Multifractal Alterations in Oral Sub-Epithelial Connective Tissue During Progression of Pre-Cancer and Cancer

Bright-field microscopy (BFM) encrypts the optical transillumination profile of the transmitted light attenuated by the complex micro-structural tissue convolutions, manifested by the dense and compact regions of the specimen under examination. The connotations of idiosyncratic tissue interaction dynamics with the onset of pre-cancerous activity are encoded in the BFM acquired oral mucosa histopathological images (OMHI). In the present study, our analysis is focused on the sub-epithelium region of the oral mucosa, which has high clinical significance but sparsely explored in the literature from the textural domain. Histopathology being the gold-standard technique till date, we have used the light microscopic histopathology images for tissue characterization. The tissue-index transmission patches (TITP) from the sub-epithelium region are cropped under the guidance of oral onco-pathologists. After that, the TITPs are characterized for its multi-scale spatial-deformation dynamics, while keeping the intrinsic anisotropic geometry, and local contour connectivity within tolerable limits. With recent studies exhibiting multifractal's potency in diverse biological system analysis, here, we exploit the 2D multifractal detrended fluctuation analysis (2D-MFDFA) on TITPs for exploring a discriminative set of multifractal signatures for healthy, oral potentially malignant disorders and oral cancer tissue sample. The predictive model's competency is validated on an experimentally collected corpus of TITP samples and substantiated via confirmatory data statistics and analysis, showing its inter-class segregation efficacy. Moreover, the 2D-MFDFA analysis evinces the complex multifractal patterns in TITPs, which is due to the presence of composite long-range correlations in the oral mucosa tissue fabric.

Dimension reduction technique using a multilayered descriptor for high-precision classification of ovarian cancer tissue using optical coherence tomography: a feasibility study

Optical coherence tomography (OCT) yields microscopic volumetric images representing tissue structures based on the contrast provided by elastic light scattering. Multipatient studies using OCT for detection of tissue abnormalities can lead to large datasets making quantitative and unbiased assessment of classification algorithms performance difficult without the availability of automated analytical schemes. We present a mathematical descriptor reducing the dimensionality of a classifier's input data, while preserving essential volumetric features from reconstructed three-dimensional optical volumes. This descriptor is used as the input of classification algorithms allowing a detailed exploration of the features space leading to optimal and reliable classification models based on support vector machine techniques. Using imaging dataset of paraffin-embedded tissue samples from 38 ovarian cancer patients, we report accuracies for cancer detection [Formula: see text] for binary classification between healthy fallopian tube and ovarian samples containing cancer cells. Furthermore, multiples classes of statistical models are presented demonstrating [Formula: see text] accuracy for the detection of high-grade serous, endometroid, and clear cells cancers. The classification approach reduces the computational complexity and needed resources to achieve highly accurate classification, making it possible to contemplate other applications, including intraoperative surgical guidance, as well as other depth sectioning techniques for fresh tissue imaging.

Fourier transform infra-red spectroscopic signatures for lung cells' epithelial mesenchymal transition: A preliminary report

Infra red (IR) spectral characterization can provide label-free cellular metabolic signatures of normal and diseased circumstances in a rapid and non-invasive manner. Present study endeavoured to enlist Fourier transform infra red (FTIR) spectroscopic signatures for lung normal and cancer cells during chemically induced epithelial mesenchymal transition (EMT) for which global metabolic dimension is not well reported yet. Occurrence of EMT was validated with morphological and immunocytochemical confirmation. Pre-processed spectral data was analyzed using ANOVA and principal component analysis-linear discriminant analysis (PCA-LDA). Significant differences observed in peak area corresponding to biochemical fingerprint (900-1800cm^-1) and high wave-number (2800-3800cm^-1) regions contributed to adequate PCA-LDA segregation of cells undergoing EMT. The findings were validated by re-analysis of data using another in-house built binary classifier namely vector valued regularized kernel approximation (VVRKFA), in order to understand EMT progression. To improve the classification accuracy, forward feature selection (FFS) tool was employed in extracting potent spectral signatures by eliminating undesirable noise. Gradual increase in classification accuracy with EMT progression of both cell types indicated prominence of the biochemical alterations. Rapid changes in cellular metabolome noted in cancer cells within first 24h of EMT induction along with higher classification accuracy for cancer cell groups in comparison to normal cells might be attributed to inherent differences between them. Spectral features were suggestive of EMT triggered changes in nucleic acid, protein, lipid and bound water contents which can emerge as the useful markers to capture EMT related cellular characteristics.