Early detection and diagnosis of oral squamous cell carcinoma (OSCC) are essential for improving patient outcomes. This study presents the development of a deep neural network model trained to classify OSCC using hyperspectral FTIR spectroscopy imaging. The network's performance was evaluated using accuracy, precision, recall, and F1-score metrics. The model demonstrated high effectiveness in classifying unfolded OSCC hyperspectral images, showing potential as a valuable tool for supporting diagnostic processes in clinical settings.

This in vitro study aimed to evaluate the preventive action of topical fluoride application combined with laser irradiation under different pH conditions using Fourier Transform Infrared Spectroscopy (FTIR) and Scanning Electron Microscopy (SEM). A total of 180 samples of human dental enamel were prepared and divided into groups: Negative Control, Fluoride (FFA 12.300 µF⁻/g), Laser (Nd:YAG 84.9 J/cm²), and Laser + Fluoride (Nd:YAG 84.9 J/cm² + FFA 12.300 µF⁻/g). The pH cycling was performed at three different pH conditions: pH 5 (below the critical pH for hydroxyapatite), pH 4.5 (below the critical pH in the presence of fluorapatite), and pH 4 (investigating acid resistance of hydroxyapatite and fluorapatite forms with laser irradiation). In the FTIR analysis, the Laser + Fluoride group demonstrated statistically significant differences compared to the Negative Control group and Fluoride group at pH 4.5 and pH 4 when evaluating the phosphate bands. Similar results were observed in the SEM analysis, where the Laser + Fluoride group exhibited lower demineralization compared to the other treatments at pH 4.5 and pH 4. In conclusion, the Laser + Fluoride group demonstrated a significant reduction in demineralization even at pH levels below the critical threshold for fluorapatite, highlighting its superior acid resistance compared to fluoride alone.

This study explores the application of Fourier-transform infrared (FTIR) spectroscopy imaging to analyze changes in the cardiac atria of mice treated with interleukin-1 beta (IL-1beta), a cytokine known to influence cardiac remodeling. By employing k-means clustering on the spectral data, we identified distinct clusters that differentiate between IL-1beta-treated tissues and controls. The analysis revealed significant spectral differences, particularly in the amide band region, which suggests variations in protein content. These preliminary findings will be expanded into a more thorough investigation of the composition of IL-1beta-treated tissues.