided a 10-20% improvement in fit to the AC reflectance as a function of spatial frequency, compared to the homogeneous model. These improved fits yielded epidermal absorption coefficients that were notably higher than the bulk μa from the homogeneous model. Fitting the extracted epidermal μa to a melanin extinction spectrum1 enabled estimation of the melanin concentration in the epidermis.
We describe the development of a perturbation-free technique to measure tissue blood flow and metabolic rate of oxygen (MRO2) using LSI and SFDI. Analytical (diffusion) and computational (Monte Carlo) models are employed to characterize the contributions of diffuse and directed flow to the measured speckle contrast. Measured flow data is combined with the deoxygenated hemoglobin concentration and a path-length factor (both obtained from SFDI) to model tissue oxygen consumption in units of M O2/min. The results of this model were comparable with values of MRO2 measured with other imaging methods (PET, MRI/MRS).
Significance: Spatial frequency domain imaging (SFDI), a noncontact wide-field imaging technique using patterned illumination with multiple wavelengths, has been used to quantitatively measure structural and functional parameters of in vivo tissue. Using SFDI in a porcine model, we previously found that scattering changes in skin could potentially be used to noninvasively assess burn severity and monitor wound healing. Translating these findings to human subjects necessitates a better understanding of the variation in “baseline” human skin scattering properties across skin types and anatomical locations.
Aim: Using SFDI, we aim to characterize the variation in the reduced scattering coefficient (μs′) for skin across a range of pigmentation and anatomic sites (including common burn locations) for normal human subjects. These measurements are expected to characterize baseline human skin properties to inform our use of SFDI for clinical burn severity and wound healing assessments.
Approach: SFDI was used to measure μs′ in the visible- and near-infrared regime (471 to 851 nm) in 15 subjects at 10 anatomical locations. Subjects varied in age, gender, and Fitzpatrick skin type.
Results: For all anatomical locations, the coefficient of variation in measured μs′ decreased with increasing wavelength. High intersubject variation in μs′ at visible wavelengths coincided with large values of the melanin extinction coefficient at those wavelengths. At 851 nm, where intersubject variation in μs′ was smallest for all anatomical locations and absorption from melanin is minimal, significant intrasubject differences in μs′ were observed at the different anatomical locations.
Conclusions: Our study is the first report of wide-field mapping of human skin scattering properties across multiple skin types and anatomical locations using SFDI. Measured μs′ values varied notably between skin types at wavelengths where absorption from melanin was prominent. Additionally, μs′ varied considerably across different anatomical locations at 851 nm, where the confounding effects from melanin absorption are minimized.
Cardiac arrest (CA) affects over 500,000 people in the United States. Although resuscitation efforts have improved, poor neurological outcome is the leading cause of morbidity in CA survivors, and only 8.3% of out-of-hospital CA survivors have good neurological recovery. Therefore, a detailed understanding of the brain before, during, and after CA and resuscitation is critical. To provide a more complete picture of CBF dynamics associated with CA and resuscitation, we postulate that both temporal and spatial CBF dynamics must be understood. To investigate spatiotemporal dynamics, we used laser speckle imaging (LSI) to image rats that underwent either 5- or 7-min asphyxial CA, followed by cardiopulmonary resuscitation until return of spontaneous circulation (ROSC). During induction of global cerebral ischemia through CA, we observed two time periods during which a decrease in CBF propagates in space in a cranial window over the right hemisphere. The first time-period is during CA and the second after the hyperemic peak, but before CBF plateaus at a hypoperfused state post-ROSC. During CA, the decrease in CBF propagates from the lateral region of the brain to the medial region of the brain. Conversely, post-ROSC, the decrease in CBF propagates from the medial region of the brain to the lateral region of the brain. We postulate that study of spatiotemporal dynamics in a global cerebral ischemia model may lead to important insight into our understanding of cerebral function during and after resuscitation from CA, which may provide clinicians with knowledge that can lead to improvements in neurological outcome.
