Description:
Summary
Measurement of retinal fluid volume could improve diagnosis and tracking of diabetic macular edema (DME); however, current methods rely on two-dimensional measurements, which have limited accuracy. The current technology uses volumetric OCT/OCT angiography scans to provide a three-dimensional and more accurate retinal fluid volume, potentially improving the ability to track DME.
Technology Overview
DME is the most common cause of vision loss in diabetic retinopathy (DR). Segmentation and quantification of retinal fluid cysts provide a more specific biomarker for DME, but traditional optical coherence tomography (OCT) cross-sections can only provide a limited two-dimensional assessment. Oregon Health & Science University researchers have developed a deep-learning algorithm called Retinal Fluid Segmentation Network (ReF-Net), which combines OCT and OCT angiography to provide a more accurate three-dimensional volumetric representations of retinal fluid volume (see Figure). The algorithm can also operate with volumetric OCT scans only, if OCT angiography scans are inaccessible. This algorithm shows a high accuracy for retinal fluid segmentation and is capable of overcoming common artifacts such as shadows and pupil vignetting. The ReF-Net algorithm could be utilized by reading centers to improve the diagnostic accuracy of diabetic macular edema by OCT systems.
Publication
Guo Y, Hormel TT, Xiong H, Wang J, Hwang TS, Jia Y. Automated segmentation of retinal fluid volume from structural and angiographic optical coherence tomography using deep learning. Translational Vision Science & Technology 2020; 9(2). Link
Licensing Opportunity
This technology is available for licensing.