This study aims to compare the effect of Aurora handheld fundus camera with traditional desktop fundus camera in the fundus photography screening of diabetic patients, and to evaluate the effect of artificial intelligence algorithm in the diagnosis of diabetic retinopathy.
Aurora handheld fundus cameras are used to take fundus photography on diabetic patients in 3 ophthalmic diabetic retinopathy screening centers in China to compare its effect with the hospital's traditional desktop fundus camera, and evaluate the auxiliary diagnostic effect of Phoebus artificial intelligence algorithm in diabetic retinopathy.
Study Type
OBSERVATIONAL
Enrollment
300
Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine
Shanghai, Shanghai Municipality, China
Image Quality of Aurora camera
Score of Image Quality
Time frame: within 3 months
Outcome of gold standard
Gold standard: 8 photographs of one patient(4 by Aurora camera, 4 by traditional camera) by ophthalmologist（double blinded）
Time frame: within 3 months
Image of Aurora camera
Images taken by Aurora camera
Time frame: 1 month
Image of traditional camera (Center 1: Canon)
Image taken by traditional camera（Center 1: Canon, pupil not dilated）
Time frame: 1 month
Image of traditional camera (Center 2: Zeiss)
Image taken by traditional camera（Center 2: Zeiss, pupil dilated）
Time frame: 1 month
Image of traditional camera (Center 3: Topcon)
Image taken by traditional camera（Center 3: Topcon, pupil dilated）
Time frame: 1 month
Outcome of artificial intelligence algorithm
Outcome of artificial intelligence algorithm(Normal or Referral Required)
Time frame: 3 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.