We have taken a three-pronged approach to improving the current standard of melanoma detection:
(a) we are developing a new hyperspectral imaging-based medical device aimed at noninvasively
detecting melanoma (b) we used a commercially available hand-held microscope with
polarization control as a dermoscope, to begin establishing an inexpensive, portable imaging capability
that could help assess the risk of a particular lesion (pigmented nevus) harboring melanoma
(c) we created an updated ABCD algorithm and user interface software that more accurately
generates a single risk number (Total Dermoscopy Score), for allowing a trained clinician
to better assess the need for seeing the patient whose internet-uploaded nevus images they are
evaluating. The hyperspectral instrument (a) is discussed elsewhere, and we focus here on (b)
and (c), in the hope of increasing melanoma awareness and early detection.
Spectral imaging involves capturing images at multiple wavelengths resulting in a data cube (x, y, λ) that allows materials
to be identified by its spectral signature. While hyperspectral imagers can provide high spectral resolution, they also
have major drawbacks such as cost, size, and the copious amounts of data in the image cube. Typically, the complete
hyperspectral data cube provides little additional information compared to only 3-8 discrete (multiwavelength) imaging
bands. We present two new approaches and related technologies where we are able to acquire spectral imaging data
stacks quickly and cost-effectively. Our two spectral imaging systems represent different approaches integrated with
standard CCD and CMOS imagers: sequential rotating filter wheels (RFWs) and lithographically patterned dichroic filter
arrays (DFAs). The RFW approach offers the ability for rapid configuration of a spectral system, and a whole new level
of self-contained image acquisition, processing and on-board display. The DFA approach offers the potential for ultra
compact imagers with acquisition of images of multiple wavelengths simultaneously, while still allowing for processing
and display steps to be built into the camera. Both approaches lend themselves production of multi-wavelength/spectral
imaging systems with differing features and advantages.
DNA methylation plays a key role in cellular differentiation. Aberrant global methylation patterns are associated with
several cancer types, as a result of changes in long-term activation status of up to 50% of genes, including oncogenes
and tumor-suppressor genes, which are regulated by methylation and demethylation of promoter region CpG
dinucleotides (CpG islands). Furthermore, DNA methylation also occurs in nonisland CpG sites (> 95% of the genome),
present once per 80 dinucleotides on average. Nuclear DNA methylation increases during the course of cellular
differentiation while cancer cells usually show a net loss in methylation. Given the large dynamic range in DNA
methylation load, the methylation pattern of a cell can provide a valuable distinction as to its status during differentiation
versus the disease state. By applying immunofluorescence, confocal microscopy and 3D image analysis we assessed the
potential of differential nuclear distribution of methylated DNA to be utilized as a biomarker to characterize cells during
development and when diseased. There are two major fields that may immediately benefit from this development: (1) the
search for factors that contribute to pluripotency and cell fate in human embryonic stem cell expansion and
differentiation, and (2) the characterization of tumor cells with regard to their heterogeneity in molecular composition
and behavior. We performed topological analysis of the distribution of methylated CpG-sites (MeC) versus
heterochromatin. This innovative approach revealed significant differences in colocalization patterns of MeC and
heterochromatin-derived signals between undifferentiated and differentiated human embryonic stem cells, as well as
untreated AtT20 mouse pituitary tumor cells compared to a subpopulation of these cells treated with 5-azacytidine for 48
hours.
We have designed and built an imaging elastic scattering spectroscopy endoscopic instrument for the purpose of
detecting cancer in vivo. As part of our testing and validation of the system, known targets representing potential disease
states of interest were constructed using polystyrene beads of known average diameter and TiO2 crystals embedded in a
two-layer agarose gel. Final construction geometry was verified using a dissection microscope. The phantoms were then
imaged using the endoscopic probe at a known incident angle, and the results compared to model predictions. The
mathematical model that was used combines classic ray-tracing optics with Mie scattering to predict the images that
would be observed by the probe at a given physical distance from a Mie-regime scattering media. This model was used
generate the expected observed response for a broad range of parameter values, and these results were then used as a
library to fit the observed data from the phantoms. Compared against the theoretical library, the best matching signal
correlated well with known phantom material dimensions. These results lead us to believe that imaging elastic scattering
can be useful in detection/diagnosis, but further refinement of the device will be necessary to detect the weak signals in a
real clinical setting.
We present an optical system for small animal imaging that can combine various in vivo imaging modalities, including
fluorescence (intensity and lifetime), spectral, and trans-illumination imaging. This system consists of light-tight box
with ultrafast pulsed or cw laser light excitation, motorized translational and rotational stages, a telecentric lens for
detection, and a cooled CCD camera that can be coupled to an ultrafast time-gated intensifier. All components are
modular, making possible laser excitation at various wavelengths and pulse lengths, and signal detection in a variety of
ways (multimode). Results of drug nanoconjugate carrier delivery studies in mice are presented. Conventional and
spectrally-resolved fluorescence images reveal details of in vivo drug nanoconjugate carrier accumulation within the
tumor region and several organs in real time. By multi-spectral image analysis of ex vivo specimens from the same mice,
we were able to evaluate the extent and topology of drug nanoconjugate carrier distribution into specific organs and the
tumor itself.
