A single-center, single-arm, retrospective pivotal trial to evaluate the efficacy of artificial intelligence-based fundus photography analysis software.
- Conditions
- Diseases of the eye and adnexa
- Registration Number
- KCT0004902
- Lead Sponsor
- Vuno
- Brief Summary
Not available
- Detailed Description
Not available
Recruitment & Eligibility
- Status
- Completed
- Sex
- All
- Target Recruitment
- 1713
1) Adults at the age 19 or above who received the fundus examination within the period from July 2016 to June 2018
2) When an abnormal finding in the fundus image can be confirmed based on the physical examination result or medical records
*If the subject received the fundus examination more than 2 times, only the first examination result is included.
- Up to 1713 images [269 images without an abnormal finding and 1444 images with an abnormal finding (including those with 2 or more abnormal findings)]
-When the clinical study is conducted, the actual number of fundus images collected will not exceed the maximum number since the fundus image database will be built per abnormal finding type. That is, a fundus image without a certain abnormal finding type can be included as a fundus image without other abnormal findings except the certain finding.
1) When a fundus image is not suitable for reading because there are serious cataracts or corneal opacities in the eye from which the fundus image was taken
2) Those who are otherwise considered by the investigator to be not suitable for participating in the study
Study & Design
- Study Type
- Interventional Study
- Study Design
- Not specified
- Primary Outcome Measures
Name Time Method AUC (area under curve): Area under the ROC (receiver operating characteristic) curve, evaluated for 12 each abnormal finding
- Secondary Outcome Measures
Name Time Method