Rapid Identification of Candida Auris by Raman Spectroscopy Combined with Deep Learning

Candida auris, first reported in Japan in 2009, is an emerging global health threat. This yeast is often multidrug-resistant, challenging to identify using standard laboratory methods, and has a propensity to cause nosocomial outbreaks despite heightened infection prevention and control measures in healthcare settings. This study introduces a portable, reagentless platform based on counter-propagating Gaussian beam Raman spectroscopy (CPGB-RS) integrated with deep learning spectral analysis. CPGB-RS provides swift and accurate identification and differentiation of C. auris from the most prevalent pathogenic Candida species: albicans, glabrata, and tropicalis, with a sensitivity of 97% and specificity of 99% when analyzing cultures. The species differentiation stems from distinct variations in the Raman spectra that arise from differences in cell wall composition (β-glucan, chitin, and mannoprotein), cell membrane components (ergosterol) and cellular energy states (mitochondrial cytochromes b and c). The platform allows automated screening on a molecular basis with software-generated diagnostic read-out in 2 minutes making it well-suited for clinical applications. Further, the technology can be extended to evaluate the effectiveness of antifungal agents leading to better patient outcomes.