Improving the nationwide screening and referral system for diabetic retinopathy using artificial intelligence integrated with hospital information systems and retinal imaging technologies Diabetic retinopathy, screening, deep learning algorithm, Hospital Information System integration, retinal camera integration, artificial intelligence-assisted diabetic retinopathy screening
Conditions
Interventions
A digital health platform integrated with hospital information systems and fundus cameras to support diabetic retinopathy screening using an artificial intelligence (AI). The system automatically anal
Experimental Device
AI-assisted DR Screening Platform with Digital Health Integration
Sponsors
Health Systems Research Institute (HSRI)
Eligibility
Sex/Gender
All
Age
18 Years to No maximum
Inclusion criteria
Inclusion criteria: 1. Patients with diabetes aged 18 years or older 2. Patients with diabetes who are listed in the civil registration and are eligible for diabetic retinopathy screening according to Ministry of Public Health guidelines 3. Able to obtain at least one retinal (fundus) image 4. Listed in the diabetic patient database at a primary care hospital that does not have a resident ophthalmologist, and who require referral for further management of diabetic retinopathy at a tertiary hospital
Exclusion criteria
Exclusion criteria: 1. Patients listed in the primary hospital database where an ophthalmologist is available on site 2. Patients previously diagnosed with any of the following conditions: macular edema from non-diabetic causes (e.g., age-related macular degeneration), radiation retinopathy, or retinal vein occlusion 3. History of retinal laser treatment or retinal surgery 4. Presence of other ocular diseases (not diabetic retinopathy) that require referral to an ophthalmologist 5. Unable to obtain retinal images from both eyes for any reason
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Successful integration of AI-assisted referral platform with hospital information systems (HIS) and fundus cameras 1 year Number and proportion of participating hospitals with confirmed HIS integration and Number and proportion of fundus cameras with automated image transfer capability into the digital platform. | — |
Secondary
| Measure | Time frame |
|---|---|
| Implementation coverage and usage of the digital platform for diabetic retinopathy screening 1 year Number of patients screened using the integrated platform; percentage relative to the total eligible diabetic population per region,Referral criteria customization within the platform based on provincial guidelines 1 year Number of provinces with successfully implemented referral rule sets in the platform | — |
Countries
Thailand
Contacts
Public ContactJirawut Limwattanayingyong
Rajavithi Hospital
Outcome results
None listed