All OCTA data and biochemical indexes of diabetic patients were acquired. A prediction model of diabetic retinopathy was built, and the random forest method was used to identify sensitive indicators.

Diabetic patients at the Department of Ophthalmology of the Seventh Affiliated Hospital of Sun Yat-sen University were selected as research participants. All OCTA data and biochemical indexes were acquired. A prediction model of diabetic retinopathy was built, and the random forest method was used to identify sensitive indicators.

Inclusion Criteria:

• Clinical diagnosis of diabetes

Exclusion Criteria:

• various other types of retinal and choroidal disease;
• history of any intraocular surgery or treatment (including intravitreal injection, retinal photocoagulation etc.
• with severe refractive medium opacity that could affect fundus examination.

diabetic patients

The research has not been completed.