Quantitative imaging of scattering changes associated with epithelial proliferation, necrosis, and fibrosis in tumors using microsampling reflectance spectroscopy

Dual modal spectroscopic tissue scanner for colorectal cancer diagnosis

BACKGROUND

Margin status is an important prognostic factor for treating colorectal cancer. This study aimed to investigate the usefulness of a multimodal spectroscopic tissue scanner for real-time cancer diagnosis without tissue staining.

PATIENTS AND METHODS

Diffuse reflectance spectra (DRS) and fluorescence spectra (FS) of < 1-mm-sized paired cancer and normal mucosa tissue were acquired using custom-built spectroscopic tissue scanners. For FS, we analyzed wavelengths and intensities at peaks and highest intensities near (± 1.25 nm) the known fluorescence spectral peaks of collagen (380 nm), reduced nicotinamide adenine dinucleotide (NADH, 460 nm), and flavin adenine dinucleotide (FAD, 550 nm). For DRS, we performed a similar analysis near the peaks of strong absorbers, oxyhemoglobin (oxyHb; 414 nm, 540 nm, and 576 nm) and deoxyhemoglobin (deoxyHb; 432 nm and 556 nm). Logistic regression analysis for these parameters was performed in the testing set.

RESULTS

We acquired 17,735 spectra of cancer tissues and 9438 of normal tissues from 30 patients. Intensity peaks of representative normal spectra for FS and DRS were higher than those of representative cancer spectra. Logistic regression analysis showed wavelength and intensity at peaks, and the intensities of the peak wavelength of NADH, FAD, deoxyHb, and oxyHb had significant coefficients. The area under the receiver operating characteristic curve was 0.927. The scanner had 100%, 64.3%, and 85.3% sensitivity, specificity, and accuracy, respectively.

CONCLUSIONS

The spectroscopic tissue scanner has high sensitivity and accuracy and provides real-time intraoperative resection margin assessments and should be further investigated as an alternative to frozen section.

On the spectral signature of melanoma: a non-parametric classification framework for cancer detection in hyperspectral imaging of melanocytic lesions

Early detection and diagnosis is a must in secondary prevention of melanoma and other cancerous lesions of the skin. In this work, we present an online, reservoir-based, non-parametric estimation and classification model that allows for this functionality on pigmented lesions, such that detection thresholding can be tuned to maximize accuracy and/or minimize overall false negative rates. This system has been tested in a dataset consisting of 116 patients and a total of 124 hyperspectral images of nevi, raised nevi and melanomas, detecting up to 100% of the suspicious lesions at the expense of some false positives.

Directional Kernel Density Estimation for Classification of Breast Tissue Spectra

In Breast Conserving Therapy, surgeons measure the thickness of healthy tissue surrounding an excised tumor (surgical margin) via post-operative histological or visual assessment tests that, for lack of enough standardization and reliability, have recurrence rates in the order of 33%. Spectroscopic interrogation of these margins is possible during surgery, but algorithms are needed for parametric or dimension reduction processing. One methodology for tumor discrimination based on dimensionality reduction and nonparametric estimation-in particular, Directional Kernel Density Estimation-is proposed and tested on spectral image data from breast samples. Once a hyperspectral image of the tumor has been captured, a surgeon assists by establishing Regions of Interest where tissues are qualitatively differentiable. After proper normalization, Directional KDE is used to estimate the likelihood of every pixel in the image belonging to each specified tissue class. This information is enough to yield, in almost real time and with 98% accuracy, results that coincide with those provided by histological H&E validation performed after the surgery.

Uniqueness in multispectral constant-wave epi-illumination imaging

Multispectral tissue imaging based on optical cameras and continuous-wave tissue illumination is commonly used in medicine and biology. Surprisingly, there is a characteristic absence of a critical look at the quantities that can be uniquely characterized from optically diffuse matter by multispectral imaging. Here, we investigate the fundamental question of uniqueness in epi-illumination measurements from turbid media obtained at multiple wavelengths. By utilizing an analytical model, tissue-mimicking phantoms, and an in vivo imaging experiment we show that independent of the bands employed, spectral measurements cannot uniquely retrieve absorption and scattering coefficients. We also establish that it is, nevertheless, possible to uniquely quantify oxygen saturation and the Mie scattering power-a previously undocumented uniqueness condition.

