Broadband Coherent Raman Microspectroscopy for the Investigation of Head and Neck Cancer advancing ultrafast spectral Histopathology

Nonlinear spectroscopic imaging techniques such as coherent Raman scattering (CRS) have proven to be powerful tools for biomedical research, providing multifaceted imaging data that holds great potential for medical diagnostics and therapy. Among these, broadband implementations such as broadband CARS and broadband SRS play a promising role in this field as they enable label-free and chemically specific imaging with subcellular resolution of a broad vibrational spectrum at high speed. In this manuscript, we explore the potential of spectral CRS techniques for the identification and classification of head and neck cancer based on their vibrational signatures. Several tissue sections from different patients were investigated by broadband CARS and a qualitative analysis already showed a strong discrimination capability. A preliminary ‘proof-of-concept’ AI-based model was also developed to automate the classification task. Furthermore, we demonstrate the application of broadband CARS and SRS in a high-speed configuration for rapid imaging of biological tissue slices. The setup achieves a dwell time of 134 µs, one order of magnitude faster than comparable multiplex CARS setups in literature. These results highlight the capability of CRS methods to advance ultrafast spectral histopathology, offering new opportunities for real-time, label-free diagnostics.