Chemometrics of biological tissues using infrared spectroscopy

Abstract

Vibrational spectroscopy techniques have been considered for the diagnosis of diseases. In particular, Fourier transforms infrared spectroscopy (FTIR) is an optical method widely adopted in biomedical research. This is because FTIR images allow the collection of various molecular information enabling a correlation between spectral characteristics and structural changes. Incorporating qualitative and quantitative aspects is essential to implementing a chemometric procedure that transcribes spectral variance into meaningful information. In the case of burns, the degree of tissue damage will strongly depend on several factors, such as severity, depth, and size, which influence the healing process. Laser scar treatment has been used for decades for post-healing remodeling. However, conventional laser techniques may not wholly normalize mature scar tissue, and a preventive approach to laser exposure during wound healing has emerged. Early intervention has already been investigated in several clinical studies, but there is no consensus. Therefore, the objectives of this study were; to investigate FTIR in biological tissue analysis to provide better chemometric strategies; to study the feasibility of FTIR to establish a profile of the skin and bone tissue, and to evaluate the effect of laser on wound healing and the development of a new chemometric methodology for automatic digital dewaxing. To this end, thermal injuries were performed on the back of Wistar rats by applying water vapor at 90ºC for 12s. Subsequently, on day three post-burn, laser irradiation was performed. In the laser procedure, 3 protocols were adopted: 1) Er:glass (SellasEvo®) with 3025FPA, 100mJ, 3s; 2) LLT with red laser with = 660nm, power 40mW; 3) Combination of the two lasers with the previous parameters. Then, a histological and chemometric analysis was performed. For the chemometric evaluation of the data, an analysis protocol was developed based on a pre-processing using OCTAVVS followed by implementing a baseline, threshold, and application of PCA. From that point, the images are classified, they can be normalized and go through the Fuzzy C-means (FCM) technique to determine the spectral differences. All-laser groups demonstrate a shrinkage of the wound region in the histopathological findings. Er:Glass, showed a smaller retraction area and less epithelium recovery. The PBM group has a dense crust. Both treatments demonstrate intense shrinkage and more consistent recovery of the epidermis. In the FCM, three trends occur 1) in the range of 900-950 cm-1 it has a significantly reduced amplitude in the control group; 2) in the range of 1200-1250 cm-1 the control group has a similar amplitude demonstrating bands of sensitive markers; and 3) the amide regions provide various spectral markers. The control group, as expected, remains with minimal changes in this FTIR region. On the other hand, in the PBM group, there is a shift from 1660 to 1679 cm-1, which may correspond to changes in the molecular structure of a protein in the context of type I collagen. FTIR is a valuable alternative for tissue research on a biochemical level through fingerprints that can be translated into morphological features.