Detection and Classification of Diabetic Retinopathy From Posterior Pole Images With A Deep Learning Model
Detection and Classification of Diabetic Retinopathy From Posterior Pole Images With A Deep Learning Model
The duration of diabetes is directly related to eye complications. Diabetic retinopathy affects 80 percent of those who have had diabetes for 20 years or more. At least 90% of new cases can be reduced with proper treatment and monitoring of the eyes. The longer a person has diabetes, the more likely it is to develop diabetic retinopathy. Each year in the United States, diabetic retinopathy accounts for 12% of all new cases of blindness. It is also the leading cause of blindness in people between the ages of 20 and 64. The most important complication of diabetes leading to vision loss is diabetic retinopathy. Depending on this, macular edema, bleeding into the retina and vitreous,neovascular glaucoma can cause blindness.
Diabetic retinopathy (DR) is a leading cause of vision-loss globally. Of an estimated 285 million people with diabetes mellitus worldwide, approximately one third have signs of DR and of these, a further one third of DR is vision-threatening DR, including diabetic macular edema (DME). Diabetic retinopathy is a retinal disease that can often be stopped with early diagnosis, but if neglected, it can lead to severe vision loss, including permanent blindness. Diabetes has high morbidity and there are millions of people who should be screened for diabetic retinopathy (DR). Annual eye screening is recommended for all diabetic patients since vision loss can be prevented if DR is diagnosed in its early stages. Currently, the number of clinical personnel trained for DR screening is less than that needed to screen a growing diabetic population. Therefore, the automatic DR screening system will be able to screen more diabetic patients and diagnose them early.
EyeCheckup is an automated retinal screening device designed automatically analyze color fundus photographs of diabetic patients to identify patients with referable or vision threatening DR. This study is designed to assess the safety and efficacy of EyeCheckup.
The study is a single center study to determine the sensitivity and specificity of EyeCheckup to diabetic retinopathy. EyeCheckup is an automated software device that is designed to analyze ocular fundus digital color photographs taken in frontline primary care settings in order to quickly screen for diabetic retinopathy (DR).
This is a prospective study to assess the safety and efficacy of EyeCheckup in screening for DR.
This study was carried out in a single center at Akdeniz University Faculty of Medicine with primary endpoints to determine the sensitivity and specificity of EyeCheckup to diabetic retinopathy in the primary care setting.
Methods and tools to be used in the study:
Clinical and laboratory tests to be performed:
After exclusions, this study will enroll up to 900 subjects who are diagnosed with diabetes by the endocrinology polyclinic and meet the eligibility criteria. Participants who meet the eligibility criteria will be recruited after obtaining written informed consent from primary health care providers. Subjects will undergo fundus photography per, Food and Drug Administration (FDA) cleared, ophthalmic cameras (product code: HKI). Images will be taken according to a specific EyeCheckup imaging protocol provided to the ophthalmic camera operator and then analyzed by the EyeCheckup device.
The photography protocol consists of two images of the ocular fundus (one optic disc nerve centered, one macula centered), obtained from both eyes of enrolled participants.
After the retinal images taken from ophthalmic cameras (product code: HKI), images are analyzed with EyeCheckup and a scan report is prepared. If it is necessary to enlarge the pupils, eye enlarging eye drops are applied and wait 15-30 minutes. This information is noted. DR is diagnosed by examination by a retina specialist with the captured images. EyeCheckup success rate is calculated by comparing both reports.
Inclusion Criteria:
Exclusion Criteria: