Multimodal Optical Imaging for Tissue Characterization and Disease Diagnosis
AuthorSawyer, Travis William
optical coherence tomography
AdvisorBarton, Jennifer K.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractEpithelial cancers are among the most dangerous forms of cancer. Of this broad group of disease, ovarian and esophageal cancer are particularly deadly, with five-year survival rates of less than 50% and 20% respectively. The primary cause of this low survival rate is due to predominantly late diagnosis. Diagnosis at early stages leads to over 90% 5-year survival rates for ovarian cancer and over 40% for esophageal cancer, but fewer than 15% of cases for these two cancers are detected early. Screening is complicated by non-specific or complete lack of symptoms, as well as the heterogeneity of the diseases. For both esophageal and ovarian cancers, many screening tests including imaging, physical examination, and blood markers tests have been investigated; however, at this time no routine screening is recommended in average-risk patients. This study evaluates the feasibility and design of instruments to use multimodal optical imaging to improve ovarian and esophageal cancer screening. This includes optical coherence tomography (OCT), multiphoton microscopy (MPM), and wide-field fluorescence imaging. The study is subdivided into four sections. In the first two sections, advanced algorithms and processing techniques are presented for rapid analysis and quantitative diagnostic evaluation for optical coherence tomography images of ovarian cancer. The results show promise for automatic processing of OCT images using segmentation, combined with highly accurate diagnostic performance in identifying diseased tissue using texture features of OCT images. The third section details the application of MPM and wide-field fluorescence imaging using exogenous contrast agents to evaluate and classify tissue health. Two tissue studies using a mouse model of ovarian cancer are presented: one ex vivo and one in vivo study. The results of both demonstrate that these modalities provide high contrast for identifying diseased tissue and that the combination of these two modalities show an improvement in diagnostic performance over a single modality. Finally, the study concludes with the design for a multimodal forward-viewing esophageal endoscope using optical coherence tomography and autofluorescence imaging. The design implements a piezo-scanning fiber to deliver the light, and spatially separates the OCT and fluorescence return signal. The design provides high resolution and is compatible with working channels in existing gastroscopes to enable easy clinical translation.
Degree ProgramGraduate College