Chronic-rhinosinusitis (CRS) is one of the most common conditions affecting ~14.2% (29.2-million) of US adults leading to estimated 18-22 million-physician office visits. It causes significant physical symptoms, negatively affects the quality-of-life and can substantially impair daily functioning. Various factors including microorganisms, allergies, and other inflammatory triggers play role in CRS. Lack of a universal marker and acknowledged difficulty in establishing the causes for the condition contributes to the poor treatment strategies and outcomes associated with CRS. Utilizing panel of sensitive markers associated with inflammatory responses in the nasal area can provide clinicians valuable information about the disease at the molecular level. The present study aims at identifying spectrochemical markers associated with the onset of CRS using data-driven Raman imaging. By combining high-resolution Raman imaging and machine learning we have developed a novel approach to obtain an integrated insight. Our findings are suggestive of differential changes in the biochemical composition of nasal tissues with CRS onset. A regression-based framework has been developed to link the inflammation score with spectral features. Support vector machine has been employed to explore the feasibility of classification. Successful recognition of these markers in nasal tissues will be helpful not only in designing automated diagnosis platforms but can also be used for identifying novel treatment strategies. Findings of this study will also serve as the foundation of our future research work on evaluating the applicability of nasal lavage for a minimally invasive method for objective CRS diagnosis.
Otitis media is one of the most common reasons for pediatrician visits, antibiotic prescription, and surgery in the pediatric population. Visible light pneumatic-otoscopy is considered the best currently available diagnostic tool for otitis media. However, it has various limitations e.g. the disposable speculum cannot create an adequate seal against the external auditory canal to obtain tympanic membrane movement. Also, lack of training for effective pneumatic-otoscopy for most clinicians is another factor. To overcome these limitations, we have recently developed an otoscope sensitive to shortwave infrared (SWIR) wavelengths of light. A SWIR otoscope could help identify middle-ear-effusions based on the strong light absorption by ear fluid. Due to a longer wavelength, light can penetrate deeper through tissue, enabling a better view behind the tympanic membrane. Here we present our preliminary findings on the feasibility of using video rate SWIR imaging in a pediatric population. A total of 74 ear video recordings were obtained in the study from 20 patients. There was an improvement in the ability to see through the tympanic membrane using the SWIR otoscope. Three patients with middle ear effusion, confirmed by pneumatic otoscopy, were all identified using both visible and SWIR otoscopy. The average contrast for visible otoscopy in the presence of middle ear effusion was 0.097 and for SWIR was 0.29. In tympanic membranes with myringosclerosis, neither technique was able to see through affected areas. However, the SWIR otoscope was able to see through dried blood, dried secretions and thin dry areas of cerumen overlying the tympanic membrane.
Changes in the cellular homeostasis in response to a stimuli, disease or therapeutic intervention are multifaceted in nature, and cannot be grasped by routinely employed targeted imaging that focuses on a small set of suspected molecules or genes. Novel approaches relying on global analysis of cellular features, from morphology to the composite biomolecular status (notably chemical composition and molecular conformation), is a pre-requisite for accurate monitoring of cellular processes. In the present study label-free profiling of normal skin fibroblasts (Hs895.Sk) exposed to sub-lethal doses of ultra-violet radiation has been performed using quantitative phase imaging and Raman spectroscopy. Spectral differences in the Raman fingerprint region indicates differences in the protein and nucleic acid composition. These differences were successfully utilized to develop an automated classification model based on principal component analysis. Distinct changes in the cellular morphology were observed and validated through quantitative phase imaging. Significant dose dependent differences in different biophysical parameters such as dry mass and matter density were observed. Combination of these two techniques, one suited for detection of subtle morphological/biophysical alterations while the other appropriate for capturing molecular perturbations, could pave the way to address issues of label-free monitoring of cellular responses in response to an external stimulus. These findings can provide an accurate understanding of different markers associated with radiation damage and would assist in providing a quantitative tool to our future studies on designing alternate diagnostic tools.
We combined Raman micro-spectroscopy and machine learning techniques to develop
a classification model based on a well-established non-alcoholic steatohepatitis (NASH) mouse
model, using spectrum pre-processing, biochemical component analysis (BCA) and logistic
regression.
Oral cancer is one of the most common cancers worldwide. One-fifth of the world’s oral cancer subjects are from India and other South Asian countries. The present Raman mapping study was carried out to understand biochemical variations in normal and malignant oral buccal mucosa. Data were acquired using WITec alpha 300R instrument from 10 normal and 10 tumors unstained tissue sections. Raman maps of normal sections could resolve the layers of epithelium, i.e. basal, intermediate, and superficial. Inflammatory, tumor, and stromal regions are distinctly depicted on Raman maps of tumor sections. Mean and difference spectra of basal and inflammatory cells suggest abundance of DNA and carotenoids features. Strong cytochrome bands are observed in intermediate layers of normal and stromal regions of tumor. Epithelium and stromal regions of normal cells are classified by principal component analysis. Classification among cellular components of normal and tumor sections is also observed. Thus, the findings of the study further support the applicability of Raman mapping for providing molecular level insights in normal and malignant conditions.
