Related Experiment Video
Updated: Aug 13, 2026
10:11
Fundus Photography as a Convenient Tool to Study Microvascular Responses to Cardiovascular Disease Risk Factors in Epidemiological Studies
Published on: October 22, 2014
19.3K
Selfie fundus imaging for diabetic retinopathy screening
Somya Kumari1, Pradeep Venkatesh2, Nikhil Tandon1
1Institution: Dr Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India.
Eye (London, England)
|October 13, 2021
Summary
Selfie Fundus Imaging (SFI) allows patients to capture their own retinal images, improving diabetic retinopathy screening accessibility and affordability. This novel approach shows high accuracy, comparable to traditional methods, for detecting diabetic retinopathy and macular edema.
Area of Science:
- Ophthalmology
- Medical Imaging
- Diabetology
Background:
- Diabetic retinopathy screening is crucial for preventing blindness, but current methods face accessibility and affordability challenges.
- Traditional screening relies on trained technicians interpreting fundus images, limiting widespread adoption.
- A novel approach, Selfie Fundus Imaging (SFI), is explored to overcome these barriers.
Purpose of the Study:
- To evaluate the feasibility and accuracy of Selfie Fundus Imaging (SFI) for diabetic retinopathy screening.
- To compare SFI-generated retinal images with those from conventional methods.
- To assess the potential of SFI in enhancing the accessibility and affordability of diabetic retinopathy screening programs.
Main Methods:
- A study involving 60 diabetic patients during COVID-19 lockdown.
- Retinal images were captured using three methods: patient-taken SFI, technician-taken smartphone photos, and a conventional digital fundus camera (gold standard).
- Sensitivity and kappa statistics were used to grade retinopathy and macular edema.
Main Results:
- Selfie Fundus Imaging (SFI) images were gradable in 98% of cases, with 82% of patients capturing images within one minute.
- SFI demonstrated 88.39% sensitivity for detecting diabetic retinopathy, with 85.86% agreement (kappa=0.77) with standard grading.
- For diabetic macular edema, SFI showed 93.67% agreement (kappa=0.91) with standard images.
Conclusions:
- Patients can capture comparable retinal images using SFI with minimal difficulty.
- SFI has the potential to significantly improve diabetic retinopathy screening by addressing availability, accessibility, and affordability barriers.
- SFI can ensure the continuity of screening programs, even during public health crises like pandemics.
