Comparative Evaluation of Fundus Image Interpretation Accuracy in Glaucoma Screening Among Different Physician Groups

Area of Science:

• Ophthalmology
• Medical Imaging
• Diagnostic Accuracy

Background:

• Glaucoma screening relies on interpreting fundus images.
• Variability in diagnostic accuracy among physicians can affect patient outcomes.
• Assessing the impact of physician specialization on glaucoma screening is crucial.

Purpose of the Study:

• To evaluate the diagnostic accuracy of glaucoma screening using fundus images among different physician groups.
• To compare the performance of non-ophthalmologists, non-specialist ophthalmologists, and glaucoma specialists.
• To identify variations in sensitivity, specificity, and predictive values.

Main Methods:

• Retrospective analysis of 69 patient eyes (25 with glaucoma).
• Fundus images captured via TRC-NW8 and interpreted by three physician groups.
• Statistical comparison of diagnostic metrics including sensitivity, specificity, PPV, NPV, and Kappa coefficient using Kruskal-Wallis and Dunn's tests.

Main Results:

• Specialists achieved higher sensitivity (0.68) and negative predictive value (NPV) compared to non-ophthalmologists.
• Non-specialist ophthalmologists also showed improved sensitivity over non-ophthalmologists.
• No significant differences in specificity or positive predictive value (PPV) were observed across groups.

Conclusions:

• Physician's background and specialization level significantly influence glaucoma screening accuracy.
• Specialized training enhances the detection of glaucoma from fundus images.
• Standardizing interpretation protocols may improve consistency in screening outcomes.