Search

The pituitary gland is a small but vital organ in the human body. It is located at the base of the brain and is often described as the master gland due to its multiple functions. The pituitary gland secretes and stores hormones, such as the thyroid-stimulating hormone (TSH), adrenocorticotropic hormone (ACTH), growth hormone (hGH), prolactin, gonadotropins, and luteinizing hormones, as well as the antidiuretic hormone (ADH). A proper diagnosis of pituitary disorders is of utmost importance as this organ participates in regulating a variety of body functions. Typical histopathological analysis provides much valuable information, but it gives no insight into the biochemical background of the changes that occur within the gland. One approach that could be used to evaluate the biochemistry of tissue sections obtained from pituitary disorders is Fourier Transform Infra-Red (FTIR) spectromicroscopy. In order to collect diagnostically valuable information large areas of tissue must be investigated. This work focuses on obtaining a unique and representative FTIR spectrum characteristic of one type of cell architecture within a sample. The idea presented is based on using hierarchical cluster analysis (HCA) for data evaluation to search for uniform patterns within samples from the perspective of FTIR spectra. The results obtained demonstrate that FTIR spectromicroscopy, combined with proper statistical evaluation, can be treated as a complementary method for histopathological analysis and ipso facto can increase the sensitivity and specificity for detecting various disorders not only for the pituitary gland, but also for other human tissues. [ABSTRACT FROM AUTHOR]

Pre-processing of Fourier transform infrared (FTIR) spectra is typically the first and crucial step in data analysis. Very often hyperspectral datasets include the regions characterized by the spectra of very low intensity, for example two-dimensional (2D) maps where the areas with only support materials (like mylar foil) are present. In that case segmentation of the complete dataset is required before subsequent evaluation. The method proposed in this contribution is based on a multivariate approach (hierarchical cluster analysis), and shows its superiority when compared to the standard method of cutting-off by using only the mean spectral intensity. Both techniques were implemented and their performance was tested in the R statistical environment - open-source platform - that is a favourable solution if the repeatability and transparency are the key aspects.

Synchrotron radiation-based Fourier transform infra-red (SR-FTIR) micro-imaging has been developed as a rapid, direct and non-destructive technique. This method, taking advantage of the high brightness and small effective source size of synchrotron light, is capable of exploring the molecular chemistry within the microstructures of microscopic particles without their destruction at high spatial resolutions. This is in contrast to traditional "wet" chemical methods, which, during processing for analysis, often caused destruction of the original samples. In the present study, we demonstrate the potential of SR-FTIR micro-imaging as an effective way to accurately identify microscopic particles deposited within latent fingerprints. These particles are present from residual amounts of materials left on a person's fingers after handling such materials. Fingerprints contaminated with various types of powders, creams, medications and high explosive materials (3-nitrooxy-2,2-bis(nitrooxymethyl)propyl nitrate (PETN), 1,3,5-trinitro-1,3,5-triazinane (RDX), 2-methyl-1,3,5-trinitrobenzene (TNT)) deposited on various - daily used - substrates have been analysed herein without any further sample preparation. A non-destructive method for the transfer of contaminated fingerprints from hard-to-reach areas of the substrates to the place of analysis is also presented. This method could have a significant impact on forensic science and could dramatically enhance the amount of information that can be obtained from the study of fingerprints.