ABSTRACT Our parent Phase II SBIR project seeks to develop effective diagnostic solutions for glaucoma, which is one of the leading causes of adult blindness in the United States and globally. Specific tasks undertaken are the collection of ocular blood flow data using our proprietary XyCAM technology from 150 human subjects in the glaucoma clinic at baseline and following hyperoxia, and the subsequent analysis of acquired data to identify biomarkers for early diagnosis of glaucoma. In particular, the plan is to evaluate the suitability of our approach in up to 30 subjects with a high degree of myopia (refractive error of more than 6 Diopters). We are requesting this administrative supplement to the parent grant to overcome certain data analysis challenges that have become apparent since we started the project. Notable among these is the need for registration of data acquired by the XyCAM to the data acquired by other instruments, and also to the data acquired by the XyCAM at other time points (such as post-hyperoxia) so that data analysis can be conducted at a regional level in a reliable and standardized manner. We propose to develop certain automated techniques by utilizing computer vision and machine learning approaches to support analyses of large volumes of data as will be collected during this study. In addition to these proposed activities facilitating project goals in the short term, we also anticipate that the developed software will prove useful more broadly in the future for multi-modal or longitudinal comparison of XyCAM-generated ocular blood flow data.

RELEVANCE TO PUBLIC HEALTH The goal of this administrative supplement is to develop data registration techniques to support effective and accurate analysis of region-specific ocular blood flow and other image data being collected from the eyes of patients with and without glaucoma at baseline and under hyperoxia conditions. The developed techniques will enable rapid and accurate comparisons between data sets and also permit combining region-specific ocular blood flow metrics as needed for this project, and also more broadly enable longitudinal research in the long run.