Duke University

Stephanie J Chiu

Cynthia A Toth

Sina Farsiu

Joseph A Izatt

Xiao T Li

Peter Christopher Nicholas

Katrina Winter

Vincent Tai

Kai Seely

Josheph A Izatt

Ryan Imperio

A retinal vessel shadow view optical coherence tomography (RVSV-OCT) image can be created by receiving, at an enhanced OCT processing system, volumetric OCT scan of a patient. The system can segment the volumetric OCT scan to determine layer boundaries and delineate a boundary of interest based on the determined layer boundaries of the segmented volumetric OCT scan. En face vascular information can be extracted to create an RVSV-OCT image by determining a first offset from the boundary of interest and a second offset from the boundary of interest; extracting volumetric data from an area between the first offset and the second offset to create a three-dimensional volume; and identifying a two-dimensional surface from the three-dimensional volume, the two-dimensional surface being the RVSV-OCT image. The RVSV-OCT image can be provided for analysis, for example, to evaluate retinal vascular disease in preterm infants at risk for retinopathy of prematurity.

En face retinal vessel shadow view optical coherence tomography images

Disclosed herein are systems and method for segmentation and identification of structured features in images. According to an aspect, a method may include representing an image as a graph of nodes connected together by edges. For example, the image may be an ocular image showing layered structures or other features of the retina. The method may also include adding, to the graph, nodes adjacent to nodes along first and second sides of the graph. The added nodes may have edge weights less than the nodes along the first and second sides of the graph. Further, the method may include assigning start and end points to any of the added nodes along the first and second sides, respectively. The method may also include graph cutting between the start and end points for identifying a feature in the image.

Segmentation and identification of layered structures in images

Segmentation and identification of closed-contour features in images using graph theory and quasi-polar transform are disclosed. According to an aspect, a method includes representing, in a rectangular domain, an image including a feature of interest. Further, the method includes determining a point within the feature of interest. The method also includes transforming the image of the feature from the rectangular domain to a quasi-polar domain based on the point. The quasi-polar domain is represented as a graph of nodes connected together by edges. The method also includes graph cutting the quasi-polar domain to identify the boundary of the feature of interest in the image.

Segmentation and identification of closed-contour features in images using graph theory and quasi-polar transform

Growing community of inventors

Durham, NC, United States of America

Stephanie J Chiu