Project Summary Glaucoma is a leading cause of vision morbidity and blindness worldwide. Early disease detection, sensitive monitoring of progression and prediction of future glaucoma progression are crucial to allow timely and personalized treatment for preservation of vision. The introduction of ocular imaging technologies significantly improves these capabilities, but in clinical practice there are still substantial challenges. At certain stages of the disease severity spectrum, particularly in the early stage and in advanced disease, there are a variety of issues that must be confronted that change over the course of the disease, including large between-subject variability, inherent measurement variability, image quality, varying dynamic ranges of measurements, minimal measurable level of tissues, etc. In addition, differences between optical coherence tomography (OCT) devices cause difficulties in clinical patient care as scan data are not interchangeable. This is due to differences in signal and image acquisition and processing, as well as the proprietary nature of the device software. In this proposal, we build on our long-standing contribution to ocular imaging and propose novel and sensitive means to detect glaucoma and identify and predict progression that are optimized to the various stages of disease severity. We will use advanced signal and image processing and image analysis techniques for OCT, a leading ocular imaging technology, to improve and harmonize images agnostic to OCT acquisition device. Commonly used parameters provided by the technologies and newly developed parameters found to have good diagnostic potential will be analyzed across the entire disease severity spectrum to identify optimal metrics for each stage of the disease. We will use state-of-the-art machine learning methods, including deep learning and diffusion modelling analysis approaches, to identify structural features embedded within OCT images that are associated with glaucoma and its progression without any a priori assumptions. This will provide novel insight into structural information and has shown very encouraging results to date. We will use recently developed analytical techniques including multidimensional information compression analysis (MICA) utilizing cluster structure function, a compression-based criterion function for optimizing visualization and clustering via low-dimensional embeddings of the OCT images as well as enhanced machine and deep learning models incorporating clinical features with OCT and visual fields to significantly improve glaucoma detection and the identification of progression and prediction of future disease progression. This program will advance the use of structural, functional, and other clinical information to produce a significant impact on the clinical management of subjects with glaucoma.