| dc.description.abstract | Raman spectroscopy is an extremely sensitive technique that has been used by many research groups to differentiate between normal and abnormal changes in biological samples. Previous results have shown that when Raman spectroscopy is used to detect cervical dysplasia, incorporating normal hormonal variations increases the classification accuracy from 88% to 94%. In this work, other variables that may affect the cervix (previous disease, proximity to disease, race, ethnicity, body mass index (BMI), obstetric history, health insurance) were evaluated. Raman spectra were acquired from patients during a routine cervical cancer screening. Accounting for previous disease, proximity to disease, BMI, and obstetric history prior to developing classification algorithms led to an increase in correctly identifying malignant areas of the cervix (to 100%). This result demonstrates that when normal variations are considered, Raman spectroscopy can be successfully implemented in diverse patient populations to detect cervical dysplasia. Raman spectra were also acquired from the cervix of pregnant mice and patients during and after their pregnancy to validate the use of this technique for detecting biochemical changes in the cervix. Spectra from mice were compared to the cervix’s structural and morphological properties. Lipid content and collagen organization significantly changed throughout pregnancy, as observed in the Raman, structural and morphological results. In the human study, Raman spectra contained a large amount of variation, particularly in peaks corresponding to protein, lipid, and collagen content. Raman spectroscopy can be used to non-invasively monitor these changes. The results from this work demonstrate that Raman spectroscopy is a versatile tool that can be utilized to successfully detect cervical dysplasia and examine the changes in the cervix during pregnancy. | |
