The Clinical Benefit of an Artificial Intelligence Software Implementation on Diabetic Retinopathy Screening
Overview
- Phase
- Not Applicable
- Intervention
- Not specified
- Conditions
- Diabetic Retinopathy
- Sponsor
- Taichung Veterans General Hospital
- Enrollment
- 1000
- Locations
- 1
- Primary Endpoint
- diagnostic accuracy
- Last Updated
- 5 years ago
Overview
Brief Summary
The investigators aim to improve the diagnostic accuracy and the clinical referral rate for diabetic retinopathy by using a deep learning-based software.
Detailed Description
Diabetic retinopathy (DR) is the leading cause of blindness among working-age patients with type 2 diabetes. According to previous studies, early screening and timely treatment can reduce the risk of worsening DR and blindness. International guidelines recommend that screening for DR be performed at least once every year for patients with type 2 diabetes. The investigators will implement a validated deep learning-based software, VeriSee®, in clinics, and evaluate the benefits on diagnostic accuracy and the clinical referral rate for diabetic retinopathy after implementation of this software.
Investigators
Eligibility Criteria
Inclusion Criteria
- •Patients with diabetes
- •Cooperation to fundal scopic examination
Exclusion Criteria
- •Diabetic duration \< 5 years in patients with type 1 diabetes
- •Pregnancy
Outcomes
Primary Outcomes
diagnostic accuracy
Time Frame: 12 months
diagnostic accuracy compared to the baseline
Secondary Outcomes
- Screening rate of diabetic retinopathy(12 months)
- Changes in HbA1c(3 months)
- Referral rate of diabetic retinopathy(12 months)