Quantifying rapidly varying perturbations in cerebral tissue absorption and scattering can potentially help to characterize changes in brain function caused by ischemic trauma. We have developed a platform for rapid intrinsic signal brain optical imaging using macroscopically structured light. The device performs fast, multispectral, spatial frequency domain imaging (SFDI), detecting backscattered light from three-phase binary square-wave projected patterns, which have a much higher refresh rate than sinusoidal patterns used in conventional SFDI. Although not as fast as “single-snapshot” spatial frequency methods that do not require three-phase projection, square-wave patterns allow accurate image demodulation in applications such as small animal imaging where the limited field of view does not allow single-phase demodulation. By using 655, 730, and 850 nm light-emitting diodes, two spatial frequencies (fx=0 and 0.3 mm−1), three spatial phases (120 deg, 240 deg, and 360 deg), and an overall camera acquisition rate of 167 Hz, we map changes in tissue absorption and reduced scattering parameters (μa and μs′) and oxy- and deoxyhemoglobin concentration at ∼14 Hz. We apply this method to a rat model of cardiac arrest (CA) and cardiopulmonary resuscitation (CPR) to quantify hemodynamics and scattering on temporal scales (Δt) ranging from tens of milliseconds to minutes. We observe rapid concurrent spatiotemporal changes in tissue oxygenation and scattering during CA and following CPR, even when the cerebral electrical signal is absent. We conclude that square-wave SFDI provides an effective technical strategy for assessing cortical optical and physiological properties by balancing competing performance demands for fast signal acquisition, small fields of view, and quantitative information content.
We introduce a tomographic approach for three-dimensional imaging of evoked hemodynamic activity, using broadband illumination and diffuse optical tomography (DOT) image reconstruction. Changes in diffuse reflectance in the rat somatosensory cortex due to stimulation of a single whisker were imaged at a frame rate of 5 Hz using a hyperspectral image mapping spectrometer. In each frame, images in 38 wavelength bands from 484 to 652 nm were acquired simultaneously. For data analysis, we developed a hyperspectral DOT algorithm that used the Rytov approximation to quantify changes in tissue concentration of oxyhemoglobin (ctHbO2) and deoxyhemoglobin (ctHb) in three dimensions. Using this algorithm, the maximum changes in ctHbO2 and ctHb were found to occur at 0.29±0.02 and 0.66±0.04 mm beneath the surface of the cortex, respectively. Rytov tomographic reconstructions revealed maximal spatially localized increases and decreases in ctHbO2 and ctHb of 321±53 and 555±96 nM, respectively, with these maximum changes occurring at 4±0.2 s poststimulus. The localized optical signals from the Rytov approximation were greater than those from modified Beer–Lambert, likely due in part to the inability of planar reflectance to account for partial volume effects.
KEYWORDS: Short wave infrared radiation, Absorption, Tissues, Tissue optics, Imaging spectroscopy, Scattering, Skin, In vivo imaging, Reflectivity, Near infrared
We present a review of short-wave infrared (SWIR, defined here as ∼1000 to 2000 nm) spectroscopy and imaging techniques for biological tissue optical property characterization. Studies indicate notable SWIR absorption features of tissue constituents including water (near 1150, 1450, and 1900 nm), lipids (near 1040, 1200, 1400, and 1700 nm), and collagen (near 1200 and 1500 nm) that are much more prominent than corresponding features observed in the visible and near-infrared (VIS-NIR, defined here as ∼400 to 1000 nm). Furthermore, the wavelength dependence of the scattering coefficient has been observed to follow a power-law decay from the VIS-NIR to the SWIR region. Thus, the magnitude of tissue scattering is lower at SWIR wavelengths than that observed at VIS or NIR wavelengths, potentially enabling increased penetration depth of incident light at SWIR wavelengths that are not highly absorbed by the aforementioned chromophores. These aspects of SWIR suggest that the tissue spectroscopy and imaging in this range of wavelengths have the potential to provide enhanced sensitivity (relative to VIS-NIR measurements) to chromophores such as water and lipids, thereby helping to characterize changes in the concentrations of these chromophores due to conditions such as atherosclerotic plaque, breast cancer, and burns.
