Diabetic Retinopathy
Conditions
Keywords
Diabetic Retinopathy, Artificial Intelligence, Deep Learning, Retinal images, Diabetic Macular Edema, Diabetic Eye Problems
Brief summary
This study is a retrospective study to assess the efficacy and accuracy of the EyeCheckup software in screening for diabetic retinopathy in healthcare settings.
Detailed description
This is a clinical diagnostic performance study for EyeCheckup with data obtained consecutively at multiple geographically different sites within the US population. These sites include all healthcare settings with a trained ophthalmic camera operator using non-mydriatic cameras that are compatible with EyeCheckup. Moreover, primary endpoints are devised that will be used to validate EyeCheckup to diagnose DR in the primary care and eye care settings. During this color fundus images representing either 4 wide field or 7 standard fields of view will be graded by a qualified professional for diabetic retinopathy according to ETDRS on DRSS scale. The human grading output will be compared with diagnosis generated for diabetic retinopathy by AI software. A comparison of both these grades will be used to determine sensitivity and specificity for the EyeCheckup software for screening for diabetic retinopathy.
Interventions
Subjects will have underwent Fundus Photography using FDA approved non-mydriatic fundus cameras compatible with EyeCheckup which are either TOPCON NW400, CANNON CR2, CANNON CR2 Plus or OPTOMED AURORA IQ
Sponsors
Ural Telekomunikasyon Sanayi Ticaret Anonim Sirketi
Study design
Observational model
CASE_ONLY
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
22 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* A diagnosis of either diabetes type 1 or diabetes type 2 * 22 years of age or older * Patients Fundus images must include at least 1 macula centered and 1 optic disc nerve centered image * All the subjects must be sequentially photographed meaning no omissions of data between a certain time frame by the sites providing images to ensure consecutive selection * No history of any other retinal vascular disease, glaucoma, or another disease that may affect the appearance of the retina or optic disc (refractive error and ocular surface disease are allowed) Subject with image taken by color fundus photography that meets the following requirements: * The resolution of image is 1024×1024 pixels or higher * The ability to provide a 7 standard or 4 wide field of view * Must obtained from either Topcon NW400 or Cannon Model CR-2 AF or CR-2 Plus AF or Optomed Aurora Cameras along with the following data: their age, ethnicity and gender, patient ocular history as evaluated of verified by the clinical sites where data is obtained
Exclusion criteria
* No diagnosis of either diabetes type 1 or diabetes type 2 * The macula, optic nerve, or other part in the image of color fundus photography is unclear to determine the disease condition as judged by the certified ophthalmologist who is grading the images.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Sensitivity and Specificity of EyeCheckup to detect more than mild diabetic retinopathy | 1 visit (1 day) |
| Sensitivity and Specificity of EyeCheckup to detect vision-threatening diabetic retinopathy | 1 visit (1 day) |
Secondary
| Measure | Time frame |
|---|---|
| Positive Predictive Value | 1 visit (1 day) |
| Negative Predictive Value | 1 visit (1 day) |
Contacts
Primary ContactRim Khazhin
Outcome results
None listed