Cristina Peris-Martínez1, Abhay Shaha2, Warren Clarida3, Ryan Amelon3, María C Hernáez-Ortega4, Amparo Navea5, Jesús Morales-Olivas6, Rosa Dolz-Marco7, Pablo Pérez-Jordá6, Frank Verbraak8, Amber A van der Heijden9. 1. FISABIO Oftalmología Médica (FOM), Valencia, España; Universidad de Valencia, Valencia, España. Electronic address: cristinaperismartinez0@gmail.com. 2. FISABIO Oftalmología Médica (FOM), Valencia, España; Universidad de Valencia, Valencia, España; IDx Technologies Inc., Coralville, United Sates of America; European Innovative Biomedicine Institute (EIBI), Castro-urdiales, España; Instituto de la retina, Valencia, España; Universidad Cardenal Herrera CEU, Valencia, España; Oftalvist, Valencia, España; Departamento de Oftalmología, VUmc, Centros Médicos de la Universidad de Ámsterdam, Ámsterdam, Países Bajos; Departamento de Medicina General y Geriátrica, Centro Médico de la Universidad VU, Ámsterdam, Países Bajos; Instituto de Investigación en Salud Pública de Ámsterdam, Centro Médico de la Universidad VU, Ámsterdam, Países Bajos. 3. IDx Technologies Inc., Coralville, United Sates of America. 4. European Innovative Biomedicine Institute (EIBI), Castro-urdiales, España. 5. Instituto de la retina, Valencia, España; Universidad Cardenal Herrera CEU, Valencia, España. 6. FISABIO Oftalmología Médica (FOM), Valencia, España. 7. Oftalvist, Valencia, España. 8. Departamento de Oftalmología, VUmc, Centros Médicos de la Universidad de Ámsterdam, Ámsterdam, Países Bajos. 9. Departamento de Medicina General y Geriátrica, Centro Médico de la Universidad VU, Ámsterdam, Países Bajos; Instituto de Investigación en Salud Pública de Ámsterdam, Centro Médico de la Universidad VU, Ámsterdam, Países Bajos.
Abstract
BACKGROUND AND OBJECTIVE: To compare the diagnostic performance of an autonomous diagnostic artificial intelligence (AI) system for the diagnosis of derivable diabetic retinopathy (RDR) with manual classification. MATERIALS AND METHODS: Patients with type 1 and type 2 diabetes participated in a diabetic retinopathy (DR) screening program between 2011-2012. 2 images of each eye were collected. Unidentifiable retinal images were obtained, one centered on the disc and one on the fovea. The exams were classified with the autonomous AI system and manually by anonymous ophthalmologists. The results of the AI system and manual classification were compared in terms of sensitivity and specificity for the diagnosis of both (RDR) and diabetic retinopathy with decreased vision (VTDR). RESULTS: 10,257 retinal inages of 5,630 eyes of 2,680 subjects were included. According to the manual classification, the prevalence of RDR was 4.14% and that of VTDR 2.57%. The AI system recorded 100% (95% CI: 97-100%) sensitivity and 81.82% (95% CI: 80 -83%) specificity for RDR, and 100% (95% CI: 95-100%) of sensitivity and 94.64% (95% CI: 94-95%) of specificity for VTDR. CONCLUSIONS: Compared to the manual classification, the autonomous diagnostic AI system registered a high sensitivity (100%) and specificity (82%) in the diagnosis of RDR and macular edema in people with diabetes. Due to its immediate diagnosis, the autonomous diagnostic AI system can increase the accessibility of RDR screening in primary care settings.
BACKGROUND AND OBJECTIVE: To compare the diagnostic performance of an autonomous diagnostic artificial intelligence (AI) system for the diagnosis of derivable diabetic retinopathy (RDR) with manual classification. MATERIALS AND METHODS:Patients with type 1 and type 2 diabetes participated in a diabetic retinopathy (DR) screening program between 2011-2012. 2 images of each eye were collected. Unidentifiable retinal images were obtained, one centered on the disc and one on the fovea. The exams were classified with the autonomous AI system and manually by anonymous ophthalmologists. The results of the AI system and manual classification were compared in terms of sensitivity and specificity for the diagnosis of both (RDR) and diabetic retinopathy with decreased vision (VTDR). RESULTS: 10,257 retinal inages of 5,630 eyes of 2,680 subjects were included. According to the manual classification, the prevalence of RDR was 4.14% and that of VTDR 2.57%. The AI system recorded 100% (95% CI: 97-100%) sensitivity and 81.82% (95% CI: 80 -83%) specificity for RDR, and 100% (95% CI: 95-100%) of sensitivity and 94.64% (95% CI: 94-95%) of specificity for VTDR. CONCLUSIONS: Compared to the manual classification, the autonomous diagnostic AI system registered a high sensitivity (100%) and specificity (82%) in the diagnosis of RDR and macular edema in people with diabetes. Due to its immediate diagnosis, the autonomous diagnostic AI system can increase the accessibility of RDR screening in primary care settings.
Authors: Soledad Jimenez-Carmona; Pedro Alemany-Marquez; Pablo Alvarez-Ramos; Eduardo Mayoral; Manuel Aguilar-Diosdado Journal: J Clin Med Date: 2021-12-21 Impact factor: 4.241