Skip to content Skip to navigation

Quantiative Retinal Imaging (QRI)

Imaging is a major component of evaluation of the retina. Beyond fundus photography, the advent of optical coherence tomography (OCT) is revolutionizing practice by permitting high-resolution, three-dimensional imaging of the retina. OCT images contain an enormous amount of information that characterize the phenotype of retinal diseases, but presently only a fraction of the information in OCT images is extracted and used for clinical decision making. Our Quantitative Retinal Image Group is developing methods to optimally leverage latent quantitative information in OCT images to enable precision care in Ophthalmology. Our group is working in the following areas, which logically comprise a pipline of processing these large datasets leading from the raw image to disease assessment and clinical decision support:

  1. OCT Image Denoising: We are developing methods to remove noise from OCT images to permit optimal visualization, segmentation, and quantitation of retinal anatomic structures and disease features

  2. Visualization: We are developing novel methods for visualizing volumetric retinal anatomic structures and imaging disease phenotypes. For example, we recently developed the Restricted Summed Voxel Projection (RSVP) method to improve the visualization of drusen and geographic lesions in AMD. We also developed a false color method of enhancing visualization of drusen and geographic atrophy and disambiguating these lesions in a single image display.

  3. Automated Retinal Segmentation: We developed automated methods to segment the multi-layered structure of the retina. This is a crucial initial step to quantifying disease features. We are also developing methods to automatically segment disease features, including drusen and geographic atrophy and to estimate the thickness of the retinal fiber layer near the optic disk.

  4. Segmentation and Quantitation of Retinal Disease Features: We are extracting and quantifying a broad variety of features of retinal desieases, specifically AMD, glaucoma, diabetic retinopathy, and retinitis pigmentosa. Use then use these features in machine learning models to assess disease status and to predict clinical outcomes and treatment responose. In recent work, we extracted a set of key features from OCT images which, in combination with selected clinical features, can predict AMD progression from dry to wet stages of the disease.

  5. Statistical Modeling and Applications: We are developing a number of clincally-useful applications that exploit our quantitative retinal imaging methods. These include:

    • En-face visualization of drusen and geographic atrophy

    • Characterization of severity of geographic atrophy

    • Prediction of AMD progression

    • Characterization of glaucoma severity and risk assessment