This technology is an artificial intelligence-based platform for improving the quality of images taken from portable OCT devices and detecting age-related macular degeneration.
Current methods for diagnosing ophthalmic disease use optical coherence tomography (OCT). However, these devices are bulky, costly and require expertise to operate. Portable OCT devices offer potential, but images acquired from these devices are lower resolution. These shortcomings have led to a gap in routine eye disease diagnosis and care in under-resourced communities. If left undiagnosed, conditions such as AMD may lead to sight loss.
This technology uses generative adversarial networks (GANs) to improve the quality of images taken from portable OCT devices. These enhanced images are then fed through a deep learning-based method for the detection of AMD.
This technology has been validated with a Duke University dataset containing 221 images from 59 individuals.
Patent pending
IR CU22266
Licensing Contact: Dovina Qu