Systems and methods for automatically generating descriptions of retinal images

摘要

Embodiments disclose systems and methods that aid in screening, diagnosis and/or monitoring of medical conditions. The systems and methods may allow, for example, for automated identification and localization of lesions and other anatomical structures from medical data obtained from medical imaging devices, computation of image-based biomarkers including quantification of dynamics of lesions, and/or integration with telemedicine services, programs, or software.

主权项

1. A computing system for automated generation of descriptors of local regions within a retinal image, the computing system comprising:
one or more hardware computer processors; and one or more storage devices configured to store software instructions configured for execution by the one or more hardware computer processors in order to cause the computing system to:
access the retinal image;generate a first filtered image using the retinal image, and a morphological filter, computed over a first geometric shape, wherein the first geometric shape is circular or a regular polygon;generate a second filtered image using the retinal image and a morphological filter, computed over a second geometric shape, wherein the second geometric shape has either an elongated or elliptical structure;generate difference image pixel values by computing differences between a plurality of first filtered image pixel values and plurality of second filtered image pixel values; andassign the difference image pixel values as descriptor values, each descriptor value corresponding to a given pixel location of the said retinal image.