Deep Learning Applied to Raman Spectroscopy for the Detection of Microsatellite Instability/MMR Deficient Colorectal Cancer

Simple Summary Colorectal cancer has several disease pathways which have implications for how patients are monitored and treated. One important pathway is caused by deficiencies to genes responsible for repairing pre-cancerous cells. Methods to detect these deficiencies exist, but are not implemented as often as recommended and could be improved. Raman spectroscopy is a technique that could provide such an improvement, having shown potential in other areas of cancer research. The full potential of Raman datasets may be achieved by exploiting modern machine learning models. We evaluated a small colorectal tissue dataset to assess the viability of some common machine learning techniques to detect different colorectal cancer pathways. We find that Raman spectroscopy in conjunction with machine learning could be a viable means of improving screening and potentially diagnostic tools and warrants further research with larger sample sizes. Defective DNA mismatch repair is one pathogenic pathway to colorectal cancer. It is characterised by microsatellite instability which provides a molecular biomarker for its detection. Clinical guidelines for universal testing of this biomarker are not met due to resource limitations; thus, there is interest in developing novel methods for its detection. Raman spectroscopy (RS) is an analytical tool able to interrogate the molecular vibrations of a sample to provide a unique biochemical fingerprint. The resulting datasets are complex and high-dimensional, making them an ideal candidate for deep learning, though this may be limited by small sample sizes. This study investigates the potential of using RS to distinguish between normal, microsatellite stable (MSS) and microsatellite unstable (MSI-H) adenocarcinoma in human colorectal samples and whether deep learning provides any benefit to this end over traditional machine learning models. A 1D convolutional neural network (CNN) was developed to discriminate between healthy, MSI-H and MSS in human tissue and compared to a principal component analysis-linear discriminant analysis (PCA-LDA) and a support vector machine (SVM) model. A nested cross-validation strategy was used to train 30 samples, 10 from each group, with a total of 1490 Raman spectra. The CNN achieved a sensitivity and specificity of 83% and 45% compared to PCA-LDA, which achieved a sensitivity and specificity of 82% and 51%, respectively. These are competitive with existing guidelines, despite the low sample size, speaking to the molecular discriminative power of RS combined with deep learning. A number of biochemical antecedents responsible for this discrimination are also explored, with Raman peaks associated with nucleic acids and collagen being implicated.