A low-cost smartphone imaging system could help community health workers screen people for cataracts and other eye conditions in areas where access to ophthalmologists is limited. In a real-world study in rural India, referral decisions made remotely from smartphone images agreed with in-person assessments in 96% of cases.
Researchers from Aravind Eye Care System and Johns Hopkins University tested the approach among 1,093 patients attending 19 rural eye camps in Tamil Nadu. The findings were presented at the 44th Congress of the European Society of Cataract and Refractive Surgeons (ESCRS). The researchers stress that the technology is intended for screening and triage, not as a replacement for a comprehensive ophthalmological examination.
Bringing screening closer to patients
Cataracts remain a major cause of blindness worldwide, with access to specialist eye care particularly limited in low-resource settings. Rural eye camps are commonly used in India to bring ophthalmologists closer to communities, but organising them requires specialists to travel from urban hospitals. Previous research from Aravind Eye Hospital found that eye camps reached only 7% of rural residents in the region studied. Attendance fell by 80% among people living more than three kilometres away. Among those who did not attend, one-third required cataract surgery.
The researchers therefore developed a portable attachment that clips onto an Android smartphone. A lens magnifies the front of the eye, while two LEDs provide illumination. A silicone scope blocks ambient light and keeps the camera at a consistent distance. Including the smartphone, the system costs less than £150. It works with a telemedicine app designed for low-bandwidth environments and is available in English and Tamil. The aim is to allow community health workers with limited ophthalmic training to capture images locally while ophthalmologists assess them remotely.
High agreement on referrals
After three hours of training, community health workers screened 1,093 patients in rural villages around five towns near Pondicherry. They were able to screen each eye in less than 2.5 minutes. Remote ophthalmologists assessed the smartphone images, and their diagnoses and referral decisions were compared with those of ophthalmologists examining the same patients at the eye camps.
The most important result concerned hospital referrals. Remote and in-person ophthalmologists agreed in 96% of cases. Agreement was also high for several lens-related diagnoses. The clinicians agreed on the presence of any cataract in 89% of cases, mature cataracts in 96%, immature cataracts in 85%, clear lenses in 89% and pseudophakia, indicating an existing lens implant, in 97%. Agreement improved as image quality increased.
According to lead researcher Prabhu Krishna Ravilla, the approach could enable ophthalmologists to concentrate their time on diagnosis and clinical decision-making rather than travelling to every screening location. In future, it could potentially support screening closer to people's homes.
Important limitations remain
The results do not yet demonstrate that smartphone screening can replace conventional eye examinations. The study did not compare the remote assessments with a masked ophthalmologist using a slit lamp, which forms part of the gold-standard comprehensive examination.
Performance was also weaker for some conditions outside the lens, including pterygium, refractive error and corneal opacity. Another complication was that eye camp doctors sometimes examined patients after pupil dilation, while community health workers captured smartphone images before dilation. This may have affected remote assessment of cataract severity. The study was conducted in rural South India, so the findings cannot automatically be generalized to other healthcare systems or populations.
ESCRS President Burkhard Dick, who was not involved in the research, described the high agreement in referral decisions as encouraging but emphasized that the approach should be considered a screening and triage tool rather than a substitute for comprehensive ophthalmological assessment.
Next steps for smartphone eye screening
The researchers plan to add slit-beam, blue-light and dilated-eye imaging to improve diagnostic performance. They also want to investigate AI-assisted cataract grading and test the platform in other geographical and cultural settings.
If further studies confirm its effectiveness, the combination of inexpensive smartphone imaging, community health workers and remote specialist assessment could provide another route to eye care for populations that currently face substantial barriers to accessing an ophthalmologist.
Smartphone diagnosis
Last year, researchers from Purdue University, Rwanda Biomedical Center and the University of Rwanda developed a smartphone-based method that could help screen children for anemia without requiring a blood test. The approach analyzes grayscale photographs of the conjunctiva, covering the inside of the eyelid and the white part of the eye. For the study, researchers captured more than 12,000 smartphone images from 565 children aged 5 to 15. Machine learning and radiomics were then used to identify patterns and structural characteristics associated with anemia.
Unlike earlier approaches based on color analysis, the method examines subtle structural changes in blood vessels using black-and-white images. This reduces sensitivity to differences in lighting and smartphone cameras, potentially making the technology more practical in low-resource settings. The researchers emphasized that smartphone screening would not replace conventional diagnostic testing. Instead, it could help identify children who require further evaluation and treatment, particularly where access to healthcare and laboratory diagnostics is limited.
References
ESCRS (press release)
Add ICT&health on Google
Show more content from ICT&health in Google Search.
