BACKGROUND: Portable retinal cameras and deep learning (DL) algorithms are novel tools adopted by diabetic retinopathy (DR) screening programs. Our objective is to evaluate the diagnostic accuracy of a DL algorithm and the performance of portable handheld retinal cameras in the detection of DR in a large and heterogenous type 2 diabetes population in a real-world, high burden setting. METHOD: Participants underwent fundus photographs of both eyes with a portable retinal camera (Phelcom Eyer). Classification of DR was performed by human reading and a DL algorithm (PhelcomNet), consisting of a convolutional neural network trained on a dataset of fundus images captured exclusively with the portable device; both methods were compared. We calculated the area under the curve (AUC), sensitivity, and specificity for more than mild DR. RESULTS: A total of 824 individuals with type 2 diabetes were enrolled at Itabuna Diabetes Campaign, a subset of 679 (82.4%) of whom could be fully assessed. The algorithm sensitivity/specificity was 97.8 % (95% CI 96.7-98.9)/61.4 % (95% CI 57.7-65.1); AUC was 0·89. All false negative cases were classified as moderate non-proliferative diabetic retinopathy (NPDR) by human grading. CONCLUSIONS: The DL algorithm reached a good diagnostic accuracy for more than mild DR in a real-world, high burden setting. The performance of the handheld portable retinal camera was adequate, with over 80% of individuals presenting with images of sufficient quality. Portable devices and artificial intelligence tools may increase coverage of DR screening programs.
BACKGROUND: Portable retinal cameras and deep learning (DL) algorithms are novel tools adopted by diabetic retinopathy (DR) screening programs. Our objective is to evaluate the diagnostic accuracy of a DL algorithm and the performance of portable handheld retinal cameras in the detection of DR in a large and heterogenous type 2 diabetes population in a real-world, high burden setting. METHOD: Participants underwent fundus photographs of both eyes with a portable retinal camera (Phelcom Eyer). Classification of DR was performed by human reading and a DL algorithm (PhelcomNet), consisting of a convolutional neural network trained on a dataset of fundus images captured exclusively with the portable device; both methods were compared. We calculated the area under the curve (AUC), sensitivity, and specificity for more than mild DR. RESULTS: A total of 824 individuals with type 2 diabetes were enrolled at Itabuna Diabetes Campaign, a subset of 679 (82.4%) of whom could be fully assessed. The algorithm sensitivity/specificity was 97.8 % (95% CI 96.7-98.9)/61.4 % (95% CI 57.7-65.1); AUC was 0·89. All false negative cases were classified as moderate non-proliferative diabetic retinopathy (NPDR) by human grading. CONCLUSIONS: The DL algorithm reached a good diagnostic accuracy for more than mild DR in a real-world, high burden setting. The performance of the handheld portable retinal camera was adequate, with over 80% of individuals presenting with images of sufficient quality. Portable devices and artificial intelligence tools may increase coverage of DR screening programs.
Authors: Fernando Korn Malerbi; Amaury Lelis Dal Fabbro; João Paulo Botelho Vieira Filho; Laercio Joel Franco Journal: Diabetes Res Clin Pract Date: 2020-08-21 Impact factor: 5.602
Authors: Yuchen Xie; Quang D Nguyen; Haslina Hamzah; Gilbert Lim; Valentina Bellemo; Dinesh V Gunasekeran; Michelle Y T Yip; Xin Qi Lee; Wynne Hsu; Mong Li Lee; Colin S Tan; Hon Tym Wong; Ecosse L Lamoureux; Gavin S W Tan; Tien Y Wong; Eric A Finkelstein; Daniel S W Ting Journal: Lancet Digit Health Date: 2020-04-23
Authors: Helen K Li; Mark Horton; Sven-Erik Bursell; Jerry Cavallerano; Ingrid Zimmer-Galler; Mathew Tennant; Michael Abramoff; Edward Chaum; Debra Cabrera Debuc; Tom Leonard-Martin; Marc Winchester; Mary G Lawrence; Wendell Bauman; W Kelly Gardner; Lloyd Hildebran; Jay Federman Journal: Telemed J E Health Date: 2011-10-04 Impact factor: 3.536
Authors: Jose R Davila; Sabyasachi S Sengupta; Leslie M Niziol; Manavi D Sindal; Cagri G Besirli; Swati Upadhyaya; Maria A Woodward; Rengaraj Venkatesh; Alan L Robin; Joseph Grubbs; Paula Anne Newman-Casey Journal: Ophthalmologica Date: 2017-07-05 Impact factor: 3.250
Authors: Ângela J Ben; Jeruza L Neyeloff; Camila F de Souza; Ana Paula O Rosses; Aline L de Araujo; Adriana Szortika; Franciele Locatelli; Gabriela de Carvalho; Cristina R Neumann Journal: Appl Health Econ Health Policy Date: 2020-02 Impact factor: 2.561
Authors: Alexander S Himstead; Janani Prasad; Sean Melucci; Kevin M Gustafson; Paul E Israelsen; Andrew Browne Journal: Int J Ophthalmol Date: 2022-06-18 Impact factor: 1.645