Depth-resolved visualization and automated quantification of hyperreflective foci on OCT scans using optical...

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Hyperreflective foci (HRF) in OCT imaging are linked to macular hyperpigmentation and increased risk of age-related macular degeneration (AMD) progression.
• Accurate visualization, localization, and quantification of HRF are crucial for monitoring AMD.
• Current methods for HRF assessment can be labor-intensive and subjective.

Purpose of the Study:

• To develop and validate an automated depth-resolved algorithm for quantifying hyperreflective foci (HRF) using optical attenuation coefficients (OACs) in OCT imaging.
• To differentiate HRF within the retina from those along the retinal pigment epithelium (RPE).
• To assess the accuracy of the automated algorithm by comparing its pigment burden measurements to manual segmentations.

Main Methods:

• Developed an automated algorithm transforming OCT scans to a linear representation, contrasted by OACs.
• Visualized and localized HRF by differentiating intra-retinal HRF from RPE-associated HRF.
• Quantified total pigment burden using en face sum projection of OAC slabs (ILM to BM) and compared with manual measurements.

Main Results:

• The automated algorithm successfully distinguished between intra-retinal HRF and RPE-associated HRF.
• A significant correlation (P < 0.001) was found between the automated and manual measurements of total pigment burden area in 24 test eyes.
• The algorithm demonstrated reliable quantification of pigment burden in swept-source OCT scans from 42 AMD patients.

Conclusions:

• The proposed automated algorithm based on OACs provides an effective method for visualizing, localizing, and quantifying HRF.
• This tool holds significant potential for advancing research in eye diseases characterized by HRF, such as age-related macular degeneration.
• The algorithm's accuracy and automation offer a valuable advancement for clinical studies and disease monitoring.