A color enhancement method to optimize the visualization of breast tumors in cancer pathology is proposed. Light scattering measurements are minimally invasive, and allow the estimation of tissue morphology and composition to guide the surgeon in resection surgeries. The usability of scatter and absorption signatures acquired with a microsampling reflectance spectral imaging system was improved employing an empirical approximation to the Mie theory to estimate the scattering power on a per-pixel basis. The proposed methodology generates a new image with blended color and diagnostic purposes coming from the emphasis or highlighting of specific wavelengths or features. These features can be the specific absorbent tissue components (oxygenated and deoxygenated hemoglobin, etc.), additional parameters as scattering power or amplitude or even the combination of both. The goal is to obtain an improved and inherent tissue contrast working only with the local reflectance of tissue. To this aim, it is provided a visual interpretation of what is considered non-malignant (normal epithelia and stroma, benign epithelia and stroma, inflammation), malignant (DCIS, IDC, ILC) and adipose tissue. Consequently, a fast visualization map of the intracavity area can be offered to the surgeon providing relevant diagnostic information. No labeling or extrinsic indicators are required for proposed methodology and therefore the possibility of transferring absorption and scattering features simultaneously into visualization, fusing their effects into a single image, can guide surgeons efficiently.
The feasibility of spatial frequency domain imaging (SFDI) for breast surgical margin assessment was evaluated in tissue-simulating phantoms and in fully intact lumpectomy specimens at the time of surgery. Phantom data was evaluated according to contrast-detail resolution, quantitative accuracy and model-data goodness of fit, where optical parameters were estimated by minimizing the residual sum of squares between the measured modulation amplitude and its solutions, modeled according to diffusion and scaled-Monte Carlo simulations. In contrast-detail phantoms, a 1.25-mm-diameter surface inclusion was detectable for scattering contrast >28% ; a fraction of this scattering contrast (7%) was detectable for a 10 mm surface inclusion and at least 33% scattering contrast was detected up to 1.5 mm below the phantom surface, a probing depth relevant to breast surgical margin assessment. Recovered hemoglobin concentrations were insensitive to changes in scattering, except for overestimation at visible wavelengths for total hemoglobin concentrations <15 μM . The scattering amplitude increased linearly with scattering concentration, but the scattering slope depended on both the particle size and number density. Goodness of fit was comparable for the diffusion and scaled-Monte Carlo models of transport in spatially modulated, near-infrared reflectance acquired from 47 lumpectomy tissues, but recovered absorption parameters varied more linearly with expected hemoglobin concentration in liquid phantoms for the scaled-Monte Carlo forward model. SFDI could potentially reduce the high secondary excision rate associated with breast conserving surgery; its clinical translation further requires reduced image reconstruction time and smart inking strategies.
Fractal analysis combined with a label-free scattering technique is proposed for describing the pathological architecture
of tumors. Clinicians and pathologists are conventionally trained to classify abnormal features such as structural
irregularities or high indices of mitosis. The potential of fractal analysis lies in the fact of being a morphometric measure
of the irregular structures providing a measure of the object’s complexity and self-similarity. As cancer is characterized
by disorder and irregularity in tissues, this measure could be related to tumor growth. Fractal analysis has been probed in
the understanding of the tumor vasculature network. This work addresses the feasibility of applying fractal analysis to
the scattering power map (as a physical modeling) and principal components (as a statistical modeling) provided by a
localized reflectance spectroscopic system. Disorder, irregularity and cell size variation in tissue samples is translated
into the scattering power and principal components magnitude and its fractal dimension is correlated with the pathologist
assessment of the samples. The fractal dimension is computed applying the box-counting technique. Results show that
fractal analysis of ex-vivo fresh tissue samples exhibits separated ranges of fractal dimension that could help classifier
combining the fractal results with other morphological features. This contrast trend would help in the discrimination of
tissues in the intraoperative context and may serve as a useful adjunct to surgeons.
