Integration of Machine Learning in Surface-Enhanced Raman Spectroscopy Biosensor for Biomedical Applications

초록

Surface-enhanced Raman scattering (SERS) is an optical analytical technique that enables the detection of specifc molecules with high sensitivity via plasmonic efect of metal nanostructures. Despite its advantages in sensing various biomolecules, the difculties in establishing reliable SERS-active substrates, as well as the complexity of interpreting SERS spectra, hin- der the practical applications of SERS-based biosensors in the biomedical feld. Recent advancements in machine learning (ML) technology have facilitated data analysis, thereby reducing these limitations of SERS-based biosensors. In this review article, the introduction of ML in the development of SERS biosensors for diagnostic platforms will be discussed. Firstly, a brief overview of the ML algorithm used in the SERS study is introduced. Two main applications of ML in SERS biosensors, ML-based design of novel SERS-active nanostructures and ML-assisted data analysis of SERS signals, will be described next, and the future perspectives and challenges of ML-integrated SERS sensors in the biomedical feld will be presented.

키워드