Automated segmentation and quantification of calcified drusen in 3D swept source OCT imaging

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Calcified drusen assessment is crucial for predicting age-related macular degeneration (AMD) progression.
• Current methods for drusen quantification can be labor-intensive and subjective.

Purpose of the Study:

• To develop and validate an automated algorithm for segmenting and quantifying calcified drusen using 3D SS-OCT.
• To improve the accuracy of drusen quantification by incorporating RPE correction.

Main Methods:

• Developed an algorithm leveraging optical attenuation coefficient (OAC) and choroidal hypotransmission defects (hypoTD) for calcified drusen segmentation.
• Implemented a novel RPE correction method to enhance segmentation accuracy.
• Validated the algorithm on 29 eyes with nonexudative AMD and calcified drusen using SS-OCT.

Main Results:

• The automated method demonstrated good agreement with human expert graders for calcified drusen area.
• Achieved a Dice similarity coefficient of 68.27% ± 11.09% for segmentation.
• Showed a high correlation coefficient (r=0.9422) and low bias for area measurements.

Conclusions:

• The proposed automated algorithm accurately segments and quantifies calcified drusen in SS-OCT images.
• This tool has the potential to aid in the clinical assessment and management of AMD.
• Automated analysis of calcified drusen can lead to more objective and efficient patient monitoring.