In India oral cancer ranks the top due to the habitual usage of tobacco in its various forms and remains the major burden. Hence priority is given for early diagnosis as it is the better solution for cure or to improve the survival rate. For the past three decades, optical spectroscopic techniques have shown its capacity in the discrimination of normal and malignant samples. Many research works have conventional Raman in the effective detection of cancer using the variations in bond vibrations of the molecules. However in addition polarized Raman provides the orientation and symmetry of biomolecules. If so can polarized Raman be the better choice than the conventional Raman in the detection of cancer? The present study aimed to found the answer for the above query. The conventional and polarized Raman spectra were acquired for the same set of blood plasma samples of normal subjects and oral malignant (OSCC) patients. Thus, obtained Raman spectral data were compared using linear discriminant analysis coupled with artificial neural network (LDA-ANN). The depolarization ratio of biomolecules such as antioxidant, amino acid, protein and nucleic acid bases present in blood plasma was proven to be the best attributes in the categorization of the groups. The polarized Raman results were promising in discriminating oral cancer blood plasma from that of normal blood plasma with improved efficiency. The results will be discussed in detail.
During metabolism the metabolites such as hormones, proteins and enzymes were released in to the blood stream by the cells. These metabolites reflect any change that occurs due to any disturbances in normal metabolic function of the human system. This was well observed with the altered spectral signatures observed with fluorescence spectroscopic technique. Previously many have reported on the significance of native fluorescence spectroscopic method in the diagnosis of cancer. As fluorescence spectroscopy is sensitive and simple, it has complementary techniques such as excitation-emission matrix, synchronous and polarization. The fluorescence polarization measurement provides details about any association or binding reactions and denaturing effects that occurs due to change in the micro environment of cells and tissues. In this study, we have made an attempt in the diagnosis of oral cancer at 405 nm excitation using fluorescence polarization measurement. The fluorescence anisotropic values calculated from polarized fluorescence spectral data of normal and oral cancer subjects yielded a good accuracy when analyzed with linear discriminant analysis based artificial neural network. The results will be discussed in detail.
Oral cancer is the most frequent type of cancer that occurs with 75000 to 80000 new cases reported every year in India. The carcinogens from tobacco and related products are the main cause for the oral cancer. ATR-FTIR method is label free, fast and cost-effective diagnostic method would allow for rapid diagnostic results in earlier stages by the minimal chemical changes occur in the biological metabolites available in the blood plasma. The present study reports the use of ATR-FTIR data with advanced statistical model (LDA-ANN) in the diagnosis of oral cancer from normal with better accuracy. The infrared spectra were acquired on ATR-FTIR Jasco spectrophotometer at 4 cm-1 resolution, 30 scans, in the 1800-900 cm-1 spectral range. Each sample had 5 spectra recorded from each blood plasma sample. The spectral data were routed through the multilayer perception of artificial neural network to evaluate for the statistical efficacy. Among the spectral data it was found that amide II (1486 cm-1) and lipid (1526 cm-1) affords about 90 % in the discrimination between groups using LDA. These preliminary results indicate that ATR-FTIR is useful to differentiate normal subject from oral cancer patients using blood plasma.
Blood plasma possesses the biomolecules released from cells/tissues after metabolism and reflects the pathological conditions of the subjects. The analysis of biofluids for disease diagnosis becomes very attractive in the diagnosis of cancers due to the ease in the collection of samples, easy to transport, multiple sampling for regular screening of the disease and being less invasive to the patients. Hence, the intention of this study was to apply near-infrared (NIR) Raman spectroscopy in the high wavenumber (HW) region (2500−3400 cm−1) for the diagnosis of oral malignancy using blood plasma. From the Raman spectra it is observed that the biomolecules protein and lipid played a major role in the discrimination between groups. The diagnostic algorithms based on principal components analysis coupled with linear discriminant analysis (PCA-LDA) with the leave-one-patient-out cross-validation method on HW Raman spectra yielded a promising results in the identification of oral malignancy. The details of results will be discussed.
Many research works based on fluorescence spectroscopy have proven its potential in the diagnosis of various diseases using the spectral signatures of the native key fluorophores such as tryptophan, tyrosine, collagen, NADH, FAD and porphyrin. These fluorophores distribution, concentration and their conformation may be changed depending upon the pathological and metabolic conditions of cells and tissues. In this study, we have made an attempt to characterize the blood plasma of normal subject and oral cancer patients by native fluorescence spectroscopy at 280 nm excitation. Further, the fluorescence data were analyzed by employing the multivariate statistical method - linear discriminant analyses (LDA) using leaves one out cross validation method. The results illustrate the potential of fluorescence spectroscopy technique in the diagnosis of oral cancer using blood plasma.
Urine is considered diagnostically important for tits native fluorophores and they vary in their distribution, concentration and physiochemical properties, depending upon the metabolic condition of the subject. In this study, we have made an attempt, to characterize the urine of normal subject and diabetic patients under medication by native fluorescence spectroscopy at 280 nm excitation. Further, the fluorescence data were analyzed employing the multivariate statistical method linear discriminant analysis (LDA) using leave one out cross validation method. The results were promising in discriminating diabetic urine from that of normal urine. This study in future may be extended to check the feasibility in ruling out the coexisting disorders such as cancer.
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.