Effectiveness and Cost-Effectiveness Evaluations of AI-Assisted Diagnostic Software (VeriSee) for Ophthalmic Disease Screening

Not Applicable

Recruiting

• Conditions: Age-Related Macular Degeneration (AMD); Diabetic Retinopathy (DR)

• Registration Number: NCT06843499
• Lead Sponsor: National Taiwan University Hospital

Brief Summary

This study aims to evaluate the effectiveness of an artificial intelligence (AI)-assisted screening system in ophthalmic diagnosis. Using AI-based fundus photography, the system will assist physicians in diagnosing three common eye diseases: age-related macular degeneration and diabetic retinopathy (DR). The AI system will analyze fundus images from participants and rapidly generate detection results for ophthalmologists' reference in making final diagnoses and clinical decisions. The study will assess the clinical benefits of the AI-assisted diagnostic system, providing scientific evidence to enhance the efficiency of ophthalmic disease diagnosis and treatment.

Detailed Description

Artificial Intelligence (AI) has shown significant potential in medical imaging analysis and disease diagnosis, particularly in ophthalmology. Substantial advancements have been made in utilizing AI for diagnosing common ophthalmic diseases, enhancing early detection and improving patient outcomes. Early diagnosis of age-related macular degeneration (AMD) and diabetic retinopathy (DR) is crucial for effective treatment and disease management.

However, current clinical diagnoses rely heavily on ophthalmologists, leading to challenges such as low patient attendance rates and unequal distribution of diagnostic resources. To address these issues, this study will provide robust evidence to further validate the diagnostic performance of AI-assisted screening and clinical effectiveness of the VeriSee AI-assisted diagnostic system in the detection of diabetic DR and AMD.

VeriSee AMD and VeriSee DR are AI-powered medical software tools designed to screen for AMD and DR, respectively. These systems employ advanced AI algorithms to analyze color fundus photography images, assess disease conditions, and evaluate image quality. By integrating this software into clinical workflows, physicians receive instant diagnostic support, improving efficiency and accessibility in ophthalmic disease screening.

Study Timeline

• Study First Submitted: 2025-02-20
• Study First Submitted QC: 2025-02-20
• Study First Posted: 2025-02-25
• Status Verified: 2025-06
• Study Start: 2025-06-02
• Last Update Submitted: 2025-06-17
• Last Update Posted: 2025-06-22
• Primary Completion: 2027-12-31
• Study Completion: 2027-12-31

Recruitment & Eligibility

• Status: RECRUITING
• Sex: All
• Target Recruitment: 1000

Inclusion Criteria

• VeriSee AMD is used in non-retinal subspecialty ophthalmology clinics for adults aged 50 and above.
• VeriSee DR is used in non-retinal subspecialty clinics for diabetic patients aged 20 and above.

Exclusion Criteria

• The patient does not agree to participate in the trial or is unable to provide informed consent.

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Primary Outcome Measures

Name	Time	Method
Sensitivity	From screening to physician-confirmed diagnosis of AMD or DR, an average of 1 month	The sensitivity of the index test (VeriSee) was calculated as the proportion of participants with reference standard-confirmed disease who were correctly identified as positive by the AI-assisted diagnostic software.
Specificity	From screening to physician-confirmed diagnosis of AMD or DR, an average of 1 month	The specificity of the index test was calculated as the proportion of participants without the target condition, as determined by the reference standard, who were correctly classified as negative by the AI-assisted diagnostic tool.
Concordance	From screening to physician-confirmed diagnosis of AMD or DR, an average of 1 month	Concordance between the AI-assisted diagnosis and the ophthalmologists' interpretation was assessed using the overall agreement rate (i.e., the percentage of cases with identical classification results).

Secondary Outcome Measures

Name	Time	Method
Total Cost Analysis (Including Direct and Indirect Costs)	From enrollment to 12 months after screening	This measure includes direct medical costs (e.g., screening, follow-up, medication, and treatment), healthcare-related indirect medical costs (e.g., IT system maintenance, healthcare personnel), and non-medical indirect costs (e.g., transportation and productivity loss due to blindness). Costs will be analyzed from both the National Health Insurance perspective and the broader societal perspective.

Trial Locations

Locations (1)

National Taiwan University Hospital; 🇨🇳 Taipei, Taiwan