Low-cost tissue simulating phantoms with adjustable wavelength-dependent scattering properties in the visible and infrared ranges

We present a method for low-cost fabrication of polydimethylsiloxane (PDMS) tissue simulating phantoms with tunable scattering spectra, spanning visible, and near-infrared regimes. These phantoms use optical polishing agents (aluminum oxide powders) at various grit sizes to approximate in vivo tissue scattering particles across multiple size distributions (range: 17 to 3  μm). This class of tunable scattering phantoms is used to mimic distinct changes in wavelength-dependent scattering properties observed in tissue pathologies such as partial thickness burns. Described by a power-law dependence on wavelength, the scattering magnitude of these phantoms scale linearly with particle concentration over a physiologic range [μs′=(0.5 to 2.0  mm−1)] whereas the scattering spectra, specific to each particle size distribution, correlate to distinct exponential coefficients (range: 0.007 to 0.32). Aluminum oxide powders used in this investigation did not detectably contribute to the absorption properties of these phantoms. The optical properties of these phantoms are verified through inverse adding-doubling methods and the tolerances of this fabrication method are discussed.

Molecular dyes used for surgical specimen margin orientation allow for intraoperative optical assessment during breast conserving surgery

A variety of optical techniques utilizing near-infrared (NIR) light are being proposed for intraoperative breast tumor margin assessment. However, immediately following a lumpectomy excision, the margins are inked, which preserves the orientation of the specimen but prevents optical interrogation of the tissue margins. Here, a workflow is proposed that allows for both NIR optical assessment following full specimen marking using molecular dyes which have negligible absorption and scattering in the NIR. The effect of standard surgical inks in contrast to molecular dyes for an NIR signal is shown. Further, the proposed workflow is demonstrated with full specimen intraoperative imaging on all margins directly after the lumpectomy has been excised and completely marked. This work is an important step in the path to clinical feasibility of intraoperative breast tumor margin assessment using NIR optical methods without having to compromise on the current clinical practice of inking resected specimens for margin orientation.

Spectral discrimination of breast pathologies in situ using spatial frequency domain imaging

Introduction

Nationally, 25% to 50% of patients undergoing lumpectomy for local management of breast cancer require a secondary excision because of the persistence of residual tumor. Intraoperative assessment of specimen margins by frozen-section analysis is not widely adopted in breast-conserving surgery. Here, a new approach to wide-field optical imaging of breast pathology in situ was tested to determine whether the system could accurately discriminate cancer from benign tissues before routine pathological processing.

Methods

Spatial frequency domain imaging (SFDI) was used to quantify near-infrared (NIR) optical parameters at the surface of 47 lumpectomy tissue specimens. Spatial frequency and wavelength-dependent reflectance spectra were parameterized with matched simulations of light transport. Spectral images were co-registered to histopathology in adjacent, stained sections of the tissue, cut in the geometry imaged in situ. A supervised classifier and feature-selection algorithm were implemented to automate discrimination of breast pathologies and to rank the contribution of each parameter to a diagnosis.

Results

Spectral parameters distinguished all pathology subtypes with 82% accuracy and benign (fibrocystic disease, fibroadenoma) from malignant (DCIS, invasive cancer, and partially treated invasive cancer after neoadjuvant chemotherapy) pathologies with 88% accuracy, high specificity (93%), and reasonable sensitivity (79%). Although spectral absorption and scattering features were essential components of the discriminant classifier, scattering exhibited lower variance and contributed most to tissue-type separation. The scattering slope was sensitive to stromal and epithelial distributions measured with quantitative immunohistochemistry.

Conclusions

SFDI is a new quantitative imaging technique that renders a specific tissue-type diagnosis. Its combination of planar sampling and frequency-dependent depth sensing is clinically pragmatic and appropriate for breast surgical-margin assessment. This study is the first to apply SFDI to pathology discrimination in surgical breast tissues. It represents an important step toward imaging surgical specimens immediately ex vivo to reduce the high rate of secondary excisions associated with breast lumpectomy procedures.

Sub-diffusive scattering parameter maps recovered using wide-field high-frequency structured light imaging

This study investigates the hypothesis that structured light reflectance imaging with high spatial frequency patterns [Formula: see text] can be used to quantitatively map the anisotropic scattering phase function distribution [Formula: see text] in turbid media. Monte Carlo simulations were used in part to establish a semi-empirical model of demodulated reflectance ([Formula: see text]) in terms of dimensionless scattering [Formula: see text] and [Formula: see text], a metric of the first two moments of the [Formula: see text] distribution. Experiments completed in tissue-simulating phantoms showed that simultaneous analysis of [Formula: see text] spectra sampled at multiple [Formula: see text] in the frequency range [0.05-0.5] [Formula: see text] allowed accurate estimation of both [Formula: see text] in the relevant tissue range [0.4-1.8] [Formula: see text], and [Formula: see text] in the range [1.4-1.75]. Pilot measurements of a healthy volunteer exhibited [Formula: see text]-based contrast between scar tissue and surrounding normal skin, which was not as apparent in wide field diffuse imaging. These results represent the first wide-field maps to quantify sub-diffuse scattering parameters, which are sensitive to sub-microscopic tissue structures and composition, and therefore, offer potential for fast diagnostic imaging of ultrastructure on a size scale that is relevant to surgical applications.

