Age Related Macular Degeneration, Diabetic Retinopathy, Glaucoma, Glaucoma Suspect, Retinal Vein Occlusion
Conditions
Brief summary
The purpose of this multi-center study is to evaluate the extent to which AI-assisted fundus image interpretation improves the diagnostic performance of ophthalmologists. Rather than assessing the standalone algorithm performance, this study aims to determine the clinical value of using AI as a decision-support tool within actual clinical workflows. At each participating institution, five ophthalmologists within three years of board certification and five ophthalmology residents will participate as readers. All readers will interpret fundus images both with and without the AI-based assistance software. The study will quantitatively compare diagnostic accuracy and reading time across the two conditions for four posterior segment diseases: diabetic retinopathy, age-related macular degeneration, retinal vein occlusion, and glaucoma.
Interventions
DEVICEVUNO Med-Fundus AI
The intervention consists of an AI-based fundus image interpretation software that provides automated outputs for 12 retinal and optic nerve findings (e.g., hemorrhage, exudates, drusen, optic disc change). The system does not generate a direct disease diagnosis. Instead, the AI displays the presence or absence of 12 predefined findings along with their lesion locations. Readers may use this finding-level information as decision-support when determining the presence of the four target diseases (diabetic retinopathy, age-related macular degeneration, retinal vein occlusion, and glaucoma).
Sponsors
Inje University
Dong-A University Hospital
Kosin University Gospel Hospital
Pusan National University Hospital
Pusan National University Yangsan Hospital
Study design
Allocation
NON_RANDOMIZED
Intervention model
SEQUENTIAL
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
Ten readers will be recruited from five participating hospital sites, consisting of: * Five ophthalmologists within three years of board certification * Five ophthalmology residents Ophthalmologists and residents of any age, sex, race, or ethnicity may participate as study readers. All readers must meet the following inclusion criteria: * Licensed physicians qualified to interpret fundus images. * Ophthalmologists within three years of board certification, or ophthalmology residents with no restriction on clinical experience. * Able and willing to complete both the unassisted and AI-assisted reading sessions. * Able to provide informed consent for participation in the reader study. * Affiliated with one of the participating clinical sites.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Performance of readers with and without AI assistance: Sensitivity | Through study completion, approximately 2 months | Sensitivity of reader diagnoses for each of the four target diseases (DR, AMD, RVO, glaucoma) and for any fundus abnormality will be assessed with and without AI assistance, using the image-level reference standard as the comparator, through two reading sessions in which all 10 readers review all cases-randomised for each reader-with a washout period implemented to mitigate recall bias. |
| Performance of readers with and without AI assistance: Specificity | Through study completion, approximately 2 months | Specificity of reader diagnoses for each of the four target diseases (DR, AMD, RVO, glaucoma) and for any fundus abnormality will be assessed with and without AI assistance, using the image-level reference standard as the comparator, through two reading sessions in which all 10 readers review all cases-randomised for each reader-with a washout period implemented to mitigate recall bias. |
| Reading time per image | Through study completion, approximately 2 months | Reading time per image will be measured during both unassisted and AI-assisted interpretation sessions. For each case, the total time from the moment the image is displayed to the moment the reader submits the final disease classification will be recorded automatically by the reading platform. Mean reading time per image will be calculated for each reader and compared between the two conditions to evaluate whether AI assistance reduces interpretation time. |
Countries
South Korea
Outcome results
None listed