Extending the wavelength range of spatial frequency domain imaging (SFDI) into the short-wave infrared (SWIR) has the potential to provide enhanced sensitivity to chromophores such as water and lipids that have prominent absorption features in the SWIR region. Here, we present, for the first time, a method combining SFDI with unstructured (zero spatial frequency) illumination to extract tissue absorption and scattering properties over a wavelength range (850 to 1800 nm) largely unexplored by previous tissue optics techniques. To obtain images over this wavelength range, we employ a SWIR camera in conjunction with an SFDI system. We use SFDI to obtain in vivo tissue reduced scattering coefficients at the wavelengths from 850 to 1050 nm, and then use unstructured wide-field illumination and an extrapolated power-law fit to this scattering spectrum to extract the absorption spectrum from 850 to 1800 nm. Our proof-of-principle experiment in a rat burn model illustrates that the combination of multispectral SWIR imaging, SFDI, and unstructured illumination can characterize in vivo changes in skin optical properties over a greatly expanded wavelength range. In the rat burn experiment, these changes (relative to normal, unburned skin) included increased absorption and increased scattering amplitude and slope, consistent with changes that we previously reported in the near-infrared using SFDI.
In this study, we describe a direct fit photon-tissue interaction model to quantitatively analyze reflectance spectra of biological tissue samples. The model rapidly extracts biologically-relevant parameters associated with tissue optical scattering and absorption. This model was employed to analyze reflectance spectra acquired from freshly excised human pancreatic pre-cancerous tissues (intraductal papillary mucinous neoplasm (IPMN), a common precursor lesion to pancreatic cancer). Compared to previously reported models, the direct fit model improved fit accuracy and speed. Thus, these results suggest that such models could serve as real-time, quantitative tools to characterize biological tissues assessed with reflectance spectroscopy.
KEYWORDS: Tumors, Tissues, Breast, Raman spectroscopy, Signal to noise ratio, Sensors, Natural surfaces, Tissue optics, Composites, Monte Carlo methods
The risk of local recurrence for breast cancers is strongly correlated with the presence of a tumor within 1 to 2 mm of the surgical margin on the excised specimen. Previous experimental and theoretical results suggest that spatially offset Raman spectroscopy (SORS) holds much promise for intraoperative margin analysis. Based on simulation predictions for signal-to-noise ratio differences among varying spatial offsets, a SORS probe with multiple source-detector offsets was designed and tested. It was then employed to acquire spectra from 35 frozen-thawed breast tissue samples in vitro. Spectra from each detector ring were averaged to create a composite spectrum with biochemical information covering the entire range from the tissue surface to ∼2 mm below the surface, and a probabilistic classification scheme was used to classify these composite spectra as "negative" or "positive" margins. This discrimination was performed with 95% sensitivity and 100% specificity, or with 100% positive predictive value and 94% negative predictive value.
There is a growing need for the development of computational models that can account for complex tissue morphology
in simulations of photon propagation. We describe the development and validation of a user-friendly, MATLAB-based
Monte Carlo code that uses analytically-defined surface meshes to model heterogeneous tissue geometry. The code can
use information from non-linear optical microscopy images to discriminate the fluorescence photons (from endogenous
or exogenous fluorophores) detected from different layers of complex turbid media. We present a specific application of
modeling a layered human tissue-engineered construct (Ex Vivo Produced Oral Mucosa Equivalent, EVPOME) designed
for use in repair of oral tissue following surgery. Second-harmonic generation microscopic imaging of an EVPOME
construct (oral keratinocytes atop a scaffold coated with human type IV collagen) was employed to determine an
approximate analytical expression for the complex shape of the interface between the two layers. This expression can
then be inserted into the code to correct the simulated fluorescence for the effect of the irregular tissue geometry.
Tissue-engineered constructs require noninvasive monitoring of cellular viability prior to implantation. In a preclinical
study on human Ex Vivo Produced Oral Mucosa Equivalent (EVPOME) constructs, nonlinear optical molecular imaging
was employed to extract morphological and functional information from intact constructs. Multiphoton excitation
fluorescence images were acquired using endogenous fluorescence from cellular nicotinamide adenine dinucleotide
phosphate [NAD(P)H] and flavin adenine dinucleotide (FAD). The images were analyzed to report quantitatively on
tissue structure and metabolism (redox ratio). Both thickness variations over time and cell distribution variations with
depth were identified, while changes in redox were quantified. Our results show that nonlinear optical molecular imaging
has the potential to visualize and quantitatively monitor the growth and viability of a tissue-engineered construct over
time.