Multi-spectral imaging provides digital images of a scene or object at a large, usually sequential number of wavelengths,
generating precise optical spectra at every pixel. We use the term "spectral signature" for a quantitative plot of optical
property variations as a function of wavelengths. We present here intelligent spectral signature bio-imaging methods we
developed, including automatic signature selection based on machine learning algorithms and database search-based
automatic color allocations, and selected visualization schemes matching these approaches. Using this intelligent spectral
signature bio-imaging method, we could discriminate normal and aganglionic colon tissue of the Hirschsprung's disease
mouse model with over 95% sensitivity and specificity in various similarity measure methods and various anatomic
organs such as parathyroid gland, thyroid gland and pre-tracheal fat in dissected neck of the rat in vivo.
The recent emergence of bright, inexpensive colored LEDs offers several advantages over traditional light sources,
including reduced size and increased portability, low power consumption and heat production, increased durability and
longer life, and high temporal resolution. We assembled a modular array of different Phillips LUMILED LUXEON
LEDs, white and seven colors with peak wavelengths between 450 and 640 nm and bandwidths of 20-30 nm. LED
illumination was fiber-optically coupled to the transmitted light path of an inverted microscope, and digital images of
sectioned human tissue stained with absorbing dyes were acquired using combinations of the white and color LEDs. The
LED array was also coupled to an endoscope and used to image human and mouse tissue in situ. Image contrast was
assessed (1) qualitatively by looking down the microscope and by viewing the digital images, and (2) quantitatively by
using entropy analysis in the real and frequency domains to assess the dependence of contrast enhancement on spatial
frequency. Contrast in image features of a given color range was enhanced by LEDs conjugate to that color, whereas
LED colors spanning a wider range enhanced contrast in the entire image, with white LEDs often maximizing contrast of
tissue. This analysis demonstrates the utility of LED illumination in modulating contrast in light microscopy and
endoscopy, which may facilitate spectral segmentation and classification of image features.
We describe a simple fluorescence microscope based on wide-field two-photon excitation. While still taking advantage
of some inherent properties of non-linear (two-photon) microscopy, such as increased penetration depth through tissue
and reduced phototoxicity, this approach provides video frame rate imaging, can be easily coupled to fluorescence
spectral and lifetime detection modules, and makes efficient use of the high average power currently available from
ultrashort pulsed lasers. For a standard histopathology specimen, we were able to identify different structures based on
spectral and fluorescence lifetime detection and analysis. We examined the use of 200fs and 2ps pulses from Spectra
Physics MaiTai and Tsunami lasers, respectively, with average power ranging from 50mW to 500mW.
KEYWORDS: Tissues, Breast cancer, Imaging spectroscopy, Image segmentation, Breast, Cancer, Tumors, Image classification, In vivo imaging, Library classification systems
Spectral imaging has recently been introduced in the biomedical field as a noninvasive, quantitative means of studying biological tissues. Many of its potential applications have been demonstrated (in vitro and, to a lesser degree, in vivo) with the use of stains or dyes. Successful translation to the clinical environment has been largely lagging, due to safety considerations and regulatory limitations preventing use of contrast agents in humans. We report experiments showing the feasibility of high-resolution spectral imaging of breast cancer without the use of contrast agents, thus completing the continuum of translational research, to in vivo imaging that will be directly applicable in the clinical environment. Our initial work focused on image acquisition using Fourier transform microinterferometry and subsequent segmentation of both stained and unstained breast cancer slides-derived image sets. We then applied our techniques to imaging fresh unstained ex vivo specimens of rat breast cancer and sentinel lymph nodes. We also investigated multiple methods of classification to optimize our image analyses, and preliminary results for the best algorithm tested yielded an overall sensitivity of 96%, and a specificity of 92% for cancer detection. Using spectral imaging and classification techniques, we were able to demonstrate that reliable detection of breast cancer in fixed and fresh unstained specimens of breast tissue is possible.
We developed a new endoscope that allows for non-contact, rapid (sub-second) acquisition of polarized spectral images of tissue invivo. The intent was to enable exploration of a variety of optical contrast mechanisms (such as light absorption, reflectance, scattering, and fluorescence) in a search for new methods of early cancer detection in a clinical setting. Our first new implementation for cancer detection is based on a body of spectroscopic work that employs elastic scattering (Mie) theory to estimate the size of bulk scatterers in a given medium - in our case, the epithelial tissue of lungs. This paper describes the novel design of the Hyperspectral Imaging Endoscope, and our initial experiences with employing it for the early detection of dysplasia and cancer in lung epithelia.
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.