Level-Set Method for Image Analysis of Schlemm's Canal and Trabecular Meshwork

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Accurate segmentation of Schlemm's canal (SC) and trabecular meshwork (TM) is crucial for understanding glaucoma.
• Ultrasound biomicroscopy (UBM) provides detailed images of the anterior segment, but analysis can be challenging.
• Quantitative analysis of TM-SC requires robust segmentation methods.

Purpose of the Study:

• To evaluate and compare the performance of different image segmentation methods for analyzing SC and TM in UBM images.
• To assess the accuracy and repeatability of K-means, fuzzy C-means, and level-set methods.
• To investigate the correlation between intraocular pressure (IOP) and TM-SC geometric measurements.

Main Methods:

• Twenty-six healthy volunteers underwent UBM imaging at varying IOPs.
• UBM images were segmented using ImageJ, K-means, fuzzy C-means, and level-set methods.
• Quantitative analysis included relative error, interclass correlation coefficient (ICC), and Pearson correlation.

Main Results:

• The level-set method demonstrated superior similarity to ImageJ results compared to K-means and fuzzy C-means.
• High ICC values for the level-set method indicated excellent repeatability (0.97, 0.95, 0.9).
• Significant negative correlations were found between IOP and SC area (-0.91), SC perimeter (-0.72), SC length (-0.66), and TM width (-0.61).

Conclusions:

• The level-set method offers improved accuracy, precision, repeatability, and efficiency for UBM image segmentation compared to other methods.
• It provides a reliable and semiautomated approach for analyzing the TM-SC region.
• This method has translational relevance for clinical applications in glaucoma assessment.