Jessica Loo1, Glenn J Jaffe2, Jacque L Duncan3, David G Birch4, Sina Farsiu1,2. 1. Department of Biomedical Engineering, Duke University, Durham, North Carolina. 2. Department of Ophthalmology, Duke University Medical Center, Durham, North Carolina. 3. Department of Ophthalmology, University of California, San Francisco, San Francisco, California; and. 4. Retina Foundation of the Southwest, Dallas, Texas.
Abstract
PURPOSE: To assess the generalizability of a deep learning-based algorithm to segment the ellipsoid zone (EZ). METHODS: The dataset consisted of 127 spectral-domain optical coherence tomography volumes from eyes of participants with USH2A-related retinal degeneration enrolled in the RUSH2A clinical trial (NCT03146078). The EZ was segmented manually by trained readers and automatically by deep OCT atrophy detection, a deep learning-based algorithm originally developed for macular telangiectasia Type 2. Performance was evaluated using the Dice similarity coefficient between the segmentations, and the absolute difference and Pearson's correlation of measurements of interest obtained from the segmentations. RESULTS: With deep OCT atrophy detection, the average (mean ± SD, median) Dice similarity coefficient was 0.79 ± 0.27, 0.90. The average absolute difference in total EZ area was 0.62 ± 1.41, 0.22 mm2 with a correlation of 0.97. The average absolute difference in the maximum EZ length was 222 ± 288, 126 µm with a correlation of 0.97. CONCLUSION: Deep OCT atrophy detection segmented EZ in USH2A-related retinal degeneration with good performance. The algorithm is potentially generalizable to other diseases and other biomarkers of interest as well, which is an important aspect of clinical applicability.
PURPOSE: To assess the generalizability of a deep learning-based algorithm to segment the ellipsoid zone (EZ). METHODS: The dataset consisted of 127 spectral-domain optical coherence tomography volumes from eyes of participants with USH2A-related retinal degeneration enrolled in the RUSH2A clinical trial (NCT03146078). The EZ was segmented manually by trained readers and automatically by deep OCT atrophy detection, a deep learning-based algorithm originally developed for macular telangiectasia Type 2. Performance was evaluated using the Dice similarity coefficient between the segmentations, and the absolute difference and Pearson's correlation of measurements of interest obtained from the segmentations. RESULTS: With deep OCT atrophy detection, the average (mean ± SD, median) Dice similarity coefficient was 0.79 ± 0.27, 0.90. The average absolute difference in total EZ area was 0.62 ± 1.41, 0.22 mm2 with a correlation of 0.97. The average absolute difference in the maximum EZ length was 222 ± 288, 126 µm with a correlation of 0.97. CONCLUSION: Deep OCT atrophy detection segmented EZ in USH2A-related retinal degeneration with good performance. The algorithm is potentially generalizable to other diseases and other biomarkers of interest as well, which is an important aspect of clinical applicability.
Authors: Laurence H M Pierrache; Bas P Hartel; Erwin van Wijk; Magda A Meester-Smoor; Frans P M Cremers; Elfride de Baere; Julie de Zaeytijd; Mary J van Schooneveld; Cor W R J Cremers; Gislin Dagnelie; Carel B Hoyng; Arthur A Bergen; Bart P Leroy; Ronald J E Pennings; L Ingeborgh van den Born; Caroline C W Klaver Journal: Ophthalmology Date: 2016-02-27 Impact factor: 12.079
Authors: J D Eudy; M D Weston; S Yao; D M Hoover; H L Rehm; M Ma-Edmonds; D Yan; I Ahmad; J J Cheng; C Ayuso; C Cremers; S Davenport; C Moller; C B Talmadge; K W Beisel; M Tamayo; C C Morton; A Swaroop; W J Kimberling; J Sumegi Journal: Science Date: 1998-06-12 Impact factor: 47.728
Authors: Samuel G Jacobson; Artur V Cideciyan; Tomas S Aleman; Alexander Sumaroka; Alejandro J Roman; Leigh M Gardner; Haydn M Prosser; Monalisa Mishra; N Torben Bech-Hansen; Waldo Herrera; Sharon B Schwartz; Xue-Zhong Liu; William J Kimberling; Karen P Steel; David S Williams Journal: Hum Mol Genet Date: 2008-05-07 Impact factor: 6.150
Authors: Dibyendu Mukherjee; Eleonora M Lad; Ryan R Vann; Stephanie J Jaffe; Traci E Clemons; Martin Friedlander; Emily Y Chew; Glenn J Jaffe; Sina Farsiu Journal: Invest Ophthalmol Vis Sci Date: 2017-05-01 Impact factor: 4.799
Authors: Emily Y Chew; Traci E Clemons; Glenn J Jaffe; Charles A Johnson; Sina Farsiu; Eleonora M Lad; Robyn Guymer; Philip Rosenfeld; Jean-Pierre Hubschman; Ian Constable; Henry Wiley; Lawrence J Singerman; Mark Gillies; Grant Comer; Barbara Blodi; Dean Eliott; Jiong Yan; Alan Bird; Martin Friedlander Journal: Ophthalmology Date: 2018-10-04 Impact factor: 12.079
Authors: Jacque L Duncan; Wendi Liang; Maureen G Maguire; Isabelle Audo; Allison R Ayala; David G Birch; Joseph Carroll; Janet K Cheetham; Simona Degli Esposti; Todd A Durham; Laura Erker; Sina Farsiu; Frederick L Ferris; Elise Heon; Robert B Hufnagel; Alessandro Iannaccone; Glenn J Jaffe; Christine N Kay; Michel Michaelides; Mark E Pennesi; José-Alain Sahel Journal: Am J Ophthalmol Date: 2020-05-22 Impact factor: 5.488
Authors: Touka Banaee; Rishi P Singh; Kathryn Champ; Felipe F Conti; Karen Wai; Jim Bena; Lucas Beven; Justis P Ehlers Journal: Ophthalmol Retina Date: 2018-01-06