Fourier transform infrared (FTIR) spectral mapping of the cervical transformation zone, and dysplastic squamous epithelium

Abstract

Objective-This paper is aimed at establishing infrared spectral patterns for the different tissue types found in, and for different stages of disease of squamous cervical epithelium. Methods for the unsupervised distinction of these tissue types are discussed.Methods-Fourier transform infrared (FTIR) maps of the squamous and glandular cervical epithelium, and of the cervical transformation zone, were obtained and analyzed by multivariate unsupervised hierarchical cluster methods. The resulting clusters are correlated to the corresponding stained histopathological features in the tissue sections.Results-Multivariate statistical analysis of FTIR spectra collected for tissue sections permit an unsupervised method of distinguishing tissue types, and of differentiating between normal and diseased tissue. By analyzing different spectral windows and comparing the results with histology, we found the amide I and II region (1740-1470 cm −1 ) to be very important in correlating anatomical and histopathological features in tissue to spectral clusters. Since an unsupervised, rather than a diagnostic, algorithm was used in these efforts, no statistical analysis of false-positive/false-negative results is reported at this time.Conclusions-The combination of FTIR micro-spectroscopy and multivariate spectral processing provides important insights into the fundamental spectral signatures of individual cells and consequently shows potential as a diagnostic tool for cervical cancer.

Molecular pathology via IR and Raman spectral imaging

“…Multivariate methods, in contrast, utilize the entire spectral vectors to create images from the hyperspectral datasets. One commonly used multivariate method is unsupervised hierarchical cluster analysis (HCA) [21,56] which calculates the similarity of all spectra in a dataset, and assigns color codes to spectral groups, or clusters, based on their similarity. Figure 5 shows such an HCA-based spectral image and, for comparison, an H & E-stained image of the same tissue section (see Section 3.5).…”

Infrared spectral histopathology (SHP): a novel diagnostic tool for the accurate classification of lung cancer

“…HCA is a well-known method to extract patterns in data sets; 45 in this particular application, HCA is used to detect spectral similarities. 46 To this end, the similarity between all pairs of spectra in a 1 mm  1 mm section of each tissue spot was computed by a metric known as Euclidean distance, 45 which results in a similarity (correlation) coefficient that ranges from 1.0 for perfectly identical spectra to 0.0 for completely dissimilar spectra. This is a computationally highly intensive step, because a 1 mm  1 mm square region of the tissue containing 25 600 individual spectra requires the computation of 25 600 2 /2 or about 300 million similarity coefficients (for each data set).…”

Automation of an algorithm based on fuzzy clustering for analyzing tumoral heterogeneity in human skin carcinoma tissue sections

“…Thus, FT-IR microimaging has demonstrated potential to provide clinically relevant diagnostic information in oncology. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] The biochemical changes related to carcinogenesis between cancerous and surrounding tissue areas are subtle. As a consequence, IR hyperspectral images need to be processed by powerful digital signal processing and pattern recognition methods in order to highlight these changes.…”