Retinal thickness measurements from optical coherence tomography using a Markov boundary model

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Optical coherence tomography (OCT) provides high-resolution cross-sectional images of the retina.
• Accurate retinal thickness measurements are crucial for diagnosing and managing ocular diseases.
• Current clinical practice relies on subjective visual assessment of retinal thickness.

Purpose of the Study:

• To develop and evaluate an automated system for detecting retinal boundaries in OCT B-scans.
• To enable objective, quantitative measurement of retinal thickness.
• To assess the accuracy and clinical utility of the automated system.

Main Methods:

• Utilized a one-dimensional edge-detection kernel to identify edge primitives.
• Employed a Markov model to organize edge primitives into coherent retinal boundary structures.
• Validated the system using 1450 OCT images, comparing automated results to manually corrected boundaries.

Main Results:

• The automated system demonstrated high qualitative agreement with true retinal structures, failing in only one image.
• Quantitative analysis showed retinal thickness measurements within 10 micrometers of manual corrections for 74% of images.
• Measurements were within 25 micrometers (10% of normal thickness) for 98.4% of images, below clinical significance.

Conclusions:

• The developed automated system for retinal boundary detection in OCT is highly accurate and robust.
• Quantitative retinal thickness measurements from the system are precise and clinically relevant.
• This technology has the potential to enhance patient care by improving upon current subjective assessments.