Exploring radiation induced lung toxicity with Raman spectroscopy

In this work, I first explore the efficacy of Raman measurements of lung tissue after it has been cryopreserved twice versus the standard practice of only one cycle. It was found the resulting spectral measurements of two cryo-cycle tissue, while being affected in count but not peak position when compared to only one cryo-cycle, was of a sufficient information quality for further experimentation. As such, the main focus of this work was to develop a method using Raman spectroscopy to measure and detect radiation induced toxicity in murine lungs. Analysis of Raman tissue data is achieved with a combination of basis restricted and unrestricted non-negative matrix factorization. Using quadratic discriminant analysis, the scores produced by non-negative matrix factorization was classified with an accuracy up to 56%, average specificity of ∼85 and average sensitivity of ∼55% . To conclude, the development of a highly accurate spectroscopic-based method for classifying radiation induced pathology in murine lungs paves the way for future work in predicting pathology development in the early stages after irradiation.Science, Irving K. Barber Faculty of (Okanagan)Computer Science, Mathematics, Physics and Statistics, Department of (Okanagan)Graduat