Pancreatic adenocarcinoma has a five-year survival rate of only 6%, largely because current diagnostic methods cannot
reliably detect the disease in its early stages. Reflectance and fluorescence spectroscopies have the potential to provide
quantitative, minimally-invasive means of distinguishing pancreatic adenocarcinoma from normal pancreatic tissue and
chronic pancreatitis. The first collection of wavelength-resolved reflectance and fluorescence spectra and time-resolved
fluorescence decay curves from human pancreatic tissues was acquired with clinically-compatible instrumentation.
Mathematical models of reflectance and fluorescence extracted parameters related to tissue morphology and
biochemistry that were statistically significant for distinguishing between pancreatic tissue types. These results suggest
that optical spectroscopy has the potential to detect pancreatic disease in a clinical setting.
A specialized transient digitizer system was developed for spectroscopic collection of fluorescence wavelength-time
matrices (WTMs) from biological tissues. The system is compact, utilizes fiber optic probes for clinical compatibility,
and offers rapid collection of high signal-to-noise ratio (>100) time- and wavelength- resolved fluorescence. The system
is compatible with excitation sources operating in excess of 25 kHz. Wavelength-resolved measurement range is 300-800
nm with ≥ 0.01 nm steps. Time-resolved measurement depth is 128 ns with fixed 0.2 ns steps. The information-rich
WTM data provides comprehensive fluorescence sensing capabilities, as demonstrated on tissue simulating phantoms.
Extracting wavelength-resolved fluorophore lifetimes illustrates the potential of using the technology to resolve
exogenous or endogenous fluorophore contributions in tissue samples in a clinical setting for tissue diagnostics and
monitoring.
Polarized Raman spectroscopy allows measurement of molecular orientation and composition and is widely used in the study of polymer systems. Here, we extend the technique to the extraction of quantitative orientation information from bone tissue, which is optically thick and highly turbid. We discuss multiple scattering effects in tissue and show that repeated measurements using a series of objectives of differing numerical apertures can be employed to assess the contributions of sample turbidity and depth of field on polarized Raman measurements. A high numerical aperture objective minimizes the systematic errors introduced by multiple scattering. We test and validate the use of polarized Raman spectroscopy using wild-type and genetically modified (oim/oim model of osteogenesis imperfecta) murine bones. Mineral orientation distribution functions show that mineral crystallites are not as well aligned (p<0.05) in oim/oim bones (28±3 deg) compared to wild-type bones (22±3 deg), in agreement with small-angle X-ray scattering results. In wild-type mice, backbone carbonyl orientation is 76±2 deg and in oim/oim mice, it is 72±4 deg (p>0.05). We provide evidence that simultaneous quantitative measurements of mineral and collagen orientations on intact bone specimens are possible using polarized Raman spectroscopy.
Polarized Raman spectroscopy is widely used in the study of molecular composition and orientation in synthetic and
natural polymer systems. Here, we describe the use of Raman spectroscopy to extract quantitative orientation
information from bone tissue. Bone tissue poses special challenges to the use of polarized Raman spectroscopy for
measurement of orientation distribution functions because the tissue is turbid and birefringent. Multiple scattering in
turbid media depolarizes light and is potentially a source of error.
Using a Raman microprobe, we show that repeating the measurements with a series of objectives of differing numerical
apertures can be used to assess the contributions of sample turbidity and depth of field to the calculated orientation
distribution functions. With this test, an optic can be chosen to minimize the systematic errors introduced by multiple
scattering events. With adequate knowledge of the optical properties of these bone tissues, we can determine if elastic
light scattering affects the polarized Raman measurements.
We report data collected with a specialized transient digitizer, high repetition rate microchip laser sources, and fiber
optic light delivery and collection for rapid remote sensing in tissue-simulating phantoms. The instrumentation is highly
suitable for eventual translation to a clinical setting owing to the speed of data acquisition and small footprint. Ranges for
data acquisition time and instrument sensitivity were determined by measuring wavelength time matrices (WTMs) from
tissue-simulating phantoms. Accuracy of WTM data was validated by comparison with Monte-Carlo simulations of
fluorescent light propagation in turbid media.
