Purchase this article with an account.
Praveen J Patel, Pearse Andrew Keane, Carlota M Grossi, Paul J Foster, Charles A Reisman, Qi Yang, Chih-Yang Chang, Kinpui Chan, UK Biobank Eyes and Vision Consortium; Feasibility of Rapid, Automated Analysis of Macular Thickness in a Population Study: Approach to Analysis of Spectral-domain OCT Images from 67,321 Subjects in the UK Biobank Study. Invest. Ophthalmol. Vis. Sci. 2014;55(13):670.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To report the feasibility of automated segmentation of spectral-domain optical coherence tomography (OCT) images collected as part of the ocular module of the UK Biobank dataset. The primary outcome was the number and proportion of images successfully segmented using remote access to the Topcon 3D OCT-1000 Mark II OCT images collected and stored by UK Biobank. This research has been conducted using the UK Biobank Resource.
A multi-disciplinary team from industry, university and NHS was established from the UK and US to access and analyze stored Topcon 3D OCT-1000 Mark II images collected as part of the UK Biobank (UKBB) study. As part of the UKBB data access rules and procedures for bulk data, the stored OCT files (source data) could not be copied, stored or removed outside the local UKBB network. Instead, researchers are given access to computers at the central UKBB data repository via remote, secure login and can then install any analysis software needed on the UKBB computers. A copy of the stored OCT image file needs to be fetched before running the segmentation analysis software. The derived data are then extracted, after which the OCT image file is deleted. Multiple logins can be implemented in parallel, increasing the processing throughput. After logging in remotely to the UKBB network, we used TABSTM (Topcon Advanced Boundary Segmentation) to analyze the stored OCT images.
A total of 134,642 macular OCT images were available for processing from 134,642 eyes of 67,321 patients. Of these images, 134,611 images were successfully processed with 31 images failing segmentation analysis due to corrupted OCT files or withdrawal of subject consent for UKBB study participation (successful analysis in 99.98% of images). The mean (±SD) age of patients was 57 (±8) years with 36,623 females and 30,698 males. Average time taken to call up an image and complete segmentation analysis was approximately 120 seconds per data set per login. The average signal strength (Q factor) for all images was 65 (±13).
Rapid, remote, automated analysis is a feasible approach to analyze macular thickness measurements from stored OCT images. This approach may be used to analyze and extract OCT derived macular thickness measurements in future population studies.
This PDF is available to Subscribers Only