[Skip to Navigation]
Original Investigation
May 11, 2023

Efficacy of Smartphone-Based Telescreening for Retinopathy of Prematurity With and Without Artificial Intelligence in India

Author Affiliations
  • 1Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland
  • 2Department of Ophthalmology, University of Michigan, Ann Arbor
  • 3Department of Pediatric Retina and Ocular Oncology, Aravind Eye Hospital, Coimbatore, India
  • 4Department of Surgical Retina, Singapore National Eye Center, Singapore
  • 5Department of Ophthalmology, University of Colorado, Aurora
  • 6National Eye Institute, National Institutes of Health, Bethesda, Maryland
  • 7National Library of Medicine, National Institutes of Health, Bethesda, Maryland
  • 8Mass General Brigham and Brigham and Women’s Hospital Center for Clinical Data Science, Boston, Massachusetts
  • 9Department of Ophthalmology, Ophthalmology and Visual Sciences, University of Illinois at Chicago, Chicago
JAMA Ophthalmol. Published online May 11, 2023. doi:10.1001/jamaophthalmol.2023.1466
Key Points

Question  Can low field-of-view, inexpensive, smartphone-based fundus imaging (SBFI) systems be effective in telemedicine screening for retinopathy of prematurity (ROP)?

Findings  This cross-sectional study compared standard widefield fundus imaging and 2 different SBFI systems taken by an operational technician-led telemedicine program in India found that neither SBFI approach was superior to the other and that both human graders and an autonomous artificial intelligence (AI)–based system could appropriately refer treatment-requiring ROP using only SBFI photos.

Meaning  The findings in this study suggest that cost-effective SBFI and AI systems may have the potential to effectively screen for treatment-requiring ROP, thereby increasing international screening accessibility.


Importance  Retinopathy of prematurity (ROP) telemedicine screening programs have been found to be effective, but they rely on widefield digital fundus imaging (WDFI) cameras, which are expensive, making them less accessible in low- to middle-income countries. Cheaper, smartphone-based fundus imaging (SBFI) systems have been described, but these have a narrower field of view (FOV) and have not been tested in a real-world, operational telemedicine setting.

Objective  To assess the efficacy of SBFI systems compared with WDFI when used by technicians for ROP screening with both artificial intelligence (AI) and human graders.

Design, Setting, and Participants  This prospective cross-sectional comparison study took place as a single-center ROP teleophthalmology program in India from January 2021 to April 2022. Premature infants who met normal ROP screening criteria and enrolled in the teleophthalmology screening program were included. Those who had already been treated for ROP were excluded.

Exposures  All participants had WDFI images and from 1 of 2 SBFI devices, the Make-In-India (MII) Retcam or Keeler Monocular Indirect Ophthalmoscope (MIO) devices. Two masked readers evaluated zone, stage, plus, and vascular severity scores (VSS, from 1-9) in all images. Smartphone images were then stratified by patient into training (70%), validation (10%), and test (20%) data sets and used to train a ResNet18 deep learning architecture for binary classification of normal vs preplus or plus disease, which was then used for patient-level predictions of referral warranted (RW)– and treatment requiring (TR)–ROP.

Main Outcome and Measures  Sensitivity and specificity of detection of RW-ROP, and TR-ROP by both human graders and an AI system and area under the receiver operating characteristic curve (AUC) of grader-assigned VSS. Sensitivity and specificity were compared between the 2 SBFI systems using Pearson χ2testing.

Results  A total of 156 infants (312 eyes; mean [SD] gestational age, 33.0 [3.0] weeks; 75 [48%] female) were included with paired examinations. Sensitivity and specificity were not found to be statistically different between the 2 SBFI systems. Human graders were effective with SBFI at detecting TR-ROP with a sensitivity of 100% and specificity of 83.49%. The AUCs with grader-assigned VSS only were 0.95 (95% CI, 0.91-0.99) and 0.96 (95% CI, 0.93-0.99) for RW-ROP and TR-ROP, respectively. For the AI system, the sensitivity of detecting TR-ROP sensitivity was 100% with specificity of 58.6%, and RW-ROP sensitivity was 80.0% with specificity of 59.3%.

Conclusions and Relevance  In this cross-sectional study, 2 different SBFI systems used by technicians in an ROP screening program were highly sensitive for TR-ROP. SBFI systems with AI may be a cost-effective method to improve the global capacity for ROP screening.

Add or change institution