Identification and examination of protein compounds utilizing Raman scattering and machine learning techniques

Proteins are important components of the human body, playing a crucial role in the functioning of cells: they catalyze chemical reactions and form cellular structures. An imbalance in protein metabolism can have serious consequences, such as impaired immunity and changes in glandular activity. Detection of various biological compounds is challenging due to their complex intermolecular relationships, and traditional methods such as immunoassays and chromatography may not always provide accurate results. The presented research aims to overcome these limitations by introducing an approach that combines Raman spectroscopy and machine learning to accurately identify protein compounds. This technique aims to minimize errors in quantitative and qualitative analysis and enable systematic investigation of protein compounds.The results obtained during testing of the algorithm on data obtained during experiments indicate the possibility of using this technique for more than 10 analyte substances and achieving an accuracy of over 90%. The methodology for working with experimental data using artificial intelligence tools thus formed can form the basis for creating effective platforms and devices for use not only in the scientific field, but also in the fields of medicine, agriculture, and food safety.