Simultaneous multiplane imaging of human ovarian cancer by volume holographic imaging

Ovarian cancer is the most deadly gynecologic cancer, a fact which is attributable to poor early detection and survival once the disease has reached advanced stages. Intraoperative laparoscopic volume holographic imaging has the potential to provide simultaneous visualization of surface and subsurface structures in ovarian tissues for improved assessment of developing ovarian cancer. In this ex vivo ovarian tissue study, we assembled a benchtop volume holographic imaging system (VHIS) to characterize the microarchitecture of 78 normal and 40 abnormal tissue specimens derived from ovarian, fallopian tube, uterine, and peritoneal tissues, collected from 26 patients aged 22 to 73 undergoing bilateral salpingo-oophorectomy, hysterectomy with bilateral salpingo-oophorectomy, or abdominal cytoreductive surgery. All tissues were successfully imaged with the VHIS in both reflectance- and fluorescence-modes revealing morphological features which can be used to distinguish between normal, benign abnormalities, and cancerous tissues. We present the development and successful application of VHIS for imaging human ovarian tissue. Comparison of VHIS images with corresponding histopathology allowed for qualitatively distinguishing microstructural features unique to the studied tissue type and disease state. These results motivate the development of a laparoscopic VHIS for evaluating the surface and subsurface morphological alterations in ovarian cancer pathogenesis.

Optimization of a widefield structured illumination microscope for non-destructive assessment and quantification of nuclear features in tumor margins of a primary mouse model of sarcoma

Cancer is associated with specific cellular morphological changes, such as increased nuclear size and crowding from rapidly proliferating cells. In situ tissue imaging using fluorescent stains may be useful for intraoperative detection of residual cancer in surgical tumor margins. We developed a widefield fluorescence structured illumination microscope (SIM) system with a single-shot FOV of 2.1×1.6 mm (3.4 mm²) and sub-cellular resolution (4.4 µm). The objectives of this work were to measure the relationship between illumination pattern frequency and optical sectioning strength and signal-to-noise ratio in turbid (i.e. thick) samples for selection of the optimum frequency, and to determine feasibility for detecting residual cancer on tumor resection margins, using a genetically engineered primary mouse model of sarcoma. The SIM system was tested in tissue mimicking solid phantoms with various scattering levels to determine impact of both turbidity and illumination frequency on two SIM metrics, optical section thickness and modulation depth. To demonstrate preclinical feasibility, ex vivo 50 µm frozen sections and fresh intact thick tissue samples excised from a primary mouse model of sarcoma were stained with acridine orange, which stains cell nuclei, skeletal muscle, and collagenous stroma. The cell nuclei were segmented using a high-pass filter algorithm, which allowed quantification of nuclear density. The results showed that the optimal illumination frequency was 31.7 µm⁻¹ used in conjunction with a 4×0.1 NA objective ( = 0.165). This yielded an optical section thickness of 128 µm and an 8.9×contrast enhancement over uniform illumination. We successfully demonstrated the ability to resolve cell nuclei in situ achieved via SIM, which allowed segmentation of nuclei from heterogeneous tissues in the presence of considerable background fluorescence. Specifically, we demonstrate that optical sectioning of fresh intact thick tissues performed equivalently in regards to nuclear density quantification, to physical frozen sectioning and standard microscopy.

Direct identification of breast cancer pathologies using blind separation of label-free localized reflectance measurements

Breast tumors are blindly identified using Principal (PCA) and Independent Component Analysis (ICA) of localized reflectance measurements. No assumption of a particular theoretical model for the reflectance needs to be made, while the resulting features are proven to have discriminative power of breast pathologies. Normal, benign and malignant breast tissue types in lumpectomy specimens were imaged ex vivo and a surgeon-guided calibration of the system is proposed to overcome the limitations of the blind analysis. A simple, fast and linear classifier has been proposed where no training information is required for the diagnosis. A set of 29 breast tissue specimens have been diagnosed with a sensitivity of 96% and specificity of 95% when discriminating benign from malignant pathologies. The proposed hybrid combination PCA-ICA enhanced diagnostic discrimination, providing tumor probability maps, and intermediate PCA parameters reflected tissue optical properties.

