A Digital Staining Algorithm for Optical Coherence Tomography Images of the Optic Nerve Head

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Optical coherence tomography (OCT) is vital for visualizing optic nerve head (ONH) structures.
• Differentiating connective and neural tissues in OCT images is challenging but critical for diagnosing pathologies like glaucoma.

Purpose of the Study:

• To develop and validate a digital staining algorithm for spectral-domain OCT images of the ONH.
• To highlight and differentiate between connective and neural tissues within the ONH.

Main Methods:

• Acquired OCT volumes from 10 healthy subjects.
• Applied adaptive compensation for enhanced deep tissue visibility.
• Utilized a novel digital staining technique based on pixel-intensity histograms to generate stained volumes (P1-P4).
• Verified digital staining with a digital phantom and compared it to k-means clustering.

Main Results:

• The algorithm successfully isolated three regions in a digital phantom.
• Digitally stained images (P1-P4) effectively highlighted connective tissues, nerve fiber layer, prelamina, and retinal layers.
• Demonstrated a significant contrast increase (3.6 ± 0.6 times) in connective tissues.
• Showed superior separation of ONH tissue layers compared to k-means clustering.

Conclusions:

• A novel algorithm for digital staining of connective and neural tissues in OCT images of the ONH has been developed.
• This technique offers potential for improved clinical management of glaucoma by enhancing visualization of affected ONH tissues.