TY - JOUR
T1 - Comparison of automated and expert human grading of diabetic retinopathy using smartphone-based retinal photography
AU - Kim, Tyson N.
AU - Aaberg, Michael T.
AU - Li, Patrick
AU - Davila, Jose R.
AU - Bhaskaranand, Malavika
AU - Bhat, Sandeep
AU - Ramachandra, Chaithanya
AU - Solanki, Kaushal
AU - Myers, Frankie
AU - Reber, Clay
AU - Jalalizadeh, Rohan
AU - Margolis, Todd P.
AU - Fletcher, Daniel
AU - Paulus, Yannis M.
N1 - Funding Information:
Acknowledgements We would like to thank the Knights Templar Eye Foundation, Research to Prevent Blindness, the Rogers Family Foundation, and the University of Michigan for their support. Additionally, we would like to thank the Eyenuk team for their willingness to partner with us and their continued commitment to ocular research.
Funding Information:
Funding This work was supported by the Knights Templar Eye Foundation Career-Starter Research Grant, the University of Michigan Translational Research and Commercialization for Life Sciences, the University of Michigan Center for Entrepreneurship Dean’s Engineering Translational Prototype Research Fund, the QB3 Bridging the Gap Award from the Rogers Family Foundation, the Bakar Fellows Award, the Chan-Zuckerberg Biohub Investigator Award, the National Eye Institute grant 1K08EY027458, and the University of Michigan Department of Ophthalmology and Visual Sciences.
Publisher Copyright:
© 2020, The Author(s), under exclusive licence to The Royal College of Ophthalmologists.
PY - 2021/1
Y1 - 2021/1
N2 - Purpose: The aim of this study is to investigate the efficacy of a mobile platform that combines smartphone-based retinal imaging with automated grading for determining the presence of referral-warranted diabetic retinopathy (RWDR). Methods: A smartphone-based camera (RetinaScope) was used by non-ophthalmic personnel to image the retina of patients with diabetes. Images were analyzed with the Eyenuk EyeArt® system, which generated referral recommendations based on presence of diabetic retinopathy (DR) and/or markers for clinically significant macular oedema. Images were independently evaluated by two masked readers and categorized as refer/no refer. The accuracies of the graders and automated interpretation were determined by comparing results to gold standard clinical diagnoses. Results: A total of 119 eyes from 69 patients were included. RWDR was present in 88 eyes (73.9%) and in 54 patients (78.3%). At the patient-level, automated interpretation had a sensitivity of 87.0% and specificity of 78.6%; grader 1 had a sensitivity of 96.3% and specificity of 42.9%; grader 2 had a sensitivity of 92.5% and specificity of 50.0%. At the eye-level, automated interpretation had a sensitivity of 77.8% and specificity of 71.5%; grader 1 had a sensitivity of 94.0% and specificity of 52.2%; grader 2 had a sensitivity of 89.5% and specificity of 66.9%. Discussion: Retinal photography with RetinaScope combined with automated interpretation by EyeArt achieved a lower sensitivity but higher specificity than trained expert graders. Feasibility testing was performed using non-ophthalmic personnel in a retina clinic with high disease burden. Additional studies are needed to assess efficacy of screening diabetic patients from general population.
AB - Purpose: The aim of this study is to investigate the efficacy of a mobile platform that combines smartphone-based retinal imaging with automated grading for determining the presence of referral-warranted diabetic retinopathy (RWDR). Methods: A smartphone-based camera (RetinaScope) was used by non-ophthalmic personnel to image the retina of patients with diabetes. Images were analyzed with the Eyenuk EyeArt® system, which generated referral recommendations based on presence of diabetic retinopathy (DR) and/or markers for clinically significant macular oedema. Images were independently evaluated by two masked readers and categorized as refer/no refer. The accuracies of the graders and automated interpretation were determined by comparing results to gold standard clinical diagnoses. Results: A total of 119 eyes from 69 patients were included. RWDR was present in 88 eyes (73.9%) and in 54 patients (78.3%). At the patient-level, automated interpretation had a sensitivity of 87.0% and specificity of 78.6%; grader 1 had a sensitivity of 96.3% and specificity of 42.9%; grader 2 had a sensitivity of 92.5% and specificity of 50.0%. At the eye-level, automated interpretation had a sensitivity of 77.8% and specificity of 71.5%; grader 1 had a sensitivity of 94.0% and specificity of 52.2%; grader 2 had a sensitivity of 89.5% and specificity of 66.9%. Discussion: Retinal photography with RetinaScope combined with automated interpretation by EyeArt achieved a lower sensitivity but higher specificity than trained expert graders. Feasibility testing was performed using non-ophthalmic personnel in a retina clinic with high disease burden. Additional studies are needed to assess efficacy of screening diabetic patients from general population.
UR - http://www.scopus.com/inward/record.url?scp=85084153551&partnerID=8YFLogxK
U2 - 10.1038/s41433-020-0849-5
DO - 10.1038/s41433-020-0849-5
M3 - Article
C2 - 32341536
AN - SCOPUS:85084153551
VL - 35
SP - 334
EP - 342
JO - Eye
JF - Eye
SN - 0950-222X
IS - 1
ER -