A prototype clinical fluorescence and reflectance spectrometer was developed and employed to detect human pancreatic
adenocarcinoma. For the first time, quantitative pancreatic tissue models and chemometric algorithms were applied to
successfully distinguish among tissue types.
Pancreatic adenocarcinoma has a five-year survival rate of only 4%, largely because an effective procedure for early
detection has not been developed. In this study, mathematical modeling of reflectance and fluorescence spectra was
utilized to quantitatively characterize differences between normal pancreatic tissue, pancreatitis, and pancreatic
adenocarcinoma. Initial attempts at separating the spectra of different tissue types involved dividing fluorescence by
reflectance, and removing absorption artifacts by applying a "reverse Beer-Lambert factor" when the absorption
coefficient was modeled as a linear combination of the extinction coefficients of oxy- and deoxy-hemoglobin. These
procedures demonstrated the need for a more complete mathematical model to quantitatively describe fluorescence and
reflectance for minimally-invasive fiber-based optical diagnostics in the pancreas.
Monte Carlo (MC) simulations are considered the "gold standard" for mathematical description of photon transport in
tissue, but they can require large computation times. Therefore, it is important to develop simple and efficient methods
for accelerating MC simulations, especially when a large "library" of related simulations is needed. A semi-analytical
method involving MC simulations and a path-integral (PI) based scaling technique generated time-resolved reflectance
curves from layered tissue models. First, a zero-absorption MC simulation was run for a tissue model with fixed
scattering properties in each layer. Then, a closed-form expression for the average classical path of a photon in tissue
was used to determine the percentage of time that the photon spent in each layer, to create a weighted Beer-Lambert
factor to scale the time-resolved reflectance of the simulated
zero-absorption tissue model. This method is a unique
alternative to other scaling techniques in that it does not require the path length or number of collisions of each photon to
be stored during the initial simulation. Effects of various layer thicknesses and absorption and scattering coefficients on
the accuracy of the method will be discussed.
Light-scattering spectroscopy has the potential to provide information about bone composition via a fiber-optic probe
placed on the skin. In order to design efficient probes, one must understand the effect of all tissue layers on photon
transport. To quantitatively understand the effect of overlying tissue layers on the detected bone Raman signal, a layered
Monte Carlo model was modified for Raman scattering. The model incorporated the absorption and scattering properties
of three overlying tissue layers (dermis, subdermis, muscle), as well as the underlying bone tissue. The attenuation of the
collected bone Raman signal, predominantly due to elastic light scattering in the overlying tissue layers, affected the
carbonate/phosphate (C/P) ratio by increasing the standard deviation of the computational result. Furthermore, the mean
C/P ratio varied when the relative thicknesses of the layers were varied and the elastic scattering coefficient at the
Raman scattering wavelength of carbonate was modeled to be different from that at the Raman scattering wavelength of
phosphate. These results represent the first portion of a computational study designed to predict optimal probe geometry
and help to analyze detected signal for Raman scattering experiments involving bone.
Photon transport in complex biological tissues is most accurately predicted via Monte Carlo (MC)
modeling methods, which often require lengthy computation times. In this report, a semi-analytical
technique (henceforth referred to as PI-scaling) was derived that combines MC simulation, absorption
scaling, and path integrals (PI) to rapidly reconstruct time-resolved reflectance from the surface of bilayered
epithelial tissue models. Comparisons to forward MC simulations indicated that the PI−scaling
method was accurate to better than 10% for tissue models where the optical properties of the
top layer did not greatly influence the time-resolved reflectance. Employing such a method should
provide a novel solution to the first step of the problem of rapid simulation of time-resolved
reflectance of photons in layered tissues.
Tissue engineered constructs can be employed to graft wounds or replace diseased tissue. Non-invasive methods are
required to assess cellular viability in these constructs both pre- and post-implantation into patients. In this study, Monte
Carlo simulations and fluorescence experiments were executed on ex vivo produced oral mucosa equivalent (EVPOME)
constructs to investigate the fluorescence emitted at 355 nm excitation from these constructs. Both simulations and
experiments indicated the need to investigate alternative excitation wavelengths in order to increase the cellular
fluorescence from these constructs, while decreasing contributions from extra-cellular fluorophores.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.