Application of machine learning-assisted surface-enhanced Raman spectroscopy in medical laboratories: Principles, opportunities, and challenges

Surface-enhanced Raman spectroscopy (SERS) has attracted increasing attention due to its rich molecular information and excellent sensitivity. However, with the advancement of SERS technology, traditional linear data processing methods no longer suffice, prompting the exploration of new approaches to enhance SERS analysis. The progress of machine learning (ML) technologies has shown immense potential in enhancing SERS capabilities through rapid analysis and automated data processing. This review focuses on summarizing advancements in applying various ML algorithms to SERS in recent years, providing guidelines for processing SERS signals and evaluating and interpreting the performance of ML algorithm. In addition, the latest developments of ML-assisted SERS in clinical applications were reported in depth. Finally, the current limitations and challenges of existing SERS research were discussed, which provided insights into future research directions. In sum, this review will lay the foundation for translating ML-assisted SERS from laboratory benches to clinically real-world situations.