Introduction
Optical spectroscopic methods are being explored as novel tools for early and non-invasive cancer diagnosis. Both ex
vivo and in vivo Raman spectroscopic studies carried out in oral cancer over the past decade have demonstrated that
spectra of normal tissues are rich in lipids while tumor spectra show predominance of proteins. An accurate
understanding of spectral features with respect to the biochemical composition is a pre-requisite before transferring these
technologies for routine clinical usage. Therefore, in the present study, we have carried out Raman and biochemical
studies on same tissues to correlate spectral markers and biochemical composition of normal and tumor oral tissues.
Materials and Methods
Spectra of 20 pairs of normal and tumor oral tissues were acquired using fiber-optic probe coupled HE-785 Raman
spectrometer. Intensity associated with lipid (1440 cm-1) and protein (1450 and 1660 cm-1) bands were computed using
curve-deconvolution method. Same tissues were then subjected to biochemical estimations of major biomolecules i.e.,
protein, lipid and phospholipids.
Results and Discussion
The intensity of the lipid band was found to be higher in normal tissues with respect to tumors, and the protein band was
higher in tumors compared to normal tissues. Biochemical estimation yielded similar results i.e. high protein to lipid or
phospholipid ratio in tumors with-respect to normal tissues. These differences were found to be statistically significant.
Conclusion
Findings of curve-deconvolution and biochemical estimation correlate very well and corroborate the spectral profile
noted in earlier studies.
KEYWORDS: Raman spectroscopy, Tumors, Control systems, Cancer, In vivo imaging, Spectroscopy, Principal component analysis, Tissues, Data modeling, Biopsy
Cancers of oral cavities are one of the most common malignancies in India and other south-Asian countries. Tobacco habits are the main etiological factors for oral cancer. Identification of premalignant lesions is required for improving survival rates related to oral cancer. Optical spectroscopy methods are projected as alternative/adjunct for cancer diagnosis. Earlier studies have demonstrated the feasibility of classifying normal, premalignant, and malignant oral ex-vivo tissues. We intend to evaluate potentials of Raman spectroscopy in detecting premalignant conditions. Spectra were recorded from premalignant patches, contralateral normal (opposite to tumor site), and cancerous sites of subjects with oral cancers and also from age-matched healthy subjects with and without tobacco habits. A total of 861 spectra from 104 subjects were recorded using a fiber-optic probe-coupled HE-785 Raman spectrometer. Spectral differences in the 1200- to 1800-cm −1 region were subjected to unsupervised principal component analysis and supervised linear discriminant analysis followed by validation with leave-one-out and an independent test data set. Results suggest that premalignant conditions can be objectively discriminated with both normal and cancerous sites as well as from healthy controls with and without tobacco habits. Findings of the study further support efficacy of Raman spectroscopic approaches in oral-cancer applications.
Keratins are one of most widely used markers for oral cancers. Keratin 8 and 18 are expressed in simple epithelia and
perform both mechanical and regulatory functions. Their expression are not seen in normal oral tissues but are often
expressed in oral squamous cell carcinoma. Aberrant expression of keratins 8 and 18 is most common change in human
oral cancer. Optical-spectroscopic methods are sensitive to biochemical changes and being projected as novel diagnostic
tools for cancer diagnosis. Aim of this study was to evaluate potentials of Raman spectroscopy in detecting minor
changes associated with differential level of keratin expression in tongue-cancer-derived AW13516 cells. Knockdown
clones for K8 were generated and synchronized by growing under serum-free conditions. Cell pellets of three
independent experiments in duplicate were used for recording Raman spectra with fiberoptic-probe coupled HE-785
Raman-instrument. A total of 123 and 96 spectra from knockdown clones and vector controls respectively in 1200-1800
cm-1 region were successfully utilized for classification using LDA. Two separate clusters with classification-efficiency
of ~95% were obtained. Leave-one-out cross-validation yielded ~63% efficiency. Findings of the study demonstrate the
potentials of Raman spectroscopy in detecting even subtle changes such as variations in keratin expression levels. Future
studies towards identifying Raman signals from keratin in oral cells can help in precise cancer diagnosis.
Oral squamous cell carcinoma is sixth among the major malignancies worldwide. Tobacco habits are known as major
causative factor in tumor carcinogenesis in oral cancer. Optical spectroscopy methods, including Raman, are being
actively pursued as alternative/adjunct for cancer diagnosis. Earlier studies have demonstrated the feasibility of
classifying normal, premalignant and malignant oral ex-vivo tissues. In the present study we have recorded in vivo
spectra from contralateral normal and diseased sites of 50 subjects with pathologically confirmed lesions of buccal
mucosa using fiber-optic-probe-coupled HE-785 Raman spectrometer. Spectra were recorded on similar points as per
teeth positions with an average acquisition time of 8 seconds. A total of 215 and 225 spectra from normal and tumor
sites, respectively, were recorded. Finger print region (1200-1800 cm-1) was utilized for classification using LDA.
