Research progress of AI assisted nano-SERS technology in rapid identification of multi species oral pathogenic biofilms.

Complex multispecies microbial communities that are oral pathogenic biofilms are linked to significant oral infections, such as dental caries, periodontitis, peri-implantitis, and chronic endodontic infection. In order to diagnose and treat these biofilms effectively, it is important that the microbial composition contained within the biofilm is quickly and correctly identified. Traditional techniques like culture-based assays, polymerase chain reaction and immunological techniques are normally very time consuming, labor intensive as well as have limited capacity to identify more than one species at a time. Recent developments of surface-enhanced Raman spectroscopy (SERS) assisted by nano-engineered plasmonic substrates offer a sensitive and label-free detection method of molecular fingerprints of bacterial cells and biofilm components. Coupled with artificial intelligence (AI) and machine learning algorithms allows automated processing of complex SERS spectral samples, which enhance fast detection and categorization of a variety of pathogens in polymicrobial biofilms. New applications on microfluidic systems, portable diagnostic systems and point-of-care systems further improve the possibility of this technology in clinical applications. The review outlines the biological properties and clinical relevance of oral biofilms, the progress in the design of nano-SERS substrates, and AI-supported spectral analysis methods. The existing problems, such as the reproducibility of the substrate, the variability of the spectral, the lack of datasets, and the obstacles of clinical translation are critically debated. The interplay of nanotechnology, SERS, and AI have great potential in the creation of the next-generation diagnostic devices that will be able to detect pathogenic biofilms in the mouth within a clinical practice promptly, sensitively, and multiplexed.