Validation of Optical Coherence Tomography Retinal Segmentation in Neurodegenerative Disease

• DOI: 10.1167/tvst.8.5.6
• Publication Date: 2019-09-11T13:31:29Z
• Authors: Bryan M. Wong; Richard W. Cheng; Efrem D. Mandelcorn; Edward Margolin; Sherif El-Defrawy; Peng Yan; Anna T. Santiago; Elena Leontieva; Wendy Lou; ; Wendy Hatch; Christopher Hudson
• Keywords: 03 medical and health sciences; 0302 clinical medicine; Articles

This study assessed agreement between an automated spectral-domain optical coherence tomography (SD-OCT) retinal segmentation software and manually corrected to validate its use in a prospective clinical of neurodegenerative diseases (NDD).The sample comprised 30 subjects with NDD, including vascular cognitive impairment, frontotemporal dementia, Parkinson's disease, Alzheimer's disease. Macular SD-OCT scans were acquired segmented using Heidelberg Spectralis. For the central foveal B scan each eye, eight lines examined determine proportion line that erroneously delineated. Errors four all spanning 6-mm circle centered on foveola. Mean volume thickness measurements for layers (total retina, nerve fiber layer [RNFL], inner layers, outer layers) obtained before after correction.The plexiform had one lowest mean error ratios (2%), while RNFL highest (23%). Agreement trained observer was excellent (ICC > 0.98) layers. differences observers ranged from -0.003 0.006 mm3. -1.855 1.859 μm.Despite occasional small errors software-generated sublayer segmentation, software-derived observer-corrected measurements.Automated generates valid NDD subjects, thereby avoiding need correct nonobvious delineation errors.