Standard-model was developed using 125 normal and 139 tumor spectra from 27 subjects. Two separate clusters with an
efficiency of ~95% were obtained. Cross-validation with leave-one-out yielded ~90% efficiency. Remaining 90 normal
and 86 tumor spectra were used as test data and predication efficiency of model was evaluated. Findings of the study
indicate that Raman spectroscopic methods in combination with appropriate multivariate tool can be used for objective,
noninvasive and rapid diagnosis.
Oral squamous cell carcinoma is one of among the top 10 malignancies. Optical spectroscopy, including Raman, is being actively pursued as alternative/adjunct for cancer diagnosis. Earlier studies have demonstrated the feasibility of classifying normal, premalignant, and malignant oral ex vivo tissues. Spectral features showed predominance of lipids and proteins in normal and cancer conditions, respectively, which were attributed to membrane lipids and surface proteins. In view of recent developments in deep tissue Raman spectroscopy, we have recorded Raman spectra from superior and inferior surfaces of 10 normal oral tissues on intact, as well as incised, biopsies after separation of epithelium from connective tissue. Spectral variations and similarities among different groups were explored by unsupervised (principal component analysis) and supervised (linear discriminant analysis, factorial discriminant analysis) methodologies. Clusters of spectra from superior and inferior surfaces of intact tissues show a high overlap; whereas spectra from separated epithelium and connective tissue sections yielded clear clusters, though they also overlap on clusters of intact tissues. Spectra of all four groups of normal tissues gave exclusive clusters when tested against malignant spectra. Thus, this study demonstrates that spectra recorded from the superior surface of an intact tissue may have contributions from deeper layers but has no bearing from the classification of a malignant tissues point of view.
Cancer is now recognized as one of the major causes of morbidity and mortality. Histopathological diagnosis, the gold standard, is shown to be subjective, time consuming, prone to interobserver disagreement, and often fails to predict prognosis. Optical spectroscopic methods are being contemplated as adjuncts or alternatives to conventional cancer diagnostics. The most important aspect of these approaches is their objectivity, and multivariate statistical tools play a major role in realizing it. However, rigorous evaluation of the robustness of spectral models is a prerequisite. The utility of Raman spectroscopy in the diagnosis of cancers has been well established. Until now, the specificity and applicability of spectral models have been evaluated for specific cancer types. In this study, we have evaluated the utility of spectroscopic models representing normal and malignant tissues of the breast, cervix, colon, larynx, and oral cavity in a broader perspective, using different multivariate tests. The limit test, which was used in our earlier study, gave high sensitivity but suffered from poor specificity. The performance of other methods such as factorial discriminant analysis and partial least square discriminant analysis are at par with more complex nonlinear methods such as decision trees, but they provide very little information about the classification model. This comparative study thus demonstrates not just the efficacy of Raman spectroscopic models but also the applicability and limitations of different multivariate tools for discrimination under complex conditions such as the multicancer scenario.
Optical spectroscopic methods are being contemplated as adjunct/ alternative to existing 'Gold standard' of cancer
diagnosis, histopathological examination. Several groups are actively pursuing diagnostic applications of Ramanspectroscopy
in cancers. We have developed Raman spectroscopic models for diagnosis of breast, oral, stomach, colon
and larynx cancers. So far, specificity and applicability of spectral- models has been limited to particular tissue origin. In
this study we have evaluated explicitly of spectroscopic-models by analyzing spectra from already developed spectralmodels
representing normal and malignant tissues of breast (46), cervix (52), colon (25), larynx (53), and oral (47).
Spectral data was analyzed by Principal Component Analysis (PCA) using scores of factor, Mahalanobis distance and
Spectral residuals as discriminating parameters. Multiparametric limit test approach was also explored. The preliminary
unsupervised PCA of pooled data indicates that normal tissue types were always exclusive from their malignant
counterparts. But when we consider tissue of different origin, large overlap among clusters was found. Supervised
analysis by Mahalanobis distance and spectral residuals gave similar results. The 'limit test' approach where
classification is based on match / mis-match of the given spectrum against all the available spectra has revealed that
spectral models are very exclusive and specific. For example breast normal spectral model show matches only with
breast normal spectra and mismatch to rest of the spectra. Same pattern was seen for most of spectral models. Therefore,
results of the study indicate the exclusiveness and efficacy of Raman spectroscopic-models. Prospectively, these findings
might open new application of Raman spectroscopic models in identifying a tumor as primary or metastatic.
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.