Radial angular filter arrays for angle-resolved scattering spectroscopy

The radial angular filter array (RAFA) consists of a series of radially-distributed micro-machined channels, where the long axes of the channels converge at a focal point. The high aspect ratio of each channel provides a means to reject photons with trajectories outside the acceptance angle of the channel. The output of the RAFA represents the angular distribution of photons emitted from the focal point. A series of RAFAs were designed, fabricated, and tested to evaluate the impact of device geometry, inter-channel cross talk, achromaticity, and channel leakage on device performance. As an application example, an RAFA was used together with an imaging spectrometer to capture angle-resolved spectra of turbid samples.

Scanning in situ spectroscopy platform for imaging surgical breast tissue specimens

A non-contact localized spectroscopic imaging platform has been developed and optimized to scan 1 x 1 cm² square regions of surgically resected breast tissue specimens with ~150-micron resolution. A color corrected, image-space telecentric scanning design maintained a consistent sampling geometry and uniform spot size across the entire imaging field. Theoretical modeling in ZEMAX allowed estimation of the spot size, which is equal at both the center and extreme positions of the field with ~5% variation across the designed waveband, indicating excellent color correction. The spot sizes at the center and an extreme field position were also measured experimentally using the standard knife-edge technique and were found to be within ~8% of the theoretical predictions. Highly localized sampling offered inherent insensitivity to variations in background absorption allowing direct imaging of local scattering parameters, which was validated using a matrix of varying concentrations of Intralipid and blood in phantoms. Four representative, pathologically distinct lumpectomy tissue specimens were imaged, capturing natural variations in tissue scattering response within a given pathology. Variations as high as 60% were observed in the average reflectance and relative scattering power images, which must be taken into account for robust classification performance. Despite this variation, the preliminary data indicates discernible scatter power contrast between the benign vs malignant groups, but reliable discrimination of pathologies within these groups would require investigation into additional contrast mechanisms.

Scatter spectroscopic imaging distinguishes between breast pathologies in tissues relevant to surgical margin assessment

PURPOSE

A new approach to spectroscopic imaging was developed to detect and discriminate microscopic pathologies in resected breast tissues; diagnostic performance of the prototype system was tested in 27 tissues procured during breast conservative surgery.

EXPERIMENTAL DESIGN

A custom-built, scanning in situ spectroscopy platform sampled broadband reflectance from a 150-μm-diameter spot over a 1 × 1 cm(2) field using a dark field geometry and telecentric lens; the system was designed to balance sensitivity to cellular morphology and imaging the inherent diversity within tissue subtypes. Nearly 300,000 broadband spectra were parameterized using light scattering models and spatially dependent spectral signatures were interpreted using a cooccurrence matrix representation of image texture.

RESULTS

Local scattering changes distinguished benign from malignant pathologies with 94% accuracy, 93% sensitivity, 95% specificity, and 93% positive and 95% negative predictive values using a threshold-based classifier. Texture and shape features were important to optimally discriminate benign from malignant tissues, including pixel-to-pixel correlation, contrast and homogeneity, and the shape features of fractal dimension and Euler number. Analysis of the region-based diagnostic performance showed that spectroscopic image features from 1 × 1 mm(2) areas were diagnostically discriminant and enabled quantification of within-class tissue heterogeneities.

CONCLUSIONS

Localized scatter-imaging signatures detected by the scanning spectroscopy platform readily distinguished benign from malignant pathologies in surgical tissues and showed new spectral-spatial signatures of clinical breast pathologies.

Portable optical fiber probe-based spectroscopic scanner for rapid cancer diagnosis: a new tool for intraoperative margin assessment

