Segmentation and quantification of blood vessels for OCT-based micro-angiograms using hybrid shape/intensity...

Area of Science:

• Biomedical Imaging
• Ophthalmology
• Microcirculation Research

Background:

• Optical coherence tomography (OCT) microangiography visualizes 3D blood vessel networks in vivo.
• Quantitative analysis of vasculature (e.g., vessel diameter, morphology) requires efficient segmentation algorithms.

Purpose of the Study:

• To develop a hybrid Hessian/intensity-based method for segmenting and quantifying blood vessels in functional OCT microangiograms.
• To improve the accuracy of vessel segmentation by combining Hessian filters with intensity-based methods.

Main Methods:

• Utilized multi-scale Hessian filters for segmenting tubular structures (blood vessels).
• Compounded Hessian-based segmentation with an intensity-based method to overcome limitations in scale parameter selection.
• Tested the algorithm on a wound healing model and validated performance using a public manual segmentation dataset.

Main Results:

• The hybrid method effectively segments and quantifies blood vessel shape and diameter from OCT microangiograms.
• Demonstrated improved robustness compared to traditional Hessian-based methods by mitigating sensitivity to scale parameters.

Conclusions:

• The developed hybrid segmentation algorithm offers a robust approach for quantitative analysis of micro-angiograms.
• This method holds significant potential for microcirculation research in ophthalmology and diagnosing retinal diseases.