Texture analysis of light scattering in tissue is proposed to obtain diagnostic information from breast cancer specimens. Light scattering measurements are minimally invasive, and allow the estimation of tissue morphology to guide the surgeon in resection surgeries. The usability of scatter signatures acquired with a micro-sampling reflectance spectral imaging system was improved utilizing an empirical approximation to the Mie theory to estimate the scattering power on a per-pixel basis. Co-occurrence analysis is then applied to the scattering power images to extract the textural features. A statistical analysis of the features demonstrated the suitability of the autocorrelation for the classification of notmalignant (normal epithelia and stroma, benign epithelia and stroma, inflammation), malignant (DCIS, IDC, ILC) and adipose tissue, since it reveals morphological information of tissue. Non-malignant tissue shows higher autocorrelation values while adipose tissue presents a very low autocorrelation on its scatter texture, being malignant the middle ground. Consequently, a fast linear classifier based on the consideration of just one straightforward feature is enough for providing relevant diagnostic information. A leave-one-out validation of the linear classifier on 29 samples with 48 regions of interest showed classification accuracies of 98.74% on adipose tissue, 82.67% on non-malignant tissue and 72.37% on malignant tissue, in comparison with the biopsy H and E gold standard. This demonstrates that autocorrelation analysis of scatter signatures is a very computationally efficient and automated approach to provide pathological information in real-time to guide surgeon during tissue resection.
A blind separation technique based on Independent Component Analysis (ICA) is proposed for breast tumor delineation
and pathologic diagnosis. Tissue morphology is determined by fitting local measures of tissue reflectance to a Mie
theory approximation, parameterizing the scattering power, scattering amplitude and average scattering irradiance. ICA
is applied on the scattering parameters by spatial analysis using the Fast ICA method to extract more determinant
features for an accurate diagnostic. Neither training, nor comparisons with reference parameters are required. Tissue
diagnosis is provided directly following ICA application to the scattering parameter images. Surgically resected breast
tissues were imaged and identified by a pathologist. Three different tissue pathologies were identified in 29 samples and
classified as not-malignant, malignant and adipose. Scatter plot analysis of both ICA results and optical parameters
where obtained. ICA subtle ameliorates those cases where optical parameter's scatter plots were not linearly separable.
Furthermore, observing the mixing matrix of the ICA, it can be decided when the optical parameters themselves are
diagnostically powerful. Moreover, contrast maps provided by ICA correlate with the pathologic diagnosis. The time
response of the diagnostic strategy is therefore enhanced comparing with complex classifiers, enabling near real-time
assessment of pathology during breast-conserving surgery.
A spectral analysis technique to enhance tumor contrast during breast conserving surgery is proposed. A set of 29
surgically-excised breast tissues have been imaged in local reflectance geometry. Measures of broadband reflectance are
directly analyzed using Principle Component Analysis (PCA), on a per sample basis, to extract areas of maximal spectral
variation. A dynamic selection threshold has been applied to obtain the final number of principal components,
accounting for inter-patient variability. A blind separation technique based on Independent Component Analysis (ICA) is
then applied to extract diagnostically powerful results. ICA application reveals that the behavior of one independent
component highly correlates with the pathologic diagnosis and it surpasses the contrast obtained using empirical models.
Moreover, blind detection characteristics (no training, no comparisons with training reference data) and no need for
parameterization makes the automated diagnosis simple and time efficient, favoring its translation to the clinical
practice. Correlation coefficient with model-based results up to 0.91 has been achieved.
Tissue ultra-structure and molecular composition provide native contrast mechanisms for discriminating across pathologically distinct tissue-types. Multi-modality optical probe designs combined with spatially confined sampling techniques have been shown to be sensitive to this type of contrast but their extension to imaging has only been realized recently. A modular scanning spectroscopy platform has been developed to allow imaging localized morphology and molecular contrast measures in breast cancer surgical specimens. A custom designed dark-field telecentric scanning spectroscopy system forms the core of this imaging platform. The system allows imaging localized elastic scatter and fluorescence measures over fields of up to 15 mm x 15 mm at 100 microns resolution in tissue. Results from intralipid and blood phantom measurements demonstrate the ability of the system to quantify localized scatter parameters despite significant changes in local absorption. A co-registered fluorescence spectroscopy mode is also demonstrated in a protophorphyrin-IX phantom.
