Approximately 80% of chronic wounds will be colonized by some form of biofilm.  The development of polymicrobial biofilms further delays healing of diabetic foot ulcers (DFUs), burns and skin ulcers amongst other chronic wound types. Due to the robust nature of biofilms, susceptibility to antimicrobials is significantly reduced making such infections extremely difficult to treat. As a result, DFUs are associated with a five-year mortality rate in 40% of patients, whilst the cost of treatment and amputation also imposes a significant economic burden on the NHS.   Raman spectroscopy is a powerful analytical method that uses inelastic photon scattering to probe molecular vibrational modes. This technique is label-free, rapid, and non-destructive, allowing real-time analysis of samples without chemical labelling or structural disruption. Raman spectroscopy was used to construct unique spectral profiles for three common skin colonizers and wound-associated pathogens: Pseudomonas aeruginosa, Staphylococcus aureus and Candida albicans.  Raman spectral data from 24-hour single, dual and multispecies biofilms were pre-processed and showed phenotypic differences in the fingerprint region (500-1800cm-1). However, subtle variations in Raman spectra are often difficult to discern by eye, making machine learning a valuable tool for spectral discrimination. Therefore, Raman spectra were then subjected to principal component analysis (PCA) which revealed spectral heterogeneity across different regions of the same biofilm. Raman spectral data from the biofilms will be used to train machine learning models—including Partial Least Squares Discriminant Analysis (PLS-DA), Random Forest and Support Vector Machine—to evaluate whether microbial species can be reliably discriminated within polymicrobial biofilms.