Abstract

Diabetic retinopathy (DR) at risk of vision loss (referable DR) needs to be identified by retinal screening and referred to an ophthalmologist. Existing automated algorithms have mostly been developed from images acquired with high cost mydriatic retinal cameras and cannot be applied in the settings used in most low- and middle-income countries. In this prospective multicentre study, we developed a deep learning system (DLS) that detects referable DR from retinal images acquired using handheld non-mydriatic fundus camera by non-technical field workers in 20 sites across India. Macula-centred and optic-disc-centred images from 16,247 eyes (9778 participants) were used to train and cross-validate the DLS and risk factor based logistic regression models. The DLS achieved an AUROC of 0.99 (1000 times bootstrapped 95% CI 0.98–0.99) using two-field retinal images, with 93.86 (91.34–96.08) sensitivity and 96.00 (94.68–98.09) specificity at the Youden’s index operational point. With single field inputs, the DLS reached AUROC of 0.98 (0.98–0.98) for the macula field and 0.96 (0.95–0.98) for the optic-disc field. Intergrader performance was 90.01 (88.95–91.01) sensitivity and 96.09 (95.72–96.42) specificity. The image based DLS outperformed all risk factor-based models. This DLS demonstrated a clinically acceptable performance for the identification of referable DR despite challenging image capture conditions.

Details

Title
Using deep learning to detect diabetic retinopathy on handheld non-mydriatic retinal images acquired by field workers in community settings
Author
Nunez do Rio, Joan M. 1 ; Nderitu, Paul 1 ; Raman, Rajiv 2 ; Rajalakshmi, Ramachandran 3 ; Kim, Ramasamy 4 ; Rani, Padmaja K. 5 ; Sivaprasad, Sobha 6 ; Bergeles, Christos 7 ; Bhende, Pramod 8 ; Surya, Janani 8 ; Gopal, Lingam 8 ; Ramakrishnan, Radha 9 ; Roy, Rupak 10 ; Das, Supita 10 ; Manayath, George 11 ; Vignesh, T. P. 11 ; Anantharaman, Giridhar 12 ; Gopalakrishnan, Mahesh 12 ; Natarajan, Sundaram 13 ; Krishnan, Radhika 13 ; Mani, Sheena Liz 14 ; Agarwal, Manisha 15 ; Behera, Umesh 16 ; Bhattacharjee, Harsha 17 ; Barman, Manabjyoti 17 ; Sen, Alok 18 ; Saxena, Moneesh 19 ; Sil, Asim K. 20 ; Chakabarty, Subhratanu 20 ; Cherian, Thomas 21 ; Jitesh, Reesha 21 ; Naigaonkar, Rushikesh 22 ; Desai, Abishek 22 ; Kulkarni, Sucheta 23 

 University College London, Institute of Ophthalmology, London, UK (GRID:grid.83440.3b) (ISNI:0000000121901201); King’s College London, Section of Ophthalmology, London, UK (GRID:grid.13097.3c) (ISNI:0000 0001 2322 6764) 
 Vision Research Foundation, Chennai, India (GRID:grid.414795.a) (ISNI:0000 0004 1767 4984) 
 Dr. Mohan’s Diabetes Specialities Centre and Madras Diabetes Research Foundation, Chennai, India (GRID:grid.410867.c) (ISNI:0000 0004 1805 2183) 
 Aravind Eye Hospital, Madurai, India (GRID:grid.413854.f) (ISNI:0000 0004 1767 7755) 
 LV Prasad Eye Institute, Anand Bajaj Retina Institute, Srimati Kannuri Santhamma Centre for Vitreoretinal Diseases, Hyderabad, India (GRID:grid.417748.9) (ISNI:0000 0004 1767 1636) 
 University College London, Institute of Ophthalmology, London, UK (GRID:grid.83440.3b) (ISNI:0000000121901201); Moorfields Eye Hospital, NIHR Moorfields Biomedical Research Centre, London, UK (GRID:grid.439257.e) (ISNI:0000 0000 8726 5837) 
 King’s College London, School of Biomedical Engineering & Imaging Sciences, London, UK (GRID:grid.13097.3c) (ISNI:0000 0001 2322 6764) 
 Sankara Nethralaya, Chennai, India (GRID:grid.414795.a) (ISNI:0000 0004 1767 4984) 
 Vision Sciences, UCL, London, UK (GRID:grid.83440.3b) (ISNI:0000000121901201) 
10  SankaraNethralaya, Kolkata, India (GRID:grid.83440.3b) 
11  Aravind Eye Hospital, Coimbatore, India (GRID:grid.413854.f) (ISNI:0000 0004 1767 7755) 
12  Giridhar Eye Institute, Cochin, India (GRID:grid.413854.f) 
13  Aditya Jyot Hospital, Mumbai, India (GRID:grid.413854.f) 
14  Dr Tony Fernandez Eye Hospital, Aluva, India (GRID:grid.413854.f) 
15  Dr Shroff’s Charity Eye Hospital, New Delhi, India (GRID:grid.440313.1) (ISNI:0000 0004 1804 356X) 
16  LV Prasad Eye Institute, Bhubaneshwar, India (GRID:grid.417748.9) (ISNI:0000 0004 1767 1636) 
17  Sri Sankaradeva Nethralaya, Guwahati, India (GRID:grid.417748.9) 
18  Sadguru Netra Chikitsalaya, Chitrakoot, India (GRID:grid.417748.9) 
19  Aurobindo Nethralaya, Raipur, India (GRID:grid.417748.9) 
20  Netra Niramay Niketan, Haldia, India (GRID:grid.417748.9) 
21  Little Flower Hospital and Research Centre, Angamaly, India (GRID:grid.460899.a) (ISNI:0000 0004 1781 2101) 
22  Netra Niramay Niketan, Haldia, India (GRID:grid.460899.a) 
23  HV Desai Hospital, Pune, India (GRID:grid.460899.a) 
Pages
1392
Publication year
2023
Publication date
2023
Publisher
Nature Publishing Group
e-ISSN
20452322
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2768985145
Copyright
© The Author(s) 2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.