Multi-spectral spatially modulated light is used to guide localized spectroscopy of surgically resected tissues for cancer
involvement. Modulated imaging rapidly quantifies near-infrared optical parameters with sub-millimeter resolution over
the entire field for identification of residual disease in resected tissues. Suspicious lesions are further evaluated using a
spectroscopy platform designed to image thick tissue samples at a spatial resolution sensitive to the diagnostic gold
standard, pathology. MI employs a spatial frequency domain sampling and model-based analysis of the spatial
modulation transfer function to interpret a tissue's absorption and scattering parameters at depth. The spectroscopy
platform employs a scanning-beam, telecentric dark-field illumination and confocal detection to image fields up to 1cm2
with a broadband source (480:750nm). The sampling spot size (100μm lateral resolution) confines the volume of tissue
probed to within a few transport pathlengths so that multiple-scattering effects are minimized and simple empirical
models may be used to analyze spectra. Localized spectroscopy of Intralipid and hemoglobin phantoms demonstrate
insensitivity of recovered scattering parameters to changes in absorption, but a non-linear dependence of scattering
power on Intralipid concentration is observed due to the phase sensitivity of the measurement system. Both systems were
validated independently in phantom and murine studies. Ongoing work focuses on assessing the combined utility of
these systems to identify cancer involvement in vitro, particularly in the margins of resected breast tumors.
Biophysical changes such as inflammation and necrosis occur immediately following PDT and may be used to
assess the treatment response to PDT treatment in-vivo. This study uses localized reflectance measurements to quantify
the scatter changes in tumor tissue occurring in response to verteporfin-based PDT treatment in xenograft pancreas
tumors. Nude mice were implanted with subcutaneous AsPC-1 pancreatic tumors cells in matrigel, and allowed to
establish solid tumors near 100mm3 volume. The mice were sensitized with 1mg/kg of the active component of
verteporfin (benzoporphryin derivative, BPD), one hour before light delivery. The optical irradiation was performed
using a 1 cm cylindrical interstitial diffusing tip fiber with 20J of red light (690nm). Tumor tissue was excised
progressively and imaged, from 1 day to 4 weeks, after PDT treatment. The tissue sections were stained and analyzed by
an expert veterinary pathologist, who provided information on tissue regions of interest. This information was correlated
with variations in scattering and absorption parameters elucidated from the spectral images and the degree of necrosis
and inflammation involvement was identified.
Areas of necrosis and dead cells exhibited the lowest average scatter irradiance signature (3.78 and 4.07
respectively) compared to areas of viable pancreatic tumor cells and areas of inflammation (5.81 and 7.19 respectively).
Bilirubin absorbance parameters also showed a lower absorbance value in necrotic tissue and areas of dead cells (0.05
and 0.1 respectively) compared to tissue areas for viable pancreatic tumor cells and areas of inflammation (0.28 and
0.35). These results demonstrate that localized reflectance spectroscopy is an imaging modality that can be used to
identify tissue features associated with PDT treatment (e.g. necrosis and inflammation) that can be correlated with
histopathologically-reviewed H&E stained slides. Further study of this technique may provide means for automated
discrimination of tissue features based on scatter and absorbance maps elucidated from reflectance spectral datasets and
provide a valuable tool for treatment response monitoring during PDT and enabling more effective treatment planning.
These results are relevant to verteporfin-based PDT trial for treatment pancreatic cancer in non-surgical
candidate cases (VERTPAC-1 University College London, PI Pereira), where individualized assessment of damage and
response could be beneficial, if this study is proven to be a well-controlled imaging tool.
We demonstrate that morphological features pertinent to a tissue's pathology may be ascertained from localized measures of broadband reflectance, with a mesoscopic resolution (100-μm lateral spot size) that permits scanning of an entire margin for residual disease. The technical aspects and optimization of a k-nearest neighbor classifier for automated diagnosis of pathologies are presented, and its efficacy is validated in 29 breast tissue specimens. When discriminating between benign and malignant pathologies, a sensitivity and specificity of 91 and 77% was achieved. Furthermore, detailed subtissue-type analysis was performed to consider how diverse pathologies influence scattering response and overall classification efficacy. The increased sensitivity of this technique may render it useful to guide the surgeon or pathologist where to sample pathology for microscopic assessment.
