Verification and application of convolutional neural networks for automated screening of diabetic retinopathy in Siriraj hospital

Not Applicable

Recruiting

• Conditions: mild NPDR; Diabetes patient withno DR mild NPDRmoderate NPDRsevere NPDRPDRother retinopathy; PDR; other retinopathy; Diabetes patient with; no DR; moderate NPDR; severe NPDR

• Registration Number: TCTR20200513003
• Lead Sponsor: Routine To Research Unit (R2R)

Brief Summary

Not available

Detailed Description

Not available

Study Timeline

• Study Start: 2020-05-13
• Study Completion: 2022-01-19
• Last Update Posted: 19 August 2024

Recruitment & Eligibility

• Status: Recruiting
• Sex: All
• Target Recruitment: 1520

Inclusion Criteria

Diabetes patient who visit outpatient department in Siriraj hospital

Exclusion Criteria

Poor quality fundus photo from diabetes patient
Inconclusive clinical diagnosis

Study & Design

• Study Type: Observational
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
accuracy of automated screening tool for detection of diabetic retinopathy at the time of intervention sensitivity, specificity, PPV, NPV , ROC curve

Secondary Outcome Measures

Name	Time	Method
Program validation at the end of intervention Accuracy