Automatic segmentation of closed-contour features in ophthalmic images using graph theory and dynamic programming

Area of Science:

• Medical Imaging
• Computational Biology
• Ophthalmology

Background:

• Accurate segmentation of anatomical and pathological features is crucial in medical imaging.
• Existing methods may have limitations in segmenting complex closed-contour structures.
• Graph theory and dynamic programming (GTDP) have shown promise in quantitative analysis.

Purpose of the Study:

• To present a generalized framework for segmenting closed-contour anatomical and pathological features.
• To extend the GTDP method for segmenting various biological structures, including cells and cysts.
• To validate the accuracy of the proposed method in ophthalmic imaging applications.

Main Methods:

• A novel transform maps closed-contour features from Cartesian to quasi-polar coordinates.
• Graph theory and dynamic programming (GTDP) are employed to segment features as layers.
• The method is applied to diverse ophthalmic image types.

Main Results:

• The generalized GTDP framework successfully segments closed-contour features.
• Demonstrated application across multiple ophthalmic image modalities.
• Quantitative validation confirmed high accuracy for retinal pigmented epithelium cell segmentation.

Conclusions:

• The presented GTDP-based framework offers a robust and accurate method for segmenting closed-contour biological features.
• The technique is versatile and applicable to various ophthalmic imaging scenarios.
• This approach enhances the quantitative analysis of cellular and pathological structures in microscopy.