There continues to be a significant clinical need for rapid and reliable intraoperative margin assessment during cancer surgery. Here we describe a portable, quantitative, optical fiber probe-based, spectroscopic tissue scanner designed for intraoperative diagnostic imaging of surgical margins, which we tested in a proof of concept study in human tissue for breast cancer diagnosis. The tissue scanner combines both diffuse reflectance spectroscopy (DRS) and intrinsic fluorescence spectroscopy (IFS), and has hyperspectral imaging capability, acquiring full DRS and IFS spectra for each scanned image pixel. Modeling of the DRS and IFS spectra yields quantitative parameters that reflect the metabolic, biochemical and morphological state of tissue, which are translated into disease diagnosis. The tissue scanner has high spatial resolution (0.25 mm) over a wide field of view (10 cm × 10 cm), and both high spectral resolution (2 nm) and high spectral contrast, readily distinguishing tissues with widely varying optical properties (bone, skeletal muscle, fat and connective tissue). Tissue-simulating phantom experiments confirm that the tissue scanner can quantitatively measure spectral parameters, such as hemoglobin concentration, in a physiologically relevant range with a high degree of accuracy (<5% error). Finally, studies using human breast tissues showed that the tissue scanner can detect small foci of breast cancer in a background of normal breast tissue. This tissue scanner is simpler in design, images a larger field of view at higher resolution and provides a more physically meaningful tissue diagnosis than other spectroscopic imaging systems currently reported in literatures. We believe this spectroscopic tissue scanner can provide real-time, comprehensive diagnostic imaging of surgical margins in excised tissues, overcoming the sampling limitation in current histopathology margin assessment. As such it is a significant step in the development of a platform technology for intraoperative management of cancer, a clinical problem that has been inadequately addressed to date.

Dark-field scanning in situ spectroscopy platform for broadband imaging of resected tissue

A dark-field geometry spectral imaging system is presented to raster scan thick tissue samples in situ in 1.5 cm square sections, recovering full spectra from each 100 μm diameter pixel. This spot size provides adequate resolution for wide field scanning, while also facilitating scatter imaging without requiring sophisticated light-tissue transport modeling. The system is demonstrated showing accurate estimation of localized scatter parameters and the potential to recover absorption-based contrast from broadband reflectance data measured from 480 nm up to 750 nm in tissue phantoms. Results obtained from xenograft pancreas tumors show the ability to quantitatively image changes in localized scatter response in this fast-imaging geometry. The polychromatic raster scan design allows the rapid scanning necessary for use in surgical/clinical applications where timely decisions are required about tissue pathology.

Automated classification of breast pathology using local measures of broadband reflectance

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.

Microscopic imaging and spectroscopy with scattered light

Optical contrast based on elastic scattering interactions between light and matter can be used to probe cellular structure, cellular dynamics, and image tissue architecture. The quantitative nature and high sensitivity of light scattering signals to subtle alterations in tissue morphology, as well as the ability to visualize unstained tissue in vivo, has recently generated significant interest in optical-scatter-based biosensing and imaging. Here we review the fundamental methodologies used to acquire and interpret optical scatter data. We report on recent findings in this field and present current advances in optical scatter techniques and computational methods. Cellular and tissue data enabled by current advances in optical scatter spectroscopy and imaging stand to impact a variety of biomedical applications including clinical tissue diagnosis, in vivo imaging, drug discovery, and basic cell biology.

Optical spectroscopy detects histological hallmarks of pancreatic cancer

An empirical model was developed to interpret differences in the experimentally measured reflectance and fluorescence spectra of freshly excised human pancreatic tissues: normal, adenocarcinoma, and pancreatitis (inflammation). The model provided the first quantitative links between spectroscopic measurements and histological characteristics in the human pancreas. The reflectance model enabled the first (to our knowledge) extraction of wavelength resolved absorption and reduced scattering coefficients for normal and diseased human pancreatic tissues. The fluorescence model employed reflectance information to extract attenuation free "intrinsic" endogenous fluorescence spectra from normal pancreatic tissue, pancreatic adenocarcinoma, and pancreatitis. The method developed is simple, intuitive, and potentially useful for a range of applications in optical tissue diagnostics. This approach is potentially applicable to in vivo studies, because it can account for the absorptive effects of blood in tissues.

Automated identification of tumor microscopic morphology based on macroscopically measured scatter signatures

An automated algorithm and methodology is presented to identify tumor-tissue morphologies based on broadband scatter data measured by raster scan imaging of the samples. A quasi-confocal reflectance imaging system was used to directly measure the tissue scatter reflectance in situ, and the spectrum was used to identify the scattering power, amplitude, and total wavelength-integrated intensity. Pancreatic tumor and normal samples were characterized using the instrument, and subtle changes in the scatter signal were encountered within regions of each sample. Discrimination between normal versus tumor tissue was readily performed using a K-nearest neighbor classifier algorithm. A similar approach worked for regions of tumor morphology when statistical preprocessing of the scattering parameters was included to create additional data features. This type of automated interpretation methodology can provide a tool for guiding surgical resection in areas where microscopy imaging cannot be realized efficiently by the surgeon. In addition, the results indicate important design changes for future systems.