Can an inexperienced observer accurately plot disc contours using Heidelberg retinal Tomograph?

Area of Science:

• Ophthalmology
• Medical Imaging
• Glaucoma Research

Background:

• Accurate optic disc contour plotting is crucial for glaucoma diagnosis.
• Heidelberg Retinal Tomography (HRT) is a key imaging tool.
• Observer variability can impact diagnostic accuracy.

Purpose of the Study:

• To assess the reliability of inexperienced observers in plotting optic disc contours using HRT images.
• To evaluate the impact of standardized training on observer performance.
• To determine interobserver agreement and variability before and after training.

Main Methods:

• Observational study involving 100 eyes from the Singapore Indian Eye Study.
• Optic disc contours plotted by two new observers on HRT-3 images, before and after a 2-hour training session.
• Comparison with an experienced glaucoma specialist's plots (gold standard).
• Interclass correlation and Bland-Altman plots used for agreement and variability analysis.

Main Results:

• Moderate-to-high agreement was observed between pre-training observers and the expert (ICC 0.76-0.99).
• Training significantly improved interclass correlation for all HRT-3 parameters for both observers.
• Statistically significant improvements were noted in mean retinal nerve fibre layer thickness and other HRT-3 parameters post-training.

Conclusions:

• Inexperienced observers can be effectively trained to plot optic disc contours on HRT images.
• Standardized training leads to improved and acceptable interobserver variability and agreement.
• This facilitates reliable use of HRT imaging in clinical practice and research.