Infiltrating neoplastic epithelia induce ultra-structure changes in tissue providing an intrinsic contrast in terms of their
local light scattering response. Imaging systems that can enhance this contrast allow for better visualization of tumor
boundaries and thus have enormous potential in guiding complex surgical procedures like breast lumpectomy. Highly
localized reflectance measurement probes can quantify scattering changes in tissues in situ, but in order to be useful in
surgical settings these techniques require an extension to imaging. A novel microsampling reflectance imaging system
has been developed to allow rapid quantitative imaging of ultra-structure associated scattering changes in tissues in situ.
The imaging system is described in terms of its design, construction and testing for multi-wavelength, telecentric, darkfield
illumination and confocal spectroscopic detection, with imaging fields of up to 1.5 cm × 1.5 cm at 100 microns
resolution. Spatial confinement of the incident and detected light allows for direct sampling of the scattering spectrum in
tissues in situ and the telecentric design ensures consistent sampling of the scattering phase function throughout the
entire imaging field. The imaging system was modeled and optimized using the ZEMAX optical design software.
Description of the design and results from the optimization process are presented.
An NIRS tomography system that can simultaneously acquire data at three wavelengths has
been developed to measure changes in physiological properties with 15 second time resolution. The
results of homogenous and heterogeneous blood phantom studies indicated that the R2 values
between average estimated total hemoglobin (HbT) values and blood concentrations are 0.99 and
0.9, respectively. In preliminary normal subject clinical trials, a cohort of normal subjects were
tested by acquiring the series images as well as the pressure is adding to and releasing from the
breast. The recovered data shown that by adding measurable pressure, HbT is reduced and the
maximum HbT reduction is correlated to the Body Mass Index.
The purpose of this study was to extract scatter parameters related to tissue ultra-structures from freshly excised breast
tissue and to assess whether evident changes in scatter across diagnostic categories is primarily influenced by variation
in the composition of each tissues subtypes or by physical remodeling of the extra-cellular environment. Pathologists
easily distinguish between epithelium, stroma and adipose tissues, so this classification was adopted for macroscopic
subtype classification. Micro-sampling reflectance spectroscopy was used to characterize single-backscattered photons
from fresh, excised tumors and normal reduction specimens with
sub-millimeter resolution. Phase contrast microscopy
(sub-micron resolution) was used to characterize forward-scattered light through frozen tissue from the DHMC Tissue
Bank, representing normal, benign and malignant breast tissue, sectioned at 10 microns. The packing density and
orientation of collagen fibers in the extracellular matrix (ECM) associated with invasive, normal and benign epithelium
was evaluated using transmission electron microscopy (TEM). Regions of interest (ROIs) in the H&E stained tissues
were identified for analysis, as outlined by a pathologist as the gold standard. We conclude that the scatter parameters
associated with tumor specimens (Npatients=6, Nspecimens=13) significantly differs from that of normal reductions
(Npatients=6, Nspecimens=10). Further, tissue subtypes may be identified by their scatter spectra at sub-micron resolution.
Stromal tissue scatters significantly more than the epithelial cells embedded in its ECM and adipose tissue scatters much
less. However, the scatter signature of the stroma at the sub-micron level is not particularly differentiating in terms of a
diagnosis.
Video rate diffuse tomography can be implemented within the magnetic resonance breast exam. The following paper outlines the basics of a spectrally encoded source set up, being designed and tested for use in breast imaging within a specialized breast surface coil. The system design maximizes input power to the breast, while confining the spectrum to a 10 nm bandwidth of near-infrared light. The center spectral band can be varied, since it is supplied by a tunable Ti:Sapphire laser. The encoding of each source is achieved by splitting the signal into individual nanometer bands through a high resolution grating, and focusing the output of this into each source fiber. This source configuration then requires spectral detection at the output, and so each detection fiber is delivered to a high resolution spectrometer to resolve the detected intensities. Breast imaging with this system has some subtle dynamic range issues, which means that light from sources farthest from the detector pickup are likely not providing useful data, but the closest 4-6 fibers near each source can provide useful data. The implementation of this is being carried out within a magnetic resonance breast array, and initial testing of the signals is shown, along with diagrams and